The most psychologically brutal children’s movie ever released was a box office success by convincing a generation of eight-year-olds that reality is optional and that a formless void is coming to erase everything they love. I was one of those eight-year-olds, and I will never fully recover; I find myself wondering if and how that might have made the difference for millions of today’s entrepreneurs.

The NeverEnding Story was a training film for anyone whose job is to keep something alive against an enemy whose entire method is convincing people to give up. That is the whole job of a founder, and it is the whole job of the people who build the economies founders operate in, so let’s talk about why a 1984 West German art film about a luckdragon is the most accurate thing ever made about startup development.

The Movie Cost More Than Any German Film in History, and the Author Wanted His Name Off It

Some background, because the origin story is a lesson in creative risk. Wait… “German”?

The NeverEnding Story was co-written and directed by Wolfgang Petersen, his first English-language film, adapted from the 1979 novel by German author Michael Ende. Petersen had just won acclaim for Das Boot, the claustrophobic submarine film, and he took the money and built a world of luckdragons, rock-eating giants, and racing snails. It was the most expensive film ever produced in Germany at the time, beating Petersen’s own previous record with Das Boot.

A “West German fantasy film” being the most expensive production in a national film industry is the cinematic equivalent of a seed-stage founder raising a Series B before shipping; audacious, slightly insane, and justified in retrospect.

You’ll love this because this is also very familiar to founders. Ende, the author, the person whose imagination the entire thing was built on, hated it. Hated it enough to sue. He was invited to a private screening, watched his life’s work rendered in plush and puppetry, and, on his own official biography site, he called it “a humungous melodrama of kitsch, commerce, plush and plastic” and dismissed it as “a cross between E.T. and The Day After.”

He lost the lawsuit.

The film kept his title and spawned two sequels he had nothing to do with.

This is a founder’s irony. The creator built a world, sold the rights to people who promised to honor it, watched them change it, and then watched the version he disowned become the thing that made his idea immortal to tens of millions of people who would never read a word he wrote. The book has sold more than 35 million copies across 40-plus languages; ask the average person on the street about The NeverEnding Story and they will hum a synth-pop theme, not quote Ende’s prose. Every founder who has watched a board, an acquirer, or a co-founder take the vision somewhere the founder never intended knows exactly what Ende felt. The idea outlived his control of it.

That is not a betrayal of the creative process; that is the creative process, and it is the first thing the movie teaches by accident before the opening credits even finish.

What the Story Actually Is (and Why It Terrified Children on Purpose)

The plot, for anyone who blocked it out as a survival mechanism, is that Bastian, a bullied, grieving kid, hides from bullies in a bookshop and steals a book called The NeverEnding Story. He reads it in a school attic, and the story inside is about Fantasia, a magical world being devoured by the Nothing, a formless void that is erasing the world piece by piece. Fantasia’s ruler, the Childlike Empress, is dying because the world is dying, and she dispatches a young warrior named Atreyu to find a cure and stop the Nothing.

Other children’s movies of the era gave kids a villain with a face of a wicked queen, a wizard, a shark, or a guy in a mask. This one gave children the terror of non-existence as the antagonist. The enemy is not evil; the enemy is absence. Worse, as the writer of this delightfully unhinged piece “8 Reasons Why The Neverending Story Is a Psychological Horror Show” wrote, the film shows you what the Nothing does to the will to live:

“It’s really not just that Artax the Horse dies, so much as how he dies and the way it’s shown. First off, it’s not like he just sprains his ankle and has to be sent to the glue factory, no, Artax dies because he is so very sad that he just decides to give up and perish. Jesus. And it’s not like this is a quick way to go. Instead, the scene just drags on and on, cutting to a wailing pre-teen Atreyu, and a horse that looks very seriously, actually afraid of drowning on that movie set in real-life. Don’t close your eyes! Death is real and there is no amount of crying and screaming that will ever prevent our slowly guttering flames. See you in the sequel!”

Atreyu’s beloved horse, Artax, sinks into the Swamps of Sadness, not because a monster kills him, but because he stops believing the journey is worth continuing; despair is literally physically fatal in this universe. Throughout Fantasia, characters who have lost their friends to the Nothing simply sit down and wait to be erased, choosing surrender over the effort of resistance. A children’s film built its central threat around the idea that giving up is the mechanism of death, and it was right.

Then the movie does the thing that made me, personally, question my own existence in a school library at age eight.

Near the end, the Empress reveals to Atreyu that he is a character in a book being read by Bastian, subject to a reader he cannot see or control.

And then she reveals that Bastian, the “real” boy, the audience surrogate, is also being watched, by the people watching the film.

The story folds reality in on itself and hands a room full of children the existential question that philosophy departments charge tuition for; am I real, or is someone reading me, and is someone watching them?? Sleep well, kids.

(And you thought Inception was trippy)

Why the Nothing Is the PERFECT Metaphor for What Kills Companies

I have spent twenty years watching startups and startup ecosystems die, and they almost never die the way people claim or think. Founders expect to lose to a competitor, a market shift, a funding crunch, a villain with a face. Cities claim they lose their talent to Austin or the Bay Area, a big company relocating all the jobs, or some external raider carrying off their best people.

That is not usually what actually happens. What usually happens is the Nothing.

A startup ecosystem, meaning the local web of founders, investors, mentors, universities, and anchor employers that determines whether ambitious companies can form and scale in a place, does not typically collapse because someone attacks it. It collapses because the people in it stop believing that building something ambitious there is a rational thing to attempt; they sit down in the Swamps of Sadness and wait to be erased. The talent doesn’t get stolen; it quietly concludes there’s no point and takes the job in a bigger city, or takes no risk at all. I wrote a whole piece arguing that nobody is coming to save your economy, and the reason that message lands is that most struggling regions are already sitting in the swamp, waiting for a corporate headquarters to relocate, or for venture capital to arrive, and rescue them the way Fantasia’s residents wait for someone else to stop the Nothing.

Watch what founders and cities do when they’ve decided the void is winning. They stop pitching. They stop shipping.

They keep the lights on and run the retention-and-expansion playbook, protecting what already exists, because protecting feels safer than creating. Existing companies are actually net job destroyers in aggregate while new firm formation is the only reliable engine of net job creation; the deeper problem underneath that fact is that a region defaulting to protection instead of creation has already conceded the fight to the Nothing. The measurable symptom is that capital stops showing up. The actual cause is that belief stopped showing up first, and capital, being a result of conviction, followed belief out the door.

Cities diagnose a money problem because founders all say they’re struggling to raise capital and go begging venture capitalists to “come support our startups,” which is the wrong approach entirely. Money is not the Nothing’s opposite; story is. Fantasia is not saved by a bigger army or a better weapon. It is saved when Bastian, the reader, is willing to believe in it out loud, to shout a name into a storm, to participate rather than spectate. The Childlike Empress cannot be cured by resources; she can only be cured by a human being deciding the world is worth continuing and lending it a new name.

If you have ever wondered why some underfunded, overlooked, “why is anyone building there” regions produce founders while richer ones produce PowerPoints, you already know the answer; one of them has someone willing to name the thing and mean it, and the other is waiting for a check to believe first.

Storytelling Is Not the Soft Part of the Job; It Is the Job

Founders love to treat narrative as the fluffy stuff, the pitch-deck garnish you add after the “real” work of product and revenue.

I find that people who disagree with my sub headline here, are the people who aren’t very good at the story or don’t know how to build an audience. Please understand what confirmation bias means; if someone sucks at storytelling, they’re likely to assert that this stuff isn’t what matters. We have three decades of tech-forward startups; that’s a lot of investors and founders who stumbled into success and now advise that marketing and storytelling aren’t important (because they don’t know how!). If you’re prepared to follow the same advice that has consistently resulted in 90% of startups failing, by all means, listen to the founder who was successful because of grit, luck, timing, and the right connections; never mind the fact that research shows that this is the most important job of a founder. The NeverEnding Story is a 94-minute case study. In Fantasia, the story is not a description of the world; the story is the world, and when the storytelling stops, the world literally ceases to exist. That is not a metaphor I’m stretching to make a point; the rule of the film happens to be the rule of company formation too.

Consider what a founder is actually doing in the earliest days, before there is revenue, product-market fit, or proof of anything.

They are asking employees to take a pay cut against a future that does not exist yet, asking investors to fund a spreadsheet full of assumptions, asking customers to trust a company that might not survive the quarter.

At least, you’d better be doing that, because if you’re just building something or launching an MVP planning to then do that, let me show you the door so you stop wasting everyone’s time.

Customer discovery is a story. Idea validation is storytelling. Every one of those asks is a story, and the founder’s ability to make people believe the ending is worth reaching is the entire job before anything exists to prove it. Most pitch decks fail because founders think in PowerPoint instead of narrative logic; the founders who raise are the ones who understand they are Bastian, that their willingness to believe out loud is what keeps the thing from dissolving.

The same is true one level up, for the ecosystem builders.

A region’s story about itself is not marketing; it is the permission structure that determines whether a talented person believes they can build something there. When we write that capital follows culture, not geography, this is why; the culture is the story a place tells about what is possible in it, and the capital is just the audience showing up once the story is good enough to believe. Richmond, Virginia forgot it ran the world’s first commercially successful electric streetcar; Buffalo forgot Tesla stood in its downtown and called it the future. Those regions did not lose their potential; they lost their story, and the Nothing crept in through the gap where the narrative used to be.

Reciting a region’s real history of invention is not nostalgia; it is Bastian shouting the Empress’s new name, refusing to let the void have the last word.

Startups Are Neverending by Design; That Is the Point of the Title

A lifestyle business or a service firm has a natural endpoint built into it; you open the shop, you serve the customers, you make a living, and the story concludes when you retire or sell.

That is a fine story, but it is a finite one, and there is nothing wrong with finite stories.

A startup is a different animal, and the difference is definitional, not a matter of size. A startup is an organization built for exponential, scalable growth aimed at changing some part of how the world works, which means the mission by design has no completion date; you don’t “finish” changing an industry the way you finish a plumbing job. The company either dies or it keeps going, keeps expanding the boundary of what it set out to alter, keeps handing the mission to the next cohort of employees and the next round of customers and eventually to acquirers or successors who carry it past the founder entirely. The venture that actually changes the world becomes, by its own logic, a neverending story; the ending keeps receding because the mission keeps growing.

And the ecosystem that produces those companies is the most neverending story of all, because it runs on a loop that has no natural stopping point. Companies create value, exit, and recycle both the money and the experience back into the next generation of founders and investors; the successful founder becomes the angel investor becomes the mentor becomes the anchor employer whose existence makes the next person’s risk survivable.

There is no final page.

There is only the question of whether this generation keeps the loop alive long enough to hand it to the next one, or whether they sit down in the swamp and let the Nothing have it. Every city, every region, and every country is, functionally, telling a story they will not live to see the end of, on behalf of founders they have not met yet. That is the job. It was always the job.

What Bastian Actually Does that Matters Here

The most quotable thing about the movie is not the theme song, though the theme song lives rent-free in the skull of everyone born before 1990 and got a fresh generation hooked when Stranger Things revived it. The most quotable thing is that Bastian does not defeat the Nothing with a sword. He defeats it by participating; by refusing to stay a spectator, by believing the world is worth saving loudly enough to act on it, by giving the Empress a new name and, in doing so, being handed the power to rebuild Fantasia grain by grain from a single surviving speck.

That is the methodology of building anything against the void, stated as a children’s fantasy so it would slip past the defenses of eight-year-olds and lodge somewhere permanent.

You do not beat the Nothing by out-resourcing it, because the Nothing is not made of resources; it is made of surrender. You beat it by being the person willing to believe out loud, to name the thing, to act as though the ending is worth reaching before you have any proof that it is. Founders do this every time they hire someone into a company that might not exist next year. Ecosystem builders do this every time they tell a forgotten region its own history back to it and dare it to believe. Investors do this every time they write a check against a story instead of a track record, which, contrary to the myth, is what the good ones actually do.

The movie was too dark for children because it told them the truth; the void is real, giving up is fatal, and no one is coming to read your story for you. It was also, for the same reason, the most useful thing anyone ever handed a future entrepreneur, because it also told them the void loses the moment someone decides to participate.

A generation watched a horse die in a swamp, questioned whether they were real, and walked out having absorbed the one lesson that separates the people who build things from the people who wait to be erased.

Go watch it again. Watch it as the person you are now, running the company or the fund or the city, and notice that you are not Atreyu anymore, the warrior inside the story doing the quest. You are Bastian in the attic, the one whose willingness to believe determines whether the world on the page gets to keep existing. What I realized thinking about this is that when the Nothing comes for the thing you’re building, and it always does, you either shout the name into the storm or are you sit down in the swamp and give up.

In 1992, a Cajun political operative named James Carville hung a sign in Bill Clinton’s Little Rock war room with three lines on it, and the middle line became one of the most quoted phrases in American political history. The sign read, in part, “The economy, stupid,” a reminder to campaign staff to stop chasing whatever scandal the news cycle handed them and stay focused on the one thing voters actually felt in their wallets. Bush had a 90 percent approval rating in March 1991 after the Gulf War; by August 1992, 64 percent of Americans disapproved of him, because the economy had turned and nobody in his campaign wanted to talk about it. Clinton won. The phrase outlived the election and became a template you’ve heard a thousand variations of since, “it’s the [whatever], stupid,” deployed any time someone wants to grab you by the collar and point at the thing everyone in the room is politely ignoring.

It’s a rude construction. That’s the point of it. The word “stupid” isn’t there to inform you; it’s there to insult you into paying attention to something you should already know. And I’m using it on purpose, because in the sector of the economy I work in, there’s a thing everyone in the room is politely ignoring, and the polite ignoring is doing measurable damage.

Startups don’t fail because founders are stupid. They fail because the ecosystem around the founder is broken, and almost nobody funding that ecosystem wants to say the word out loud.

The Thing Everyone Blames Is Almost Never the Thing

Ask a room full of economic developers, or investors, or the founders themselves, why startups fail, and you’ll get a predictable liturgy.

No product-market fit.

Ran out of money.

The founder was too stubborn, or not stubborn enough, or hired wrong, or scaled too fast.

Every one of those explanations puts the failure inside the company, which is convenient, because if the failure lives inside the company, then the people who built the programs, wrote the checks, and set the policy around the company get to keep their jobs and their self-image intact.

I’ve spent over two decades doing regional ecosystem work across cities and governments, and in almost all of my experience, that’s not where the failure lives. Founders fail for lack of good mentorship, bad startup programs that reward attendance over outcomes, inexperienced investors who don’t know what they’re funding, and regulatory and policy friction that treats a pre-revenue venture like a corner store with a payroll. The founder is frequently the most competent actor in the entire arrangement, working inside a support structure that’s actively working against them and calling itself help.

Here’s the empirical part, and it should make everyone in the accelerator business uncomfortable. A landmark National Bureau of Economic Research and NYU Stern study analyzed 750,000 U.S. startups across 329 accelerators and found that 60 to 80 percent of accelerators generate negative value for founders compared to going it alone.

Read that again. Please.

The majority of the programs built specifically to help founders succeed leave those founders worse off than if they’d never walked in the door.

That’s not a founder-quality problem; those are the same founders who would have done better with no program at all. That’s an ecosystem-quality problem wearing a founder-quality costume.

When most of the help is worse than no help, you don’t have a founder shortage. You have an infrastructure that’s failing the people it was built to serve and then blaming them for the failure.

Why “It’s the Founder” Is Such a Comfortable Lie

I believe that the reason the founder-blame explanation is so sticky is that it’s cheap. It’s easy.

If a startup fails because the founder was flawed, then the accelerator that took their equity, the angel group that gave them a term sheet they didn’t understand, the university that licensed away their IP, and the city that funded a coworking space instead of a capital strategy all get to keep their narrative.

The founder was the variable; everything else was sound. Next founder, please.

It’s the same move a casino makes when it tells you the house didn’t beat you, you just played badly; technically sometimes true, structurally always self-serving, and it conveniently never examines whether the game itself was ever winnable as designed.

The founder-blame story survives because it’s the only explanation that requires no one with a budget to change anything. And that’s exactly why it deserves the Carville treatment, because the underlying cause is the same one hiding under “it’s the founder” and “it’s the ecosystem” both; the people who fund this stuff would rather locate the failure anywhere except in the system they’re responsible for, and both stories are just different ways of not looking there.

Startups don’t fail for lack of founder ability; they fail because we built the support system for the ribbon-cutting, not for the founder.

Consider four parties involved in failures, and notice that none of them are things a founder can fix by being smarter or trying harder.

• Mentorship that isn’t. Most ecosystems staff their mentorship with whoever showed up, which means a founder trying to raise a seed round gets paired with a retired corporate VP who has never once been on the receiving end of a term sheet. There’s real research on this! A study of 157 startups in Canadian accelerators found that the quality of the mentor-entrepreneur relationship and the intensity of investor access were what actually moved the needle on follow-on funding, not the mere existence of a program. A mentor who has done the specific thing the founder is trying to do is worth fifty who are “happy to help.”

• Programs that reward activity instead of outcomes. An accelerator that measures itself by cohort size, demo-day attendance, and press mentions is optimizing for the things that look good in a report to whoever funds it, which are almost perfectly uncorrelated with whether any founder in the room built a company. I’ve written before about why the ethical thing to do is measure and expect outcomes rather than activity, and why a region that can’t tell you its founder progression, capital deployed, and local recirculation of exits doesn’t actually know whether its ecosystem works.

• Inexperienced investors. A first-time angel who made money in real estate and now wants to “do some tech deals” is not a neutral party in a founder’s life; they’re a governance risk with a checkbook. They push for the wrong metrics, panic at the wrong moments, and structure terms that quietly poison the next round. Capital is not the same thing as smart capital, and a region full of the former while starving for the latter will produce a lot of funded companies that die anyway.

• Policy and regulatory friction. When licensing regimes, related law complexity, and compliance burdens treat a two-person pre-revenue venture like an established employer, the founder spends the scarce, irreplaceable early months doing paperwork instead of finding customers. I’ve mapped how structural policy reform separates the states that help founders from the ones actively undermining them; the headaches are invisible to the people who impose it (or worse, ignored) while these issues are existential to the founders who absorb them.

Capital Is the Symptom. Belief Is the Disease.

When a startup dies and everyone does the post-mortem, the finding is almost always “couldn’t raise the next round.” Ran out of money. And so the region concludes it has a capital problem, lobbies for a fund, cuts a ribbon on it, and waits for the ecosystem to bloom.

Your city does this, I know it does. I’ve watched dozens of cities this year celebrate that they secured the funding or grants to do something.

Money is a symptom that shows up late.

Capital doesn’t fail to appear because there isn’t enough of it in the world; there is a staggering amount of it, sitting in safer assets, waiting for a reason to move. Capital fails to appear in a region because the region hasn’t built the belief structure that makes local risk feel rational to the people holding the money. The narrative, the expectation, the cultural permission to bet on a founder who might fail; those come first, and capital follows them. Chase the money problem and you’ll fund a vehicle that has nothing to carry. Chase the belief problem, the culture and the expectations and the stories a region tells about what’s possible there, and the capital shows up on its own, because now there’s a reason for it to. Dr. , who built a scorecard to measure how connected the actors in a region actually are, reiterated our concerns in my piece on why government avoids the word “ecosystem,” “the technology is rarely the constraint; the relationships and experience around it” are.

The same logic applies to capital; the check is rarely the constraint, the belief that makes the check rational is.

While celebrating that capital was awarded, many of us are desperately wondering, “are entrepreneurs more successful there?” “Will that funding create jobs?”

Government Keeps Funding the Building and Skipping the Ecosystem

If failure is systemic, you’d expect the people with the biggest budgets to be systematically fixing it. They are systematically doing the opposite.

The short version of why the government won’t focus on ecosystems is that policymakers, faced with a systemic problem and a budget cycle that demands a photograph before the next election, fund the single most visible, most fundable, most photogenic piece of the system and skip everything that piece depends on to function. They fund the Innovation Hub and skip the deal flow that would give anyone a reason to show up to it. They fund the university research grant and leave the commercialization pathway broken, so the money produces papers instead of companies. They fund the fab and skip the ten-year workforce pipeline, then act surprised when it sits half-staffed.

Every one of those is the same mistake the founder-blame story makes but with more zeroes. Taxpayer zeroes.

Take a systemic problem, apply a linear solution to one node, and act shocked when the system doesn’t respond. The founder gets blamed for the failure of an ecosystem nobody funded, and the program gets a ribbon for producing an outcome it was never capable of producing alone. Both are ways of refusing to look at the system, which is the actual patient, while performing surgery on whichever organ happens to photograph best.

What “We Should Do” Instead

If you run economic development, invest early-stage, or build programs, the move is to stop making an excuse out of founders and start auditing the things around them.

• Staff mentorship by relevance, not availability. A founder raising a seed round should be paired with someone who has raised a seed round, full stop. If your mentor bench can’t clear that bar, your program’s first job is to go recruit people who can, not to pretend proximity to success is the same as experience of it.

• Measure your ecosystem the way you’d measure a company. Founder progression, capital deployed locally, whether exits recirculate into the next generation of founders, talent density in sectors where the region has a real edge. If you’re reporting cohort headcounts and event attendance, you’re measuring your own activity, not anyone’s outcomes, and you already know which one is easier to fake.

• Fund the belief infrastructure before the vehicle. Before you stand up a fund, ask whether the region has the culture, the stories, and the local operators that make deploying that fund rational. Capital dropped into a place that doesn’t believe in its own founders evaporates; capital dropped into a place that does, multiplies.

• Treat policy friction as an ecosystem input, not background noise. Every hour a founder spends on licensing, compliance, and politcy complexity that made sense for a corporation and no sense for a pre-revenue startup is an hour stolen from the only work that matters early, which is finding out whether anyone wants the thing.

• Widen who counts as a founder before you decide the pipeline is thin. A lot of the most capable builders in any region aren’t visible to the incubator circuit, didn’t come out of the flagship school, and don’t look like the founder archetype the ecosystem is scanning for. That’s a discovery and access gap, not a shortage of talent, and the regions that fix it find they had far more raw material than ever considered.

The founder in your city is more capable than your ecosystem is giving them room to prove.

The next time a promising company folds and the post-mortem lands on the founder, run the audit one level out. Ask what the mentorship actually was, who the money came from, what the program measured, and how much of that founder’s first year got eaten by friction nobody with a budget ever tried to remove. If the answer is “a lot,” then you didn’t watch a founder fail, you watched an ecosystem fail, and charged the founder for it.

To look at founding teams that imploded eighteen months in, as we do in our respective work, you will usually find the same wreckage; one founder who could not stop selling the dream, one founder who could not stop believing it, and a cap table that now belongs to lawyers or limbo.

The story spun is that they “weren’t aligned” or “had different visions.” We all acknowledge that teams often fall apart and cause the failure of a venture, but we never research why that’s the case and how to avoid it.

The more accurate story is that two people with specific psychological wiring found each other, gave each other exactly what each was missing, and then destroyed the thing they built together because the wiring that attracted them was never built to sustain a partnership.

That pattern is not random.

It is predictable, it shows up in dating and marriage with the same M.O., so, a startup economist and a founder therapist have been talking about it and decided to write this up, because it lives at the intersection of both fields. I’ll weigh in about what an entrepreneur actually is and why entrepreneurs gravitate toward a particular kind of cofounder. Monica will explain, academically, why that gravitational pull exists, why it feels like a perfect match right up until it isn’t, and what to do about it.

What an Entrepreneur Actually Is (and Why That Matters Here)

Start a company, file the paperwork, and the state hands you an LLC and the title “founder.”

That title is a designation; it is not an actual job, it is not a formal role, and it should hold no psychological distinction. It merely means, “person who started,” and though it is, being a founder should not be construed to convey anything else about a person.

In Startup Ecosystems, the case is made repeatedly that a small business owner and an entrepreneur are running two fundamentally different economic logics; the small business is designed for stability, predictable income, and conservative risk, while a startup is “a temporary organization searching for a scalable and repeatable business model under conditions of genuine uncertainty,” designed to fail frequently and pivot aggressively. The person psychologically suited to run the first is not automatically suited to run the second, and vice versa.

If we can agree on that, let’s push further on why the title of this section started with the word entrepreneur while I first drew your attention to founder. The entrepreneur, in the clinical and research sense, is a personality type; not a résumé line, and not a title you earn by incorporating. The traits cluster around risk tolerance, novelty seeking, low need for structure, high tolerance for ambiguity, and an almost compulsive orientation toward action over deliberation. Those traits describe roughly one in twelve people well, and they describe a specific neurological profile even better; which is where Monica’s field enters, because a large share of the people we call “entrepreneurs” are carrying a diagnosable version of the wiring the rest of the population carries in trace amounts.

Entrepreneurs, being wired for conviction and action, are terrible at one specific thing; they are terrible at sitting still long enough to notice that the person across the table is selling them a version of themselves rather than a genuine partnership.

The Draw Toward the Narcissist

Founders don’t wake up wanting a narcissist as a cofounder. They wake up wanting the thing a narcissist reliably projects; unshakable conviction, effortless confidence, the capacity to walk into a room of skeptical investors and behave as though the outcome is already decided. If you are a distracted, idea-generating founder who cannot finish a sentence without three new tangents (I say this with affection, and from experience), the person who radiates certainty looks like the missing half of your brain.

The engineer founder feels this even more acutely. She can build anything; she cannot bring herself to stand on a stage and oversell it, because overselling feels like lying and lying feels bad. So, when someone appears who can pitch, sell, fundraise, and glad-hand without any apparent discomfort, she reads that as complementary skill rather than as a warning.

In Rocket Fuel, Gino Wickman and Mark Winters describe the “Visionary” and “Integrator” pairing that makes companies work; “with no Visionary, an Integrator can’t rise to his or her full potential,” and vice versa. The tragedy is that the same complementary hunger that makes a genuine Visionary-Integrator pairing productive is precisely the hunger a narcissist is built to exploit, because the narcissist is fluent in becoming whatever the other person is missing.

The confidence is real, in a sense; it just isn’t attached to competence, accountability, or the capacity to weather being wrong in public.

And that gap is where the founding team eventually breaks. Monica can take it from here.

The Research — Monica Michael Smith

This is a timely post. Just this past week, the father of one of the most successful entrepreneurs I have ever worked with passed away. She was the reason I started a side business counseling and coaching entrepreneurs while I was still teaching at the university, so long ago. The relevance to this article is that our close bond has been so deeply severed by a narcissist that the two of us aren’t even speaking about this tragedy in her life.

My brilliant friend and I both carried the same easily exploited belief: that we could be pulled from the coal of hard work, networking, and content creation and emerge as the sparkling diamond business the world needs. A narcissist spoke with a level of certainty that made us both believe he was the gatekeeper to that dream. We believed him.

As is true of most Cluster B personality disorders (which Narcissism falls under), his pattern of triangulating our relationship, gaslighting, and running a smear campaign to cover his own inability to deliver on his promises left a fracture in our relationship that is still present today. What I hear from more founders and entrepreneurs than you’d expect is that this kind of toxic partnership steals something more painful than money. It steals the zeal, the trust, and the vision that once made life feel vivid.

Paul’s understanding of this pattern goes deeper than anything I could put into words. It’s felt, it’s seen, and it continues to show up session after session with the founders I work with. And that makes sense, because the majority of our decision making comes from the subconscious far more than we realize. Harvard Business School professor puts the figure at roughly 95 percent of our cognition happening below conscious awareness. As children, we create codes, small equations, to stay safe, loved, and secure. Sometimes those equations are inaccurate, and they keep computing into how we relate as adults. The founder who is drawn to the narcissist is usually running an old equation. So is the narcissist.

According to Ross Rosenberg’s The Human Magnet Syndrome: The Codependent Narcissist Trap, the pull between the giver and the taker, the servant and the master, the controller and the controlled is more than an idea. It’s emotional physics.

Entrepreneurs Run Higher on ADHD, and Why that is the Entry Point

The link between entrepreneurship and ADHD is no longer speculative. A large-scale study of more than 9,800 individuals, published in Small Business Economics, found that adults with clinically diagnosed ADHD were nearly twice as likely to act on entrepreneurial intentions as those without it. A 2019 companion study, also in Small Business Economics, documented meaningfully elevated rates of ADHD and related conditions across founder populations. A 2025 meta-analysis by Tran and Wiklund in Entrepreneurship Theory and Practice refined the finding into something uncomfortable: ADHD symptoms and diagnosis are positively related to entrepreneurial behavior, yet negatively linked to outcomes.

The wiring that pushes someone to start also tends to work against them when it’s time to finish.

In my practice, I see this show up as more than a productivity problem. It shows up as a relational one. Founders with ADHD often arrive already fluent in codependency, long before a narcissist ever enters the picture. Overextending, over-functioning, and seeking external structure and approval are common adaptations for a brain that struggles to self-regulate internally, and those same adaptations are the exact raw material codependency is built from. What looks like devotion or loyalty is frequently a maladaptive coping strategy the ADHD brain built years earlier, just to feel steady.

That’s what makes rejection sensitive dysphoria worth mentioning here, even though it isn’t a formal DSM-5 diagnosis. Dr. William Dodson, one of the clinicians who first described it, notes that roughly a third of his adult ADHD patients call it the most impairing part of their condition, an intense, almost physical reaction to perceived criticism or rejection. He describes it as so consuming that “a person can’t continue to function.”

Consider that with how a narcissist operates and the pattern becomes visible. The ADHD brain tends to experience intermittent approval with unusual neurochemical intensity. Rejection lands harder, so warmth after coldness can register as rescue rather than as manipulation. Founders with ADHD, in other words, don’t fall for narcissists because they’re naive. Their neurology makes intermittent approval disproportionately rewarding, and a narcissist’s cycle of idealization and withdrawal is almost perfectly shaped to exploit that exact dynamic, especially in someone whose codependent patterns are already primed to absorb it.

The founder isn’t weak. The founder is running a nervous system, and a set of coping strategies built to protect it, that a specific personality type is unusually good at playing.

Why the Narcissist Looks Like a Good Partner

Here is the part that founders resist hearing; the narcissist often is delivering value, at least early.

Grandiose narcissism, the confident, admiration-seeking variety, is not associated with low emotional intelligence in the way people assume. An analysis of 32 studies in Management Research Review found that emotional intelligence was positively related to grandiose narcissism and negatively related to vulnerable narcissism. The grandiose narcissist reads the room, mirrors what you need, and performs certainty with exceptional skill.

The catch is what that emotional intelligence is for. Research distinguishing self-reported from performance-based EI finds that narcissists report high emotional intelligence but perform worse on objective tests; the ability to perceive and use emotion is deployed toward managing their own image and extracting admiration, not toward genuine mutual understanding.

Studies of narcissism and relationship quality confirm the cost in relationships; narcissism mediates the relationship between emotional intelligence and the quality of interpersonal relationships, and it mediates it negatively. What wins the partnership is not what sustains it, it destroys it.

Why it Dissolves, in Marriage and in the Cap Table Alike

I tell my clients that nice isn’t good; someone can be endlessly charming, but when your reality doesn’t match their reality, are they actually good to you? This is where the codependent-narcissist and ADHD-entrepreneurial pairing, whether in a marriage or a founding team, follow the same arc.

The relationship appears to meet each person’s needs; the codependent partner gets someone to caretake and organize around, the narcissist gets a reliable source of admiration and labor, the entrepreneurial founder gets the executive confidence they lack.

It looks, from the outside, like each partner is benefiting from the relationship.

At least until the emotional intelligence gap starts to widen.

The partner with the disorder, whether narcissistic grandiosity or the emotional dysregulation that rides alongside untreated ADHD, sabotages the relationship precisely because the underlying deficit was never in charisma or drive; it was in the capacity to regulate, to repair after conflict, and to tolerate not being the center of the story. When the partner finally voices a need, or the founder finally pushes back, the narcissist experiences it as an intolerable threat and responds with the deflection, blame-shifting, and withdrawal that the psychological research on these pairings documents in detail. The partnership doesn’t fail because it stopped meeting needs. It fails because the person with the lowest emotional intelligence in the room cannot survive the moment the relationship asks something of them.

In a marriage, we call that a divorce. On a founding team, we call it a founder dispute, a forced buyout, or a company that quietly dies while two people who once finished each other’s sentences stop speaking entirely.

Same Mechanics, Startup Stakes

Paul again. What Monica describes in dating and marriage is not an analogy for startup teams; it is the same phenomenon. The reason cofounder conflict is one of the leading causes of startup failure is not that founders pick badly on skills. It is that they pick on unmet psychological need, and unmet need is exactly the vulnerability a narcissist is built to fill and then exploit.

I have argued elsewhere that teams should be selected on values first, will second, and skills last, because skills are trainable and the other two aren’t. What Monica adds is the personality underneath that hierarchy; when you select on skills first, you are unusually susceptible to the person who performs the missing skill most convincingly, and the most convincing performer of confidence, sales ability, and executive certainty is very often the person you should least want in your equity structure.

The pitch that dazzles you in the coffee shop is the same instrument that will be used against you in the board meeting when your partner can handle their emotions.

This is also why the pairing survives long enough to do real damage. It indeed works at first.

The distracted visionary and the certain operator ship something.

The engineer and the salesperson raise a round.

The apparent success validates the choice and buries the warning signs, right up until the company faces its first genuine test of whether these two people can regulate themselves through being wrong together.

The Fix Is Clinical

’s field has a name for the way out of this, and it is not “vet your cofounder harder,” (though you should).

It is emotional intelligence work, and what she calls empowerment therapy; the practice of noticing the old equation you are still running and rewriting it. The founder who understands their own rejection sensitivity stops mistaking a narcissist’s intermittent warmth for partnership. The entrepreneur who learns to hold a boundary stops selecting for people who need them to have none. Emotional intelligence is not a soft skill here; it is the capability to develop that breaks the pattern of so many startups failing.

I thought of a larger implication, one that drew me to want to write this with a professional, and it is one that startup ecosystems and economic development policy have almost entirely ignored.

If entrepreneurs run higher on ADHD, if that wiring draws them toward partners who predictably detonate the company, and if the difference between a founder who succeeds and one who fails often comes down to emotional regulation rather than market insight, then mental healthcare is not a wellness perk for founders. Mental healthcare is investment in innovation.

Let me restate that for the legislators in the room hoping to drive innovation: Mental healthcare is investment in innovation.

We spend public money on accelerators, grants, and incubators optimized for incorporation rather than for the psychological capacity to actually build; and we spend almost nothing treating the conditions, on both sides of these pairings, that consistently convert promising teams into failed ones.

• Treat the untreated ADHD

• Give the codependent operator the tools to stop caretaking a personality disorder

• Get the grandiose narcissist into the kind of work that builds actual, rather than performed, emotional intelligence

• Rethink the laws surrounding the agreements in partnerships, because our creators shouldn’t lose everything because of someone with a mental health disorder.

Every one of those interventions has a measurable return in ventures that survive their own founding team.

That is a distinct competitive advantage for an economy; one that results in innovation, creates wealth, and creates jobs.

The founders who build companies that last are rarely the ones with the most conviction in the room. They are the ones who did the work to tell the difference between a partner and a performance; which makes me curious, which one you went into business with, and whether the exhaustion you feel is the work, or the wiring of your team. Please, share.

Share Startup Economist

Nobody is coming to build your rural economy, and the people still waiting for a corporate headquarters to relocate to a county of 12,000 residents are waiting for a bus that stopped running in 1985.

That is not pessimism; it is the recognition that shows up in John Shepard’s work on rural entrepreneurship, where the planner and economic developer notes, “For too long, we’ve confused industrial recruitment with economic development. Nobody’s coming to save your hometown. And that’s OK. We can save ourselves.”

He is right about the recruitment critique, but related jobs data misleads people in most cities, causing a small business framing that cuts to the heart of why small towns struggle; the correction matters for rural strategy.

Most cite that small businesses (under 500 employees) account for about 62.7% of net new jobs since 1995, since that’s what the SBA Office of Advocacy promotes. The more meaningful data is about firm age, and it points jobs at startups specifically. Kauffman Foundation economist ’s 2010 study of Business Dynamics Statistics covering 1977 through 2005 found that firms in their first year create an average of 3 million jobs annually, while every other age group, from companies in their first full year to firms founded two centuries ago, collectively sheds about a million jobs a year (SSRN). and reach the same conclusion from the opposite angle in Where Will the Jobs Come From?; without startups, net job creation in the United States would be negative in all but a handful of years. It is firm age, not firm size, that drives net job creation, a point I’ve argued in Why Cities Should Invest in Startups and in my critique of size-obsessed state economy rankings. Age, of course, encompassing both new businesses and startups, meaning we need a far greater understanding and appreciation of the startups, since they draw the attention and capital, while making it easy (and affordable) to start new businesses.

Overwhelmingly most cities fail this. Worse, my goal herein, is that Rural communities suffer because of it.

Existing firms, the retention-and-expansion base that traditional economic development protects, are net job destroyers in aggregate; useful and worth keeping, but not where the growth comes from. Cities attracting that companies relocate are indeed manifesting local jobs, but they aren’t creating jobs and they aren’t creating any appealing reason for people to live and work there, other than the fact that there is a job (ironically, a job that a large company is effectively trying to eliminate).

Growth comes from new firm formation.

So, a rural strategy cannot be only about supporting the small businesses already on Main Street; it has to build the startup culture that produces new firms in the first place, while simultaneously supporting the survival of those new businesses (because, as we’ll get to, most of them fail). The rural difficulty is that both halves of that, the culture that produces founders and the support that keeps new firms alive, are structurally harder to assemble in low-density geography.

Here is where rural gets treated as a smaller version of urban, and that thesis fails every time. Rural entrepreneurship is not a metro startup ecosystem with less parking; it has genuinely different needs, a genuinely different labor market for the people who make ecosystems work, and genuinely different economics governing whether a person can afford to take the risk at all.

We should treat those three differences as the actual design constraints, because they are.

Difference One: Rural Markets Have Distinct Needs, Not Diluted Ones

The mistake is assuming rural challenges are urban challenges scaled down. Much the same as the mistake of thinking copying Silicon Valley is what’s going to work for you, rural entrepreneurship has different challenges that happen to occur in less dense geography. A tech startup in a rural county looks nothing like a tech startup in San Francisco, and Shepard points that out when he notes that these ventures “tend to have a unique profile in Silicon Prairie from Silicon Valley.” Rural opportunity runs through home-based businesses and services, artisan food producers reaching niche markets, and tourism ventures like boutique lodging and cultural festivals, alongside the occasional high-growth company.

Much of this work grows from what Shepard describes as asset-based community development, a strategy that leverages what a community already possesses rather than chasing what it lacks.

This is, not surprisingly, consistent with the pressure in the book Startup Ecosystems; that you can’t be a “tech” or “startup” ecosystem because what is valued is the focus on your unique strengths and opportunities. Leverage what your community already possesses rather than chasing what it lacks.

The three assets he has identified are the raw material of any rural strategy:

• Natural capital (the landscapes, resources, and climate a place already has, which no incentive package can relocate elsewhere)

• Cultural capital (heritage, traditions, and craftsmanship, the things that make a place identifiable rather than interchangeable)

• Human capital (the specific skills, relationships, and creativity already present in the population, including people who don’t yet know they’re founders)

That last category maps directly to what I call invisible talent; the founders who aren’t obviously founders yet, the people outside the existing network of “known” entrepreneurs and ecosystem leaders.

In a rural community, where the trust networks are smaller and denser, the person capable of building something is often the mechanic, the extension agent, or the woman running books for four different Main Street shops. The discovery problem is more acute in rural markets because there are fewer formal on-ramps that would surface that person to a program, and because the startup culture that would tell that person “you could build a company” is thinner to begin with.

Shepard’s framing on this distinction also lines up with a theme I’ve argued repeatedly, most recently in America’s Energy Belt; that ecosystems are built around what a region already does at industrial scale, not around generic “innovation” imported from a conference. A rural region that leans into its actual identity (a legacy of craftsmanship, an outdoor lifestyle, an Indigenous heritage) becomes, in Shepard’s words, “not only more resilient, but also more attractive to newcomers, retirees, and lifestyle migrants seeking authenticity, affordability, and connection.” The distinction is not decorative; it is the difference between a place capital can locate and a place won’t.

Difference Two: You Cannot Solve This With a Startup Program, Because the People Who Staff One Aren’t There

This is the difference most economic development plans refuse to confront, because confronting it means admitting the program-in-a-box doesn’t work in a county of 8,000 people (frankly, the program-in-a-box rarely works well even in a city of 800,000 people). And it is doubly binding once you accept that new firm formation, not existing-business retention, is what actually creates jobs; producing new founders depends on a specific input that is scarce in rural markets, and it is not money. It is people who have actually done the thing.

Access to startup-experienced people is the condition that gets most consistently misunderstood. Regions point to their business advisors, their university entrepreneurship centers, their corporate mentorship rosters, and claim the box is checked.

It usually isn’t.

The relevant experience is not experience helping small businesses or managing corporate innovation; it is experience working inside startups, making decisions under resource constraints, navigating the specific operational and psychological challenges that distinguish an early-stage venture from every other kind of organization. A mentor who has only worked in operating companies will default to corporate-scale advice that a startup cannot implement and should not try to.

Failed founders are more likely to succeed and raise capital during their second attempt. For the same reason, the mentors, program operators, advisors, and local startup community leaders, are the people who have been through the experience.

Now take that scarcity and apply it to a rural geography, where the density of experienced founders, senior operators, and startup-fluent talent is a fraction of what a major city offers. The mentor pool is thin. The “experienced team member you can hire away from the company down the street” doesn’t exist because the company down the street doesn’t exist. This is why parachuting an accelerator into a small town so often produces motion without outcomes; the methodology arrives but the capital density (the investor networks ready to fund what emerges) and the sector-experienced mentors (the elements that made the original accelerator model work), don’t transfer in the replication.

The rural answer is not to counterfeit density; it is to build the trust networks Shepard describes into deliberate connective tissue, and to import experienced operators through remote ecosystem development, fellowships, remote mentorship, and regional (not municipal) pooling of scarce expertise. A single county cannot assemble a mentor bench; a multi-county region sometimes can.

Difference Three: The Economics of Risk Are Different Because the Fallback Is Different

Entrepreneurship is a risk decision, and risk decisions depend on what happens if you lose.

Startup Ecosystems framed it this way, “Innovative employers are the safety net that makes risk rational. Founders who know that if their startup fails they can go back to a meaningful, sector-relevant job will take risks that founders without that fallback cannot afford to take.”

In a rural market, that fallback is frequently absent. There is no cluster of AI companies to rejoin if your AI startup folds; there may be no sector-relevant employer within ninety minutes. The affordability that looks like an advantage cuts both ways. As I write in the book, a startup in an affordable region raising a quarter-million-dollar seed round is competing against a startup in an expensive city raising a million dollars for the same market opportunity, and affordable regions also lack the density of high earners who can write angel checks from their own portfolios.

Cheap to live is not the same as easy to fund, and it is definitely not the same as safe to fail.

This is exactly where I want to bring in , associate director of research at the Center for Business and Economic Research at Ball State University and a rural sociologist who studies how households actually make ends meet.

Wornell offered the qualification the small-business-boosterism crowd needs to hear.

She agrees with the asset-based approach and the rejection of smokestack-chasing, then adds the context that is usually missing:

“We can’t lose sight of the fact that individuals take enormous risks when they open businesses, the majority of which will fail or close within 10 years of opening (up to 25% in the first year alone). Moreover, extensive research in this area shows that while entrepreneurs report higher job satisfaction, they typically also earn less, work more, and are more stressed than their non-business owning peers, all of which has long-term implications for them and their families.”

Her point maps directly onto the firm-age finding.

If new firms are the only reliable engine of net job creation, and most new firms fail, then a rural development strategy is betting on a high-failure activity by definition. They must. Wornell continued, “Smart development doesn’t mean supporting the opening of businesses; it means supporting the entrepreneurial endeavor from start to finish: from building business plans and marketing materials, to appropriate growth and budgeting. It means identifying and supporting people in our communities with great business ideas and inclination but without the capital and resources to make those dreams a reality.”

That is the invisible talent problem stated by someone who studies rural households for a living; the people with the idea and the inclination but without the capital and network are exactly the ones a discovery-oriented ecosystem is supposed to find. And her warning about “silver bullet” solutions is the same warning I’ve built into my critique of programs that generate activity without changing conditions. Wornell’s caution: “Small business development at all costs will turn out exactly like industrial recruitment at all costs; a losing proposition for everyone.”

What Belief Does to Capital

Now to what communities get most wrong, and the reason I keep writing about this.

Rural regions, like most struggling ecosystems, diagnose their problem as a shortage of money. It almost never is.

A city seeking venture capital for its founders without first examining the structural gaps that prevent venture capital from forming locally is not solving a capital problem; it is misdiagnosing one.

Capital is a symptom, not a cause. It shows up where invention and signal already exist, a point made about Richmond’s ecosystem and one that applies more in rural markets, “Capital follows value; it does not create it. Fundraising is a symptom, not a strategy.” The root cause of capital not appearing is rarely the money itself; it is the culture, the expectation, and the narrative that determine whether anyone believes building something ambitious in this place is worth the risk. Follow the belief problem, not the money problem, and the sequence starts to make sense.

Rural communities can, in fact, thrive before the venture capital arrives, because the venture capital was always going to be downstream of the conditions anyway.

Ecosystems compound through a feedback loop between value creation and capital formation; companies create real value, exit, and recycle both the money and the experience back into the next generation of founders and investors. That loop is fragile in its early stages, and in rural markets you build the early stages with the assets you have, not the capital you wish you had. The plumbing comes first. Simplified fund-formation rules, procurement pathways that let a local startup win its first small government contract as an early-revenue signal, sector concentration around what the region actually does, and honest, distinct promotion of that sector. None of these require a check to get started, and all of them make the eventual check more likely.

A Model That Gets the Sequence Right: Patterson Fellows

I want to feature a program that appears to understand the three rural differences instead of ignoring them.

The Patterson Family Foundation has launched the Patterson Fellows program, a three-year fellowship built to support high-caliber entrepreneurs solving systemic challenges in rural communities.

It reads less like a grant and more like an attempt to manufacture the missing conditions on purpose, and notice that it targets new-venture founders rather than the existing-business base, which is the right target if job creation is the goal.

Selected Fellows receive:

• $100,000 per year for three years, a $300,000 total investment per Fellow, in annual cohorts of five

• Access to mentors, industry experts, and experienced board members

• Participate in the Pipeline Fellowship program

• Leadership development and quarterly curriculum modules

• Connections to investors, ecosystem partners, and corporate leaders

• Grant writing and storytelling support

• Access to the Patterson Family Foundation’s broader network and resources

Priority areas include health, agriculture, housing, connected commerce, and systems change. Fellows can be located anywhere, but their solution must focus on the foundation’s catchment; the rural counties of Kansas and western Missouri with fewer than 50,000 residents.

Application deadline is August 23rd at midnight. Learn more here.

Notice what the money is buying. It is not just risk capital; it is three years of the fallback problem partially solved (a founder who knows $100,000 arrives annually can take risks a founder living quarter to quarter cannot), plus deliberate importation of the mentor density that rural geography does not naturally supply, plus network access that substitutes for the local trust networks.

Jill Meyer, senior director of entrepreneurship at the Patterson Family Foundation and a leader of the Patterson Fellows initiative, framed the intent, “We are thrilled to launch the Patterson Fellows Program and support cutting-edge innovations that address the systemic challenges facing rural communities. I look forward to learning about the technologies, ideas, and solutions being developed by visionary leaders who are committed to making a meaningful difference.”

The program answers Wornell’s challenge, too. It is not “support the opening of businesses at all costs”; it is a three-year, start-to-finish structure with curriculum, mentorship, and the explicit goal that each Fellow finishes with a self-sustaining business. That is what supporting the entrepreneurial endeavor from start to finish looks like when someone actually funds it, and it is aimed at exactly the high-failure, high-payoff activity (new firm formation) that the jobs data says matters most.

What Rural Communities Should Do

I’ve spoken with most of the major family foundations throughout the United States and I hope you will take a moment to share this with them, should you have such connections. All of them are passionate about solving the challenges of rural entrepreneurship, few of them are doing it as intentionally or meaningfully as the Patterson Fellows program. Let’s change that.

More, stop asking where the capital is and start building the conditions that produce new firms and keep them alive, in the specific order that rural constraints demand:

• Inventory the assets, not the deficits. Name the natural, cultural, and human capital already present; Shepard’s asset-based frame is the correct starting diagnostic, and it costs nothing to run.

• Build the startup culture, not just the small-business support. Retention and expansion of existing firms keep a community stable, but existing firms don’t create net new jobs; new firms do. Both matter, and only one of them requires deliberately manufacturing belief that founding a company here is a rational thing to attempt.

• Solve the fallback before you recruit founders. Identify or attract the anchor employers and remote-work pathways that make failure survivable, because without a survivable downside the risk math tells rational people to stay employed.

• Import the experience you cannot grow locally. Pool startup-experienced mentors regionally, use fellowships like Patterson’s to bring in operators, and stop pretending a business advisor is a startup advisor.

• Find the invisible founders. Build deliberate on-ramps for the people with the idea and the inclination but not the network, because in a small population the next founder is almost certainly someone no program has met yet.

• Fix the plumbing before you chase the check. Simplify the structures, open procurement to early-revenue signals, concentrate on the sector you actually have, and let capital find the signal you’ve made legible.

The rural regions that will thrive are the ones that stop treating their smallness as a capital deficiency and start treating it as a design problem with knowable inputs.

Wornell is right that betting a community’s future on people’s dreams and risks is only smart development if it benefits the owner and the community both. The way you make sure it does is by building the conditions that make the risk rational before you ask anyone to take it, and by accepting that the jobs you want come from the new firms you’re afraid to bet on, not the established ones you’re comfortable protecting.

If your county has been waiting for the money to show up first, ask instead what a founder in your community can already fall back on, who can actually mentor them through the parts that break, and whether anyone has gone looking for the person with the idea who would never call themselves an entrepreneur. Those three answers determine more about your rural economy than any grant you’re still waiting to hear back on.

A couple of weeks ago, someone I trust pointed me to Aravind Srinivas on the Joe Rogan experience and said the whole discussion was worth it, that he talks about U.S. innovation, our culture, and our venture capital ecosystem. She was right, and it got me thinking hard enough to write this.

Srinivas’s perspective is that the American dream, the thing every founder abroad romanticizes, isn’t a policy or a funding mechanism at all; it’s a temperament a place learned and kept. “I always thought America’s the only country where you can come here and have an idea, and people listen to you and encourage you to go pursue it. The risk-seeking culture is just incredible, everywhere else, you kind of are either explicitly or implicitly forced to defer to authority.” The CEO of Perplexity, Srinivas shared in one breath where most economic development strategies get it backward.

The Instrument Is Not the Music

Every economic development office that ever tried to “attract venture capital” made the same mistake a person makes buying a piano and expecting to become a musician.

The instrument is not the music. Venture capital is not a resource you import, a fund you stand up, or an asset class you court with a tax abatement; it is the downstream artifact of a culture that already produces companies worth funding.

Srinivas said as much when asked why the rest of the world doesn’t simply adopt the values that so many people cross oceans to reach. His answer wasn’t an echo of what I hear most city officials proclaim, “better incentives” or “more capital,” it was, “It’s hard. You know, a lot of it is cultural,” and then he added something notable about the United States, explaining the draw to come here, “People who get rich here actually want to encourage and be part of somebody else’s crazy journey. Because it’s hard to pursue all crazy bets yourself. So, it’s an ecosystem. And once something becomes an ecosystem, there’s network effects. So, it’s very hard to copy that elsewhere.”

He’s describing a compounding social asset, not a financial one. Capital is the sediment left behind by a culture that had already been building, failing, and circulating talent for decades before anyone romanticized it. This is exactly how Silicon Valley actually formed, and the real sequence runs opposite to the mythology. Defense procurement during and after World War II concentrated advanced research contracts in the region, and those contracts demanded technical performance rather than symbolic innovation; firms like Hewlett-Packard built real products for demanding customers who required things to work. Revenue preceded the venture mythology by decades. As I’ve argued in why Silicon Valley cannot be copied, venture capital evolved locally because outlier returns occurred locally first, and the exits recycled experienced operators into investors who funded the next batch of experiments.

Failing Up Is a Credential, Not a Confession

Failing up is a hinge the entire culture of entrepreneurship turns on, and Srinivas quantified it with a Jeff Bezos line, “Where else would you be able to go raise a few million dollars for an idea that has a 5 to 10% chance of working and then fail at it and still go and raise another few million dollars for your next idea. Nowhere else.”

That is not a description of an abundant capital market; plenty of countries have capital.

It is a description of a culture that has decided a smart failure is a credential rather than a scarlet letter.

This is the part outsiders never quite believe until they’ve lived it. Worse, this is the part criticized by people who don’t live in it because it isn’t ideal to their needs; the rub in that is that criticism of this is WHY so many regions of the world mitigate the risk. In most of the world, a failed venture is a mark you carry; in startup communities that work, it’s a line on your profile that makes the next venture easier, because you’ve been tested and you came back. Venture capital is simply the financial expression of a population’s collective tolerance for asymmetric bets. Where that tolerance is high and demonstrated repeatedly, capital forms to service it. Where it’s low, no amount of imported fund-of-funds money changes the underlying temperament.

Failing up isn’t recklessness rewarded; it’s a culture pricing courage correctly.

Defining What’s Doing All the Work

Culture is doing more work than most people give it credit for. Culture in an ecosystem is not the coffee shops, the co-working aesthetic, or the number of founder happy hours on the calendar. Culture is the set of beliefs a place holds about risk, ambition, and permission.

It is whether a talented engineer at your region’s biggest employer believes that leaving to start a company is a rational, respectable move or a reckless one that will embarrass her family.

Regions where the dominant culture is passive, academic, or risk-averse produce entrepreneurs who ask permission, and asking permission is not a startup trait.

For the Systems Thinkers in the Room

If you read an ecosystem the way an engineer reads a schematic, and I suspect a lot of you do, this reframing hopefully reads as good news rather than fatalism.

That, culture sounds soft, unmeasurable, impossible to influence, which is why analytical people dismiss it and reach for the lever they can pull today, usually a fund. But Srinivas isn’t describing a mood; he’s describing a system with feedback loops, network effects, and initial conditions. That is a designable object. You cannot mandate a culture into existence, and you cannot buy one, but you can absolutely engineer the conditions under which one compounds, which is the entire premise of the six considerations of the economic development of startups.

A system runs on a loop that only turns one direction, like this flywheel analogy I love to use to help founders understand fundraising. Companies create real value, eventually exit, and capital returns to founders and early investors who then become the next generation’s angels, mentors, and fund managers. Experience disseminates. Capital becomes locally informed rather than externally imported, better calibrated to the specific conditions of the place. The ecosystem strategist’s job is not to inject capital at the end of that loop; it’s to protect the early, fragile turns of it and stop suffocating them with programs that reward activity over consequence, a failure mode I’ve called the startup theater. Systems thinkers get this immediately; that if you can’t optimize the output variable (i.e. better startups or more funding), you fix the constraint upstream, and the constraint here is almost always belief, expectation, and narrative rather than dollars.

Capital Is a Mirror, and So Is Everyone Else

Srinivas’s argument connects to something more human than portfolio theory. He spent a good portion of the conversation on what he calls the “curiosity premium,” the observation that “the most effective people, the most successful people have always been the most curious people, the ones who have been good at asking the best questions.” His reasoning for why that trait pays off isn’t what you might expect; it’s about relationships. “People who continuously ask questions tend to do better. They make more money. They have a higher quality of life. They have more compounding relationships. People find them more interesting, and so they compound their relationships over time.”

Compounding relationships. That phrase is the missing piece in most capital-formation strategies; I know it, because almost all of you talk about referrals as though the “warm intro” swill is actually how startups get funded.

Capital flows through trust networks before it flows through pitches; the research bears this out. Studies of investor behavior consistently find that funding decisions track social proximity and familiarity, a pattern sociologists call homophily, the tendency to trust and back people who resemble oneself. One analysis describes how investors may be more likely to fund entrepreneurs with similar backgrounds, experiences, or characteristics, even when that preference is not consciously acknowledged. Capital behaves like a social organism, not a calculator; it clusters where relationships already exist.

The optimistic and strategic implication is that a region’s most underpriced asset is not a bigger fund; it’s a denser, warmer, more curious web of relationships among the people who build.

This is why “invisible talent” matters so much. Invisible talent doesn’t mean marginalized or underrepresented people; that’s a separate conversation with separate intentions. Invisible talent means the founders who aren’t obviously founders yet: the domain expert running operations at a manufacturing firm who has quietly solved a problem worth a company, the engineer three layers deep in an anchor employer who has never once been to a pitch night. They’re invisible because the discovery mechanism is a relationship network before it’s a financial one, and the network was never built to see them.

Fix the relationships and you widen the aperture; the capital follows the people, as it always has.

Why “Attracting VC” Almost Always Fails

A region decides it wants venture capital, so it launches a government-backed fund, flies delegations to San Francisco, or offers incentives to lure a coastal firm to open a satellite office.

Can you see why this is backwards? Why every city focused on and saying they need to prioritize venture capital, is doing it wrong?

It has tried to import the conclusion while ignoring the culture that produces it. Tune in to Srinivas’s chat because he illuminated it differently than I have been pushing: you can move money across a border in an afternoon, but you cannot move a network effect, and you certainly cannot move a population’s relationship to risk.

When you do that, and I have countless experiences with cities trying this, the capital shows up, finds no dense web of trust to flow through, no repeated pattern of celebrated failure to reference, no compounding relationships to underwrite, and it behaves accordingly, which is to say timidly or not at all.

The better strategy is patient and mostly free.

Build the conditions that make a founder’s bet rational, principally:

1. An anchor employer whose presence makes the downside survivable, since founders who know they can return to a meaningful, sector-relevant job take risks that founders without that fallback cannot afford.

2. Reward the curious questioners rather than the confident answerers, a point Srinivas makes about education that applies just as well to ecosystem design when he argues “the smartest person in the room is the one who asks the most interesting questions.”

3. Protect the early turns of the value-to-capital loop.

4. Help the relationships compound by promoting things, removing silos and barriers, and creating connections

Do that, and the capital precipitates out of the culture the way it did everywhere it ever actually worked. Skip it, and you’ve bought a very expensive piano.

When a mayor or an economic development director asks me how to attract venture capital, I tell them the question is already pointed the wrong way. The right question is what kind of culture your region has toward risk, ambition, and the people who aren’t obviously entrepreneurs yet, and whether your relationships are dense and warm enough that capital has somewhere trusted to flow. Srinivas is right that this is hard, and he’s right that it’s cultural. He’s also, in an encouraging sense, describing something you can build, one relationship and one survivable bet at a time, which is the only way a culture ever gets built anywhere.

If your region has been chasing funds and delegations for a decade and still watches founders take the first flight out, the constraint was never actually the money; it was the culture the money reflects, and culture is the one thing a committed community actually gets to author. How to start? Who in your ecosystem is building something and doesn’t yet know anyone who’d back them, and what it would take not to refer them, but to change the circumstances so that investor KNOW and WANT them?

Most conversations confirm what you already believe; you walk in with a point of view, you make your case, the other person agrees or pushes back, and you leave more or less intact. Twenty-five years of doing this work has given me a lot of those conversations.

Some conversations are different. They don’t argue with what you know; they reorder it. They hold up a mirror you weren’t expecting and say, look at what you’re actually doing here, and look at what it costs when you do it alone.

Then I sat down with .

Ruthann hosts Unlock Your World and wrote the book by the same name: A Funny, Heartfelt, and Seriously Effective Guide to Happier People and Better Teams. Her work is built on a simple argument; that kindness is not a character trait we hope people happen to have, it’s infrastructure: Something you design, build, and hold accountable to outcomes. The teams that genuinely function, the workplaces people actually want to stay in, the cultures that produce something real rather than just describing themselves as good, are built on self-love, authentic connection, and the discipline of honest communication. Not despite those things, because of them.

I walked into the conversation shared here thinking we were in adjacent fields; I walked out understanding we had been working on the same problem from opposite ends of the building for our entire careers.

Tune in on YouTube if Preferred

Ninety Percent – For Thirty Years.

If you’ve followed my work for any length of time, you already know the number. Ruthann knew it too, and she didn’t need me to explain why it matters. What she did, almost immediately, was connect it to something I haven’t frequently framed before; that the environment removes potential from people the same way the startup ecosystem removes potential from founders.

Same mechanism. Different room. That startup ecosystem failure drives founder failure.

Most startups fail… er go…

That reframe in our conversation was a perfect alignment between us because economics isn’t just the study of how money moves, it’s the study of how systems function or break down, how policy helps or harms, how the environment shapes outcomes that we then mistakenly (or intentionally) attribute to individuals. About 20 years ago, I arrived at a troubling conclusion; that a meaningful share of startup failure isn’t about bad founders, it’s about bad conditions. Conditions we created, conditions we’re maintaining, and conditions we largely refuse to examine honestly.

Ruthann has spent her career in education and organizational leadership and arrived at the same place from a different direction; that most systems, schools, companies, teams, families, aren’t failing because the people in them lack potential. They’re failing because the environment removes it.

The Three Things We Know, and Keep Ignoring

When I laid out the three primary causes of startup failure in our conversation, Ruthann didn’t push back, she nodded. She had seen all three, in different forms, in classrooms and staff rooms and the organizations she has advised.

The first is overregulation. Legislators who don’t understand what entrepreneurship actually requires write policy that works against it. This isn’t malicious, it’s ignorance; well-intentioned ignorance that produces damaging outcomes. Most of my policy work lives here; in translating the economics of innovation into language that lawmakers can act on before they accidentally make things worse.

The second is marketing illiteracy. I’ve made this argument more times than I can count, that marketing is not advertising, that most founders skip the discipline entirely and go straight to running ads, and I’m going to keep making it until the failure rate moves. It hasn’t, so here we are. What Ruthann added was the human version of the same problem; that most leaders “market” their culture the same way bad founders market their product. They announce what they’ve built and wait for people to believe it. The consumer landscape has become extraordinarily complex, people are reachable in thousands of different places now, and leaders face the same fragmentation inside their own organizations. Knowing where and how to actually reach people, whether customers or employees, is a real discipline, not a weekend project and not a feeling.

The third cause is one that makes people uncomfortable: personality. Substantial research, including work correlated with findings out of Oxford, identifies the specific personality traits that predispose someone to entrepreneurial success. There are also traits that are statistically linked to struggle in this environment. We know this yet we largely ignore it because we prefer the story that anyone can be a founder. I suppose that anyone *can* but failing to acknowledge the differences in how people are wired results in generic startup development organizations and investors who judge based on pitches rather than capabilities. Some people are far more likely to succeed; some people are genuinely wired to sell, to fundraise, to operate under radical uncertainty, to make decisions with incomplete information. Others are brilliant in ways that are not those ways, and that brilliance does not transfer automatically.

A software engineer who builds something extraordinary but cannot sell it, cannot raise capital for it, and cannot articulate why the market needs it is not likely to succeed. Not because they aren’t talented! Because the talent they have isn’t the talent the situation requires. Seeing that clearly, without pity or apology, is the beginning of actually helping someone.

Nice Is Not the Same as Kind

Ruthann draws a distinction I have been reaching for without exploring it enough to get the point across in entrepreneurship; that being nice is not the same as being kind.

Nice is agreeable. Nice is encouraging and supportive.

Nice validates what’s already in front of you. Nice is the advisor who tells a founder their pitch is great when it demonstrably isn’t. Nice is the manager who gives the promotion to the enthusiastic employee because enthusiasm is visible and measurable performance sometimes isn’t. Nice is the policymaker who funds the program that looks good without asking whether it does good.

Kind has boundaries. Kind delivers truth in service of someone’s actual growth. Kind is harder, less comfortable, and far more useful.

In my work, too much niceness is actively damaging. Founders get encouraged when they should get redirected. Ecosystems get celebrated when they should get diagnosed. Startup programs receive continued funding when the data, if anyone looked at it realistically, would suggest the model isn’t working.

When I sit down with a founder for the first time, I ask them directly, “do you want to tell me what you’re doing because you’re excited about it, or do you want me to fix it?” Because in 100% of cases, something important is broken. That’s not a performance of toughness and it isn’t cruel criticism, it’s arithmetic. If 90% of ventures fail and I’m talking to you in the early stages, the probability that everything is working correctly is ZERO. The kindest thing anyone can do is help you find what isn’t working before the market does.

Ruthann has built an entire philosophy around this; she calls kindness infrastructure, not ornamentation. The organizations that embed honest feedback, genuine care, and authentic communication into how they actually operate don’t just feel better to work in, they perform better, because they’re working with real information instead of the curated version everyone agreed to believe.

Transparency Is a Competitive Advantage

One of my favorite moments in our conversation came from her asking about transparency and honesty, specifically the gap between how founders describe their companies and where their companies actually are.

Most early-stage pitches I hear describe a fully operational venture. The business model is settled, the platform is built, and the team (such as it is), is in place; then you ask a few questions and discover it’s two people, six months in, who have decided what they hope all of that will eventually look like.

The problem isn’t optimism; optimism is a feature in this industry. The problem is that advisers, investors, and potential partners can only help you based on what they actually know. Fabricated clarity produces genuine misdirection. When investors say, “let’s reconnect in six months,” it’s rarely because timing is bad. It’s because they figured out that the story doesn’t match the stage, and they’d rather wait for the story to become true than explain why it currently isn’t.

Setting the stage accurately (here’s where we are, here’s what we have, here’s what we’re genuinely uncertain about) is not a vulnerability; it’s the prerequisite for getting real help.

Ruthann made this point through her own experience; the people she has valued most, in classrooms and in professional settings, were the ones who told her the truth. An employee who came to her having interviewed elsewhere and said simply, “I looked around. I’m staying;” that’s loyalty built on clarity, not performance. And it’s only possible when the leader created the kind of environment where that honesty felt safe to offer.

Why Schools Are the Problem, Not the Solution

So, our conversation went this direction, exploring a passion of mine that parallels my work. Ruthann spent her career in education because, growing up, she didn’t quite fit in with the way school is designed. And I respect that motivation more than almost any I can think of; she went back to fix from the inside what had failed her from the outside. I had the same experience. Traditional education didn’t work for me, not because I was incapable, but because it kept answering questions I hadn’t asked while ignoring the ones I was actually asking.

Here’s what I believe, and the research supports it; the American school system was deliberately designed to produce workers for the industrial economy. That was the explicit intent in the post-war era! The structural architecture of public and university education hasn’t meaningfully changed since. Standardized tests, grade-level compliance, the scantron mentality don’t just fail to develop entrepreneurial thinking. They actively condition it out of people.

Think about a three-year-old.

Every child draws, every child builds, and every child is endlessly creative and endlessly curious, operating without anyone telling them they can’t be. Those tendencies, the building, the questioning, the willingness to try something and see what happens, are the raw material of entrepreneurship. The system schools those tendencies out methodically, over twelve or sixteen years, until we have a workforce that is confused by the invitation to take initiative.

And then we run startup programs and wonder why the output is poor.

Ruthann’s instinct as an educator was to create space for the different, to let students read upside down in their chairs if that’s how they absorbed information, to prioritize the outcome, which is learning, over the process, which is compliance. That isn’t just a teaching philosophy, it’s an operating philosophy, and it applies everywhere people are trying to build something that didn’t exist before.

Silos Are the Economy’s Most Expensive Habit

I have an entire section of Startup Ecosystems on silos, and I’m glad Ruthann raised it, because I don’t think most people appreciate how much damage they actually do.

Universities run parallel research programs with public funding, never comparing notes, because collaboration might compromise their competitive positioning in grant applications. Startup accelerators in the same city fight over the same pool of founders and collecting exclusive mentors, competing for volume rather than cooperating for impact. Investors often operate in their own networks, unaware of other capital sources that might be more appropriate for the ventures they’re evaluating. Information that could accelerate everyone gets trapped behind organizational walls that exist for reasons nobody can articulate anymore.

The internet was supposed to fix this; when it arrived, we called it the information age, free access to all the world’s knowledge, on demand, in real time.

What we got instead was the silo age; the Google silo, the Facebook silo, the TikTok silo, and the SaaS and Apps silos. Each capturing data, each generating intelligence, each declining to share meaningfully because the data is the product.

What Ruthann does in her work is break down the human version of this. What I do in economic development is break down the organizational and data version. In both cases, the approach is the same -> name what the silo costs, make that cost visible and personal to the people inside it, and then help them see what becomes possible when the walls come down.

Most people don’t want to do damage; they just haven’t been told clearly enough that what they’re doing is doing damage. Once they genuinely understand it, not defensively, but really understand it, they want to fix it. The door open and that’s where the work begins.

The Real Signal That Something Is Working

Near the end of our conversation, Ruthann asked for a perspective on the biggest signal is that a culture is actually healthy, not just described as healthy, but genuinely functioning. How do we avert startup ecosystem failure, for example.

My answer doesn’t come from a dashboard.

In startup ecosystems, I look at the recycling of capital. Do the people who have wins, exits, successes, meaningful returns, put money back in? Do they stay engaged, continue mentoring, continue investing, continue showing up? Or do they take their success and disappear into a larger house and a quieter life?

When they disappear, it tells you something important; that the environment wasn’t worth participating in beyond the transaction.

When they stay and reinvest, it tells you the ecosystem created something worth contributing to, worth being part of, worth building more of.

In organizational culture, the parallel is direct. Do your employees participate beyond what’s required of them? Do they come to the optional event? Do they take on the project that wasn’t in their job description? Do they show up to the conference because they’re genuinely curious, not because someone is watching?

If yes, something’s working.

If you have to mandate everything, incentivize everything, track whether people are showing up to the things that are supposedly enjoyable, you already know the answer.

Kindness Is Not a Personality Trait

I came into this conversation prepared to talk about ecosystems and failure rates and root-cause analysis of why startups struggle.

I left thinking about something Ruthann said that sounded almost simple; that kindness should not merely be a character trait we admire in people who happen to have it, it should be part of the infrastructure we build deliberately throughout society.

You don’t hope roads appear, you design them, fund them, build them, and measure whether they actually connect the places they need to connect. Kindness as infrastructure means you don’t leave communication, honesty, and genuine human consideration to personalities that want to be nice while hoping you working with the people who actually help you. You build systems that make honest feedback normal, you create environments where transparency is the default, not the exception people have to be brave enough to choose. You measure whether your culture is actually working, not with a survey that captures what people are willing to say, but with behavioral signals that show what people actually do. You keep education and startup ecosystems critical, constructively critical, to keep people oriented to what it means to be creative, risk taking, individuals who learn from a kind society that recognizes reality such as it is.

That’s what Unlock Your World is building toward, I find; not a motivational framework, but an operating philosophy for how organizations could actually function if they took human connection as seriously as they take quarterly targets.

Me, walking into rooms to tell people their ecosystems are broken, and a kindness advocate who wrote a book about making workplaces genuinely human have been working on the same problem all along.

I would call that alignment; and as Ruthann and I kept agreeing with one another, alignment is where everything starts.

Richmond ran the first commercially successful electric streetcar system in the world in 1888, and most people who live there today have no idea.

Frank Julian Sprague, perhaps the founder who seeded Richmond startups, wired twelve miles of hilly, muddy track and made electric traction work at scale when the rest of the planet was still arguing about whether it could be done; the IEEE later designated it an engineering milestone, and GE bought the business two years later. That is what we study before anyone talks about accelerators, grants, or pitch nights; Richmond has always been a place where somebody solves a hard problem under pressure, and the rest of the country adopts the answer. The question for founders, investors, and economic development professionals in Central Virginia is not whether the region can produce that again; it demonstrably has, repeatedly. The question is whether the ecosystem being built around it is designed to reward consequence or merely to reward activity.

A History of Solving Problems Nobody Else Would Touch

More than Sprague, Richmond’s founding story is not a single triumph; it is a stack of people building economic power where the system was structured to deny it. Maggie Lena Walker chartered the St. Luke Penny Savings Bank in Jackson Ward, a historic district in Richmond, in 1903, becoming the first Black woman to establish and serve as president of a bank in the United States. She did not do it as a charity project; the Federal Reserve Bank of Richmond explains Walker’s strategy as the deliberate use of economic power, redirecting money saved from a streetcar boycott into Black-owned businesses. That is an entrepreneur’s instinct; here, identifying a market denied access to capital, building the institution that serves it, and enabling others to create wealth and opportunity.

Jackson Ward earned the name “the Harlem of the South” because that entrepreneurial instinct was dense there; banks, insurers, a department store, a newspaper, all within blocks of each other. Density is the thing that makes ecosystems work, and Richmond had it in one neighborhood a century before anyone was writing papers about “innovation districts.” The lesson is not nostalgia. The lesson is that Richmond’s creative capacity has never been the constraint. What gets constrained, over and over, is whether the capital, the policy, and the narrative line up behind that capacity or fight it.

Invention as the Region’s Native Language

Sprague’s streetcar and Walker’s bank are bookends of the story here. Richmond and its surrounding corridor have produced advances in transit engineering, finance, pharmaceuticals, and consumer products, and the modern version of the story is being written in health sciences and fintech. Invention is the input that drives entrepreneurship and, eventually, draws capital; it is not the other way around. Investors do not arrive and conjure inventions out of a region that has none. They follow the signal that invention throws off. Richmond throws off that signal reliably; the work of an ecosystem is to make sure the signal is legible to the people holding checks.

Why a Region an Hour from Washington Should Think About Policy

You might already know from my work in Austin, or other ecosystem assessments, that I am a critic of city-centric startup ecosystems; you might think regionally because a city in-and-of-itself is biased and centric in ways that are irrelevant to founders and investors.

Here is the geography that makes Richmond unusual. Washington, DC sits roughly a hundred miles north; Norfolk and Newport News anchor the Hampton Roads corridor to the southeast, home to 757 Accelerate and the 757 Collab thanks to the VIPC | Virginia Innovation Partnership Corporation; Charlottesville and its 434 Catalyst accelerator sit to the west. Richmond is the center of gravity for a regional triangle that includes the seat of federal economic policy, a major defense and port economy, and a research-university corridor.

Most cities have to manufacture a reason to care about federal policy. Richmond commutes to it.

That proximity matters because entrepreneurship in America is being shaped by policy decisions that most founders never read until the decisions squash them.

For example. In June 2026, SBA loan-eligibility changes tied to immigration status were reshaping who can access startup capital, at a moment when, by the nonpartisan National Foundation for American Policy’s estimate cited that immigrants and their children have launched two-thirds of America’s billion-dollar startups.

When the federal government adjusts who qualifies for the most accessible tier of business credit, it is making a startup-formation decision whether it calls it that or not.

I will continue to argue at length that governments confuse motion with consequence, funding the visible and avoiding the structural. In The Missing Infrastructure of Entrepreneurship, “local policy keeps looking like a real estate brochure or a conference schedule when it should look like infrastructure that works as hard as the entrepreneurs it claims to support.” The reason activity wins over structure when cities and states are trying to help entrepreneurs is not stupidity; it is incentive. There is no ribbon cutting for simplifying a securities exemption. Which is exactly why a summit of ecosystem builders, sitting an hour from the agencies that write the rules, is a rare chance to talk about the rules instead of the pep rally.

The Policy Levers That Actually Move Founders

Immigration, trade, healthcare, and the SBA are not separate topics; they are four faces of whether the environment makes the downside of entrepreneurship survivable.

Yes, downside. Most entrepreneurs struggle and fail; efforts to encourage and support them, without addressing the causes of hardship and difficulty, is little more than setting people up to fail.

Immigration policy determines whether the most entrepreneurial population in the country can legally build here; National Bureau of Economic Research work published on Canada’s Start-up Visa found that immigration policy directly changes where immigrant founders choose to locate, meaning the US loses founders to jurisdictions with clearer paths. Trade policy determines whether a hardware or manufacturing startup can source and sell across borders without pricing itself out. Healthcare policy determines whether a founder with a family will leave a salaried job to take a risk, because in America health coverage is still largely bolted to employment. And the SBA determines who gets the first, cheapest dollar of credit. While we’re in this article for Richmond, given the proximity, we can’t ignore that the federal administration could shape a bold economy for startups, trade posture and the design of public-private partnerships either level the field for founders or tilt it toward incumbents.

Entrepreneurship is the largest single engine of net new job creation, and jobs, frequently celebrated or worrying, are actually a lagging indicator, not a lever we pull directly. We do not create jobs by demanding them. The White House celebrated job growth is not a result of the President. We create the conditions under which companies form, and the jobs follow the productivity.

A policy conversation held in Richmond, within arm’s reach of Washington, that treats founders as the mechanism of job creation rather than as its beneficiaries would be worth more than a dozen innovation-hub announcements.

The Richmond Summit: Startup Champions Network Comes to Central Virginia

The Startup Champions Network Richmond Summit runs September 29th through October 1st, 2026, in Richmond, and it is not a conference in the tradeshow sense. Built around ecosystem builders, the people who run the accelerators, incubators, funds, and community organizations, rather than around sponsored keynotes, SCN is direct about the goal of being there, “pitch-free and built for ‘vulnerable practice over performative success,’” which is a rare and valuable posture in a field addicted to good-news press releases. The theme is “In Community,” chosen because Richmond earned it; the summit will ground attendees in the history of Jackson Ward, walk the Arts District, and put ecosystem builders in rooms with the founders actually shaping the region.

The local hosts anchor that. People like Ashley Ray who has supported communications and planning for SCN across thirteen summits and works on the ground in Historic Jackson Ward; Debbie Irwin, now Executive Director of Lighthouse Network, centers her work on developing the whole founder rather than merely launching companies. Foundations EDC, Inncuvate, and the VCU da Vinci Center are involved and supporting this coming summit, which tells you the university, and the regional economic-development apparatus get it and are in the room.

Learn more about the Summit

An Agenda Built for Conversations Conferences Avoid

There is a Founder Happy Hour in a local distillery to open the summit; an RVA All Day immersion in Jackson Ward’s history; an InterCultural Unity session that promises “a clear-eyed look at where things actually stand today”; Punk Rock Banking, which introduces the financial institutions doing the unglamorous community lending work; a firsthand look at Lighthouse Network’s founder-development model; Collective Problem Solving structured as peer-to-peer with no keynotes; and Roses & Thorns, an SCN staple where builders talk candidly about what keeps them up at night. This is three days with the people who build startup ecosystems for a living, in a city that has something real to show them.

The proximity to Washington is the opportunity I would press hardest. A gathering of the country’s top ecosystem builders, one hundred miles from the agencies writing SBA and SEC rules, should spend some energy on how federal policy shapes founder outcomes, because the builders in that room are the ones who watch policy hit founders first. Laurie Supinski of Start Garden in Grand Rapids, who has spent years connecting entrepreneurs to the tangled map of regional support resources, put the core builder problem in one line, “a lot of times, entrepreneurs don’t know what the resources are.”

That is a discovery failure, and discovery failures are exactly what a policy-literate ecosystem is supposed to fix. Ecosystem building requires that we fix that. Bring that to Richmond.

The Macroeconomics: What Government Is Actually For in an Ecosystem

Let me be precise about the role of government as we get back to focusing on the city of Richmond.

Government does not create value in a startup ecosystem, and it cannot fund its way to one. What it can do is build the plumbing; the fund-formation rules, the securities exemptions, the procurement access, and the M&A legal capacity that let capital recycle after exits. When a region tries to substitute public capital for those structural reforms, it fills a leaking pipe rather than fixing it. Capital that enters through a favorable program generates visible deployment and short-term momentum, but it does not compound, because the architecture that would let capital recycle and attract private participation is not in place.

Early revenue from a demanding customer is one of the strongest signals risk capital responds to, and government can be the largest demanding customer, particularly around here. Procurement rules that default to established vendors and impose compliance burdens pre-revenue companies cannot meet, effectively gate local startups out of that signal. A startup that has won even a small public contract has demonstrated something a startup with only a deck has not. Reforming procurement to allow pilots and smaller contract thresholds is capital-formation work that does not spend a dollar of public investment. For a region wired into federal, state, and municipal buyers, that is not a theoretical opportunity; it is the single highest-leverage reform available.

Notable Richmond Companies and the Anchor Effect

Richmond’s macroeconomic base is not startup-native, and that is a strength most cities would envy. CarMax, Dominion Energy, and Genworth were born and scaled here; Capital One put down major roots in the region. Those anchors matter because innovative employers are the safety net that makes founder risk rational. A founder who knows that if the startup fails there is a meaningful, sector-relevant job to return to will take risks a founder without that fallback cannot afford.

You cannot build a startup ecosystem in a sector where no anchor employer exists, because the potential founders in that sector will not take the risk. Richmond’s fintech and health-sciences density around Capital One and Virginia Commonwealth University‘s VCU Health why its startups cluster where they do.

The startup layer on top of those anchors is found through roughly $180 million allocated across dozens of deals in 2025, with capital concentrating around VCU’s health-sciences corridor: TemperPack in recyclable thermal packaging, Naborforce in eldercare, and BRAINBox Solutions, Inc. in blood-biomarker diagnostics. New Richmond Ventures’ portfolio gives us insight to SVT Robotics in warehouse orchestration and Murphy’s Naturals in consumer products show the range.

But the point is not the leaderboard, it is that the sectors track against the anchors; this is how a healthy ecosystem concentrates talent, capital, and opportunities so that entrepreneurs thrive just as they did in Jackson Ward.

The Six Parts of Startup Ecosystems, Applied to Richmond Startups

Startup Ecosystems: Understanding Why Startups Thrive and Ecosystems Fail, is organized as six arguments. Richmond is a useful test case for every one of them, because it does some of them well and stumbles on others in ways that are fixable.

Part One: The Lie of Innovation

The book’s first section argues that “innovation” has become a word that regions use to look busy without accepting consequence; activity is politically safer than value, universities oversell commercialization, and metrics reward the wrong behavior. Richmond’s exposure here is the same as everywhere else, but its history is an antidote they should embrace. Sprague and Walker were consequence, not activity; nobody handed them a metric to hit. The region’s risk is measuring itself by cohort counts and event attendance rather than by companies that reach escape velocity. The chapters on institutional fear of reallocation and the commercialization myth apply directly to how VCU’s IP and da Vinci Center work should be judged; a large commercialization office is valuable, but it is not the same thing as ideal conditions for founders, and confusing the two is the classic error.

Part Two: The Things We Refuse to Distinguish

Startups are not small businesses; risk capital is not development capital; job creation is a lagging indicator, not a goal; and talent follows opportunity, not programs. This is where Richmond’s policy environment either helps or hurts, and it is where we have to be more engaged that policy language is the most underrated lever in economic development. If Richmond’s statutes and grant language treat a scalable health-IT startup and a neighborhood retail business as the same thing, the support gets optimized for the lowest common denominator and serves neither. The regional advantage is that Activation Capital and Startup Virginia clearly understand the distinction; the risk is whether the surrounding public policy does.

Part Three: Capital Is a Signal, Not a Gift

Capital follows value; it does not create it. Fundraising is a symptom, not a strategy. The whole section of Startup Ecosystems is a corrective to the belief that the missing ingredient in any ecosystem is money. Richmond’s founders who complain about capital scarcity are, in many cases, describing a signal problem dressed up as a supply problem. The regional capital does exist; NRV, Virginia Venture Partners, and VIPC’s programs are deploying. What determines whether more arrives is the quality of what Richmond is producing, not the volume of the ask.

Part Four: Ecosystems Fail for Structural Reasons, Not Moral Ones

Soft landings and geography illusions, generic accelerators as machinery of activity, public capital misaligned by political incentives; systems behave exactly as designed. If Richmond’s outcomes are falling short for anyone there, the answer is not that founders did not want it badly enough or that the region needs more inspiration. It is that the incentives produced exactly what they were built to produce. Redesign the incentives or expect the same outputs.

Part Five: Marketing Is the Missing Discipline

Marketing is market discovery, not promotion; technologists misprice demand; narrative discipline drives capital formation. Richmond’s specific weakness, I think, is narrative. Activation Capital said it directly in its own ecosystem materials, because of the region’s population size, its metropolitan area is often left off national rankings, and the region needs to “change the conversation about how we stack up” and write its own story with actionable data. That is a marketing problem in the truest sense, and it is the discipline most people everywhere underinvest in.

Part Six: What Actually Works

Conditions before programs; capital formation as policy architecture; density, optionality, and reallocation; KPIs that actually matter; expect consequence. Richmond’s Jackson Ward history proves it already knows what density looks like. The modern task is to rebuild that density deliberately, measure the things that matter (formation rates, locally deployed capital, founder-to-investor ratio, and whether exit capital recirculates), and stop celebrating the things that do not.

Get a copy of the book

Every Startup Development Organization in the Richmond Region

So, who’s doing the good work? The support layer in Central Virginia is denser than its national reputation suggests (which is why there is that marketing and narrative challenge). Understanding who does what is the first step for any founder or funder trying to navigate it.

• Activation Capital is the connective tissue. An independent authority of the Commonwealth (formally the Virginia Biotechnology Research Partnership Authority), it operates the 34-acre Virginia Bio+Tech Park and functions as the region’s ecosystem-development organization, providing social, knowledge, and financial capital to clusters and support organizations. Its Ecosystem Direct Investment Fund deployed $1.261 million to organizations supporting founders, including Lighthouse Labs, Startup Virginia, and SCORE Richmond.

• Lighthouse Network (formerly Lighthouse Labs) is Richmond’s nationally recognized accelerator, rebranded in 2025 to emphasize long-term founder development over cohort acceleration. It partnered with the Health Innovation Consortium to launch Virginia’s only health-focused accelerator.

• Startup Virginia is a nonprofit high-growth incubator based at the Michael Wassmer Innovation Center at “1717” in Shockoe Bottom, supporting founders from idea stage through growth.

• The Health Innovation Consortium, founded by VCU, VCU Health, and Activation Capital in 2019, brings health innovations to market and co-runs the health accelerator with Lighthouse.

• The Dominion Energy Innovation Center supports energy and advanced-technology startups.

• The VCU da Vinci Center anchors university-based venture creation and is a Richmond Summit sponsor.

• SCORE Richmond provides mentorship and programming, funded in part by Activation Capital.

• The Jackson Ward Collective supports Black-owned businesses in the region and was a partner in Activation Capital’s diverse-founder pilot with Opportunity Hub.

• Regionally connected programs extend the reach: 757 Accelerate and 757 Collab in Norfolk, Catalyst in Charlottesville, and RAMP/Verge in Roanoke-Blacksburg all operate within Richmond’s statewide orbit.

As always, these assessments are based on what our research uncovers so if we’re missing something, that should be a signal to the ecosystem that what’s missing isn’t well enough known. Certainly, share those gaps in the comments but more importantly, everyone, note that what it means is that the website, social media, content, or mainstream media, are failing to promote what matters.

The Regional Capital Stack

Richmond’s funding sources run deeper than the “small but growing” cliché admits, though they concentrate in specific sectors.

• NRV (New Richmond Ventures), founded in 2011 by Jim Ukrop, Bob Mooney, and Ted Chandler, is the region’s flagship venture firm. It has made roughly 27 investments with 7 exits, invests primarily at Series A across healthcare, food and beverage, and e-commerce, and raised a $33 million Early-Stage Growth Fund from 83 investors across the state. Managing Director Laura Markley described one portfolio company as a “unique solution… created by a founder team with decades of experience as owners and operators in the industry,” which is the operator-led thesis that makes local venture credible.

• Virginia Venture Partners, the seed and early-stage investment arm operated through the Virginia Innovation Partnership Corporation and based in Richmond, has made hundreds of investments across Virginia technology, life-science, and cleantech companies, with Launch grants and convertible-note programs for early-stage founders.

• VIPC’s Regional Innovation Fund provides competitive grants to entrepreneur-support organizations; Startup Virginia received $100,000 to strengthen Richmond’s ecosystem.

• Virginia Invests / Virginia Catalyst and the broader VIPC apparatus are designed to activate $250 million in private investment and engage 1,000 Virginia entrepreneurs.

• Angel and club-style capital historically ran through NRV’s original “club investing” model, continues through regional angel activity; the Activation Capital EDIF functions as ecosystem-level angel capital deployed into the organizations rather than directly into companies.

An ecosystem works when founders know who their investors are without needing referrals.

What Richmond Does Well, and Where It Has to Improve

Richmond does the anchor-and-density thing well. It has genuine corporate anchors in fintech and energy, a research university generating health-sciences IP, a connective ecosystem-development authority in Activation Capital, and a historical proof point in Jackson Ward that density and capital recycling are in the region’s DNA.

Its support organizations understand the startup-versus-small-business distinction that almost all regions botch. And its geography, wired into Washington, Norfolk, and Charlottesville, gives it a policy and customer-access advantage almost no comparable city has.

Where it has to improve is narrative and structure.

On narrative, Richmond undersells itself so consistently that it falls off national rankings its actual output would justify. That is a market-discovery failure, and the region’s builders need to treat storytelling as a discipline, not an afterthought. On structure, the open question is whether public policy in the Commonwealth matches the sophistication of its support organizations, particularly on procurement access, fund-formation friction, and whether statute distinguishes scalable startups from small businesses. The capital exists; the anchors exist; the talent exists. What determines Richmond’s next decade is whether it builds the conditions that let all three compound, or whether it settles for the activity that photographs well.

Sitting a hundred miles from the people who write the rules, Richmond has less excuse than most to get the structure wrong, which means it has more potential to fix a lot for entrepreneurs everywhere.

The builders convening in Richmond this fall could use your voice. If you are championing startups, an investor deciding where to deploy, or an economic developer working out how to really be meaningful to entrepreneurs, the exercise for you is not to admire Richmond but to look at this as a case study; ask which of the six failures your own region is committing right now, and whether the people around you would rather fix the structure or cut another ribbon.

CNBC published its annual list of America’s worst state economies this month, and the reaction split predictably along tribal lines, suggesting one thing clearly; this is probably an assessment worth scrutiny. Why bother? What we might have here is a lesson in media bias or critical thinking, perhaps it’s a great way to share how to really assess an ecosystem, or we’ve been given an opportunity to spark a meaningful discussion of what’s really going on in these states and the economy. What we start with is asking, “Does the methodology actually measure whether an economy is good or bad?” It does not. What CNBC built is a machine that rewards states for being big and old, penalizes them for being smaller or young, and then launders the result through enough sub-metrics to make it look like econometrics.

It is the ranking equivalent of judging a marathon by who weighs the most. Oklahoma, North Dakota, New Hampshire, Alaska, South Dakota, Kansas, Louisiana, West Virginia, Maryland, and Rhode Island, you’re put on notice: CNBC thinks you’re a mess. We think they’re wrong.

Start with what CNBC says it measured. The Economy category carries 16.6% of a state’s total score in the 2026 study, and to measure each state’s economy CNBC considers job growth, economic growth, and the number of major companies headquartered in the state, along with fiscal health, budget situation, long-term obligations, debt ratings, and the health of the residential real estate market. Read that list again slowly, because three of those inputs are functions of how large and established a state already is, not how well its economy is performing for the people and entrepreneurs in it. “Number of major companies headquartered in the state” is a legacy-asset count certainly driven by Big City = More Headquarters. Rhode Island did not become CNBC’s worst economy in 2026 because a founder in Providence struggles to raise a seed round; it landed there because it posted weak growth and generated relatively few new businesses, and because it has roughly the corporate-headquarters footprint you’d expect from the smallest state in the union.

You are being told that small equals bad.

The GDP Fallacy Has a Nobel Prize Attached to Its Refutation

There is something that economists settled more than fifteen years ago, which apparently never reached the CNBC newsroom (or even the White House for that matter). Using aggregate economic size as a proxy for economic quality is a known, named, catalogued error. It’s idiotic frankly and I’m happy to say so harshly in hopes that Washington will stop claiming wins based on GDP.

In 2008 the French government commissioned Joseph Stiglitz, Amartya Sen, and Jean-Paul Fitoussi, two of them Nobel laureates, to examine whether GDP is a reliable indicator of economic and social progress. Their conclusion, published in 2009 and later expanded in the book Mismeasuring Our Lives, was that it is not. Stiglitz put it bluntly in Scientific American, arguing that striving to grow GDP is not the same as ensuring the well-being of a society, and that the number does not capture health, opportunity, or the quality of life. Organisation for Economic Co-operation and Development (OECD) follow-on work is even more to the point for our purposes; the High Level Expert Group showed how over-reliance on GDP misled policy makers who did not see the 2008 crisis coming.

A state with a $1.8 trillion economy and a state with a $48 billion economy are not competing on the same field. Averaging their headquarters counts into a competitiveness score tells you which one is bigger, not which one is better at turning opportunities into new jobs.

Now, CNBC will protest that it did not use raw GDP, and technically that’s true.

But “economic growth,” “number of major companies headquartered,” and “residential real estate market health” are all size-and-maturity correlates wearing a growth costume. As one recent GDP-rankings analysis put it, total GDP mostly measures size; Vermont, Wyoming, and Alaska posted the smallest total GDPs in 2025, yet total output is not a living standard. Think of it like ranking restaurants by the square footage of the building; The Cheesecake Factory wins; the Michelin-star barbecue place I have 2 blocks down the road from me, with room with twelve seats, would be in last place. We have measured something real, but we have measured here the wrong thing.

The Nameplate Fallacy: Counting Headquarters Rewards Inheritance

, who spends his days supporting scale-ready startups, put it this way,

“A state can be rich in nameplates and in decline, or nameplate-poor and rapidly building durable new-firm density. A headquarters count cannot tell those two states apart, because it measures inheritance, not trajectory.”

He’s right, and the mechanics in doing this make it worse.

A headquarters count is a stock accumulated over decades, while job growth and GDP growth are one-year flows. Blending a decades-long legacy stock with twelve-month flows in a single composite scores a state on what it inherited alongside what it did last year, then reports the average as if it were one coherent thing.

In his Substack, Brian argues that tethering a region to large established firms builds fragility, because the region’s fate rides on a distant board’s next optimization decision. Maryland is the live case in CNBC’s own 2026 results, dinged for flat growth that traces to over-reliance on government jobs and federal funds. That’s a distinct slice of the state economy that a headquarters count can’t see, and the ranking has no recognition that government focus or federal funding for the economy as such, elevating Maryland from a “worst” because of stable jobs that have nothing to do with headquarters.

Volume Is Not Quality, and the Research Says So

CNBC does include job growth, which is good, and then undercuts it by weighting headquarters counts alongside it, which is self-defeating.

If you actually care about job creation, the research has consistently proven job growth comes from new business creation, not big-employer counts. Kauffman Foundation economist Tim Kane’s 2010 study of Business Dynamics Statistics covering 1977 through 2005 found that firms in their first year add an average of 3 million jobs annually, while every other age category, from firms in their first full year to firms founded two centuries ago, collectively shed a net million jobs a year. Kane is careful to note that age-zero firms enjoy a definitional advantage, since a firm that didn’t exist last year cannot have destroyed a job; the pattern holds anyway, in all but seven years on record. His conclusion should be tattooed on the wall of every economic development office in America, that states and cities with job creation policies aimed at luring larger, older employers cannot help but fail, not just because they are zero-sum, but because they are not built on realistic models of employment growth.

So a ranking that scores states partly on how many large companies are already headquartered there is rewarding the asset that, on average, destroys net jobs.

Brian cautioned in a way that I vehemently and frequently argue too, that while job growth comes from new ventures, we must stop treating new businesses and startups as the same, “The instinct to count new firms is not wrong in principle; the problem is resolution. Roughly half of net job creation comes from the 1 to 5% of new firms that are high-growth gazelles, and half from the 95%-plus Main Street firms. A raw formation count treats every new firm as an identical unit, whether it is built to scale or built to stay small.”

The economics back us. Jorge Guzman and Scott Stern, in The State of American Entrepreneurship (American Economic Journal: Economic Policy, 2020), argue that assessing entrepreneurship requires not simply counting quantity but estimating the initial quality of new ventures, and they build a quality-adjusted measure precisely because raw counts mislead. The same distortion shows up on CNBC’s “best states” side, where South Carolina’s headline growth was substantially a function of worker in-migration; the economy expanded by adding bodies rather than by raising output per person. A count cannot separate growth-by-headcount from growth-by-capability.

The Score Grades the Marketing Pitch

Let me hit on a point personally meaningful because I think most will miss this; this should end any debate about our criticism of CNBC. CNBC sets the Economy category’s 16.6% weight by how frequently states mention the economy in their own economic-development marketing pitches. The instrument’s weights are derived from the subjects’ self-promotion. It measures the salience of state PR and then grades states against a weight their own PR produced. It is the macro cousin of the self-commissioned impact study, a number the beneficiary helped set, about itself.

Worse, it’s how cities are saying they’re the next Silicon Valley, Innovative, or focused on entrepreneurs, then concluding it’s true because they say so.

Anyone who worked in search two decades ago recognizes this instantly.

In the early days of SEO, engines ranked pages partly on self-declared signals, the meta-keywords tag and raw keyword density. So operators stuffed keywords into the page and hid them by setting text color to match the background, white words on a white field, invisible to a human and fully legible to the crawler. The page was optimizing the signal it declared about itself rather than the target, actual relevance. CNBC is doing the state-ranking version of exactly this; it reads what each state says about itself, counts how often the economy comes up, and lets that self-declared salience set the weight. The states are writing the meta-keywords tag, and the ranking is trusting it.

Google publicly stopped trusting the meta-keywords tag in 2009. CNBC’s methodology is more than a little out-dated.

What killed keyword stuffing is that algorithms replaced self-reported density with signal a site could not declare about itself, inbound links from other sites, an external judgment the subject did not control (if you’re an SEO, bear with me, I know that’s not exactly right, it’s the point that counts). That is the move an honest state-economy score needs, from signals the state announces about itself to outcomes measured externally that it cannot game.

Where CNBC Is Backward-Looking, and Where North Dakota Gets Punished for Saving

Grading the economy partly on last year’s growth measures the past; it says nothing about durability or trajectory. Brian develops the mechanics of this in his companion piece.

North Dakota posted the lowest growth in the nation last year, so it lands near the bottom. Yet CNBC itself notes it built reserves in the flush years and could run nearly a year on its fund balance if revenue vanished. North Dakota’s mistake, apparently, was saving money. Can you see why I want to wring someone’s neck?

In Fairness, What CNBC Gets Right

We are not arguing CNBC included nothing good. Credit where it’s due, and there are three real things:

• Small-business survival rate is included at all, which is a genuine outcome signal and rarer in these rankings than it should be.

• Fiscal health, pension underfunding, and debt ratings are legitimate durability indicators.

• Tariff exposure and federal-budget-cut exposure gesture at concentration risk, even if no explicit fragility index results from them.

The problem is not the absence of good signals; it is that the outcome-flavored signals are underweighted and buried while the proxy-and-stock metrics carry the headline. This is what to watch for in considering if there is an influence of press releases, advertiser interests, or political interests, crafting the narrative more than the meaningful research properly applied.

Which makes the first ask nearly free; reweight this! CNBC already gathers survival rates and fiscal-fragility data. Promote those from the footnotes to the headline, demote the nameplate count, and the score sharpens without a single new data source.

What an Outcome-Based Score Would Actually Measure

For anyone willing to build that, which is exactly what I’m demanding startup economies do in the Startup Ecosystems, measure based on outcomes, not activities and counts that aren’t applied per capita. Brian put together a constructive alternative that follows one governing rule; every input must be a signal the state cannot declare about itself:

• High-growth (gazelle) firm share, not raw formation count, quality-adjusted along the lines of the Guzman and Stern model.

• New-firm survival at three and five years, measured from formation, as a durability signal rather than a birth signal.

• Per-capita productivity and wealth creation, decomposed from aggregate GDP so headcount-driven growth is separated from capability-driven growth.

• A concentration and fragility term, the share of jobs and output tied to a single dominant employer or to federal funding, the term that would have flagged Maryland’s exposure before it surfaced as a growth number.

Brian and I don’t fully agree on how to weight the first two, and we’d rather say so than pretend to have a consensus we don’t have. I aggressively favor the fact that startups are the cause of almost all net job creation because the small set of high-growth young firms also is a reflection of a culture of entrepreneurs, who in turn spur more. Brian gives more weight to the broad new-business base because of the far more likely survival rate of new businesses that aren’t classically defined startups. Each of us is arguing slightly for and against our own opinions, but that’s what economists do; when the data is meaningfully relevant, we can argue over nuance, we just want CNBC to set, roughly, the same standard.

Why This Matters Beyond Being a Case Study in Media Misguidance

A ranking this visible is not neutral commentary; it is an instrument, and instruments set incentives.

The influence isn’t hypothetical; when the Quality of Life rankings published, Florida’s governor dismissed them as “nonsense” on X, arguing that people wouldn’t be moving to Tennessee in droves if it were really the worst state to live in. That’s a headcount argument, the same fallacy the ranking itself runs on, which Brian takes apart with South Carolina in his companion piece. That is itself the tell; a governor who stops to swat at a ranking is a governor who knows it carries weight. When an economy score rewards headquarters counts and one-year formation numbers, it tells fifty economic-development offices to chase nameplates and juice formation figures, which is to optimize the proxy. It never tells them to build the durable gazelle density that actually creates wealth, because the instrument cannot see it. Stiglitz’s warning is that what we measure affects what we do, and if we measure the wrong thing, we will do the wrong things.

If you run an economic development organization, or you’re a founder deciding where to build, or an investor deciding where to deploy, build your own scorecard from the inputs that predict outcomes; rate of new business formation, early-stage survival, gazelle share, and the regulatory friction a founder hits between idea and incorporation. Those tell you whether an economy is alive.

Headquarters counts and debt ratings tell you how old it is.

The Variable CNBC Left Out, and Why That Puts the Wrong States at the Bottom

Search CNBC’s Economy methodology for a measure of regulatory burden and you will not find one. Fiscal health is in there, debt ratings are in there, residential real estate is in there, and the compliance friction a founder hits between idea and incorporation is nowhere. CNBC keeps a separate Business Friendliness category that touches legal and regulatory frameworks, which means the outlet knows the variable exists but I fear considers it too minimal a factor in reporting a state’s economy.

Which is how South Dakota lands on a worst-economies list while also ranking near the top of the best states for startups on tax structure and being among the least regulated states. South Dakota has almost no startup ecosystem yet but also almost no friction, which makes it a blank canvas rather than a miss, which should rank it meaningfully higher. Oklahoma is the same error with more evidence against it. CNBC cites moderate growth, a stressed housing market, and federal-cut exposure, all flow-and-exposure observations about last year, while Oklahoma is starting to crush on the things that predict next year. How the hell is it on this list unless the list is just wrong?

Run Oklahoma and South Dakota on gazelle share and three-year new-firm survival instead of nameplate count and one-year flow, and at least one lands meaningfully higher than its 2026 grade. We’ll grade that publicly against next year’s data, whichever way it breaks.

We audited a cable network’s spreadsheet because the spreadsheet moves money; a state chamber cites it in testimony, an economic development office builds next year’s strategy around climbing three spots, and a site selector puts the slide in front of a client. Reach is doing the work that rigor should be doing, which is why ecosystem builders and economists should be guiding policy and investment decisions rather than outlets whose incentive structure rewards an annual list that fifty states will either brag about or angrily rebut. Build your own scorecard from what predicts outcomes; formation rate, survival at three and five years, gazelle share, concentration risk, and the regulatory burden a founder actually carries. When a metric cannot tell the difference between an economy failing its founders and one that is simply small and unencumbered, what have you been using it to decide?

While looking into entrepreneurship education, the most-cited piece of evidence that keeps coming up is one that educators might not like to hear in as much as it requires we think differently than what we’re doing now. George Land and Beth Jarman, administered a divergent-thinking assessment, originally built for NASA to identify creative engineers, to a cohort of 1,600 young children in 1968 and then re-tested the same children as they aged. The results, published in their book Breakpoint and Beyond, are why we’re here:

98% of children aged 4–5 scored at “genius level” in divergent thinking; by age 10 that fell to roughly 30%; by age 15 to about 12%; and among 280,000 adults tested, only 2% scored at that level.

Read that again please because I want you to be alarmed, concerned, and consider that we’re doing it wrong.

Divergent thinking is the ability to look at one problem and generate many possible answers rather than hunting for the single correct one; this is the cognitive raw material of entrepreneurship, and it collapses with years of schooling.

Land wrote, highlighted in a great piece by The Open School:

“What we have concluded is that non-creative behavior is learned.”

This is a study I’d like us to be intellectually honest about because you should never repeat a statistic you can’t defend, and if you cite Land, someone will eventually challenge you. I’m citing Land, and reviewing this work, because schools are increasingly asking me about entrepreneurship education and they struggle to understand what I mean when I reply from my experience, “you can’t teach entrepreneurship,” when seemingly every university and public school curriculum now has an entrepreneurship class taught by the Professor of Entrepreneurship.

An appeal to common practice doesn’t validate that it works. The existence of astrology departments wouldn’t prove astrology predicts the future.

And I say that because this research in particular requires some critical thinking and our own observation. The finding lives in a trade book rather than a peer-reviewed journal, and skeptics have gone looking for the original NASA dataset without finding it. So, let’s cite it for what it is; a widely-referenced observation, not a controlled published experiment. The directional claim, which I argue we can agree with, that the trait declines as schooling accumulates, does not rest on Land alone; Sir Ken Robinson built the most-watched TED talk in history on the same argument:

“We are educating people out of their creative capacities.”

This Was the Design, Not an Accident

The model of mass education we still run was built at the turn of the twentieth century for a specific purpose, and that purpose was not producing founders. It was producing employees; punctual, rule-following, standardized, interchangeable employees who could staff an industrial economy. We teach children to:

• Pass the test rather than question whether the test measures anything worth knowing

• Follow the instructions rather than notice the instructions are wrong

• Absorb the assigned material and reproduce it on demand

• Then get the job, build the career, and be stable and productive

Every one of those is a virtue in an employee. Not one of them describes an entrepreneur.

This design bias runs deeper than the classroom, found in healthcare policy I recently explored while trying to unpack why education was designed to create employees.

American employer-sponsored health insurance, the thing that keeps millions tethered to jobs they would otherwise leave, was not designed around patients. It was designed around employers. During World War II, the Stabilization Act of 1942 froze wages; the Act explicitly excluded from that freeze “insurance and pension benefits in a reasonable amount to be determined by the President.” Employers, forbidden from competing for scarce workers on salary, competed on benefits instead. The USC Schaeffer Institute documented that companies “began to offer non-wage benefits, including health insurance,” and coverage exploded, from 12 million Americans in 1940 to over 70 million a decade later. An entire pillar of American life was retrofitted into an incentive for employment.

The point is not the healthcare; the point is that our foundational systems, education included, were engineered to manufacture and retain productive employees, and they are extremely good at it.

The Dissonance Is the Tell

So we run people through a machine optimized to produce stable employees, and then a strange thing happens; they feel a tug. The machine worked, they got the degree and the job and the benefits, and something still itches. That itch is the gap between what they were trained to be and something they were never allowed to develop, and the startup boom of the last fifteen years gave the itch a name.

Suddenly, everyone wants to be an entrepreneur or wants their kid to be one.

Let me define the term the way my own work defines it, because the programs selling it rarely bother.

An entrepreneur is not a person who starts a company. An entrepreneur is a person who sees a problem and organizes resources to fix it under conditions of genuine uncertainty.

Startups are one vehicle for that; they happen to be the vehicle that generates most net new jobs, which is why they hog the spotlight. Research from the Kauffman Foundation has long shown that new and young firms, not small businesses generally and not big incumbents, account for the lion’s share of net job creation in the United States. But the entrepreneurial act is the fixing, not the incorporating. The confusion between “entrepreneur” and “startup founder” is the confusion that makes school programs fail, because they teach the vehicle and skip the driving.

A Textbook Product-Market Fit Failure

Here is what schools have actually done, described in the language of the thing they claim to teach:

1. They identified demand (“people want entrepreneurship”)

2. They built a product to match the stated demand

3. They never ran customer discovery to test whether the product produces the outcome the customer is actually buying

Parents and teenagers say they want entrepreneurship; schools happily supply a thing labeled entrepreneurship; everyone pays; and the value the customer believes they are purchasing, the capacity to actually build or fix something, is almost never in the box. It is product-market fit theater. The label matches the demand while the contents match neither.

Why is the box empty? Because what gets taught is age-inappropriate, person-inappropriate, and outcome-inappropriate, usually all three at once. Walk into a typical program and you will find two things masquerading as entrepreneurship:

• Business. The business plan, the five-year projection, the market-sizing exercise. Business is real and teachable, and we already teach it, correctly, in business classes, marketing, finance, and management. It is not entrepreneurship.

• The Silicon Valley scalable-startup framework. Usually some flavor of Lean Startup, taught as if running a build-measure-learn loop turns a sixteen-year-old into a founder. Lean Startup is genuinely useful for a specific kind of venture; the accelerator-and-scalable-startup model it belongs to is relevant to a vanishingly small fraction of the businesses any economy actually produces.

Teaching that model as the definition of entrepreneurship is like teaching Formula One pit strategy as the definition of driving.

“Startups are not small businesses. They are temporary organizations searching for scalable, repeatable, business models under conditions of genuine uncertainty.” – via Startup Ecosystems

Most students being sold “entrepreneurship” will never run one of those, and pretending otherwise wastes their time and their tuition.

Still, We Have to Teach It, Because the Alternative Is Worse

None of this is an argument to stop. It is an argument to stop doing it stupidly. Entrepreneurs fix things, startups generate the majority of net new jobs, and an economy that wants to create wealth and let people thrive in a future nobody can predict cannot afford a population that does not understand entrepreneurship. The catch, and I am being generous rather than blunt, is that the professionals who most need to understand it do not. The lawyers, the politicians, the researchers, the healthcare workers, the finance people, the software developers, and the investors who surround and shape entrepreneurship largely do not understand what they are shaping. That is the actual crisis, and it points directly at the fix.

The best thing we could do for the economy is to teach entrepreneurship seriously while being ruthlessly clear that teaching entrepreneurship does not mean teaching people to start startups. Entrepreneurship is a field of study, a specialization that attaches to other disciplines, not a factory for founders. Once you accept that, the structure practically designs itself.

First, Separate the Founders from the Field

Actually enabling founders is not a four-year degree and not a two-year minor; it is a bootcamp or an incubator. An intensive, hands-on, do-the-thing environment. Trying to credential someone into founding is a category error; you enable founding by putting people in conditions where they build, fail, and build again. Keep that separate, route your actual aspiring founders there, and stop pretending a diploma produces a founder.

Everything else, the actual field of study, sorts into three two-year tracks. And before anyone misreads it; none of this is teaching business. Business stays in business, small business, marketing, finance, and management programs where it belongs.

Three Two-Year Entrepreneur Studies

1. Understanding it (work on it). This lives inside political science, business, finance, and law, and teaches those professionals what they need to know about entrepreneurship as a phenomenon:

• The history and the psychology of it

• The difference between startups and intrapreneurship

• Why deep tech is a fundamentally different animal from a scalable software startup

• Why angel investing runs on different logic than venture capital, and why both differ from lending to a main-street business

This track produces the entrepreneur specialist. It is the lawyer who works in entrepreneurship, the city Head of Innovation who actually studied the thing they are appointed to grow, the banker or investor who can tell a restaurant loan from a seed round because they were taught the difference rather than left to guess. This is history, economics, and context.

2. Applying it (work in it). A two-year program, deliverable in K-12 at age-appropriate depth or as a college minor, built on the recognition that most people will never start a venture, but many will work inside one, where the rules are genuinely different. It overlaps with track one; you still need finance and legal minds, but the output inverts:

• Not the law firm that specializes in startups, but the legal mind who works for one

• Not the investor who understands startups, but the same finance graduate serving as VP of Finance inside one

• The salesperson, marketer, and software developer who learn to operate in a startup rather than a corporation

Let me say bluntly this observation that most keep quiet; if you are a marketer or a software developer selling yourself as someone who can do so for everyone, we both know you cannot, because working for a startup is a different world from working for an established business. The economy needs the marketing graduate with a minor in entrepreneurship, and the software engineer who is not merely a computer-science degree, but an engineer specialized for entrepreneurial environments. This is methodologies, due diligence, marketing, and psychology.

3. Studying it (work for it, on behalf of entrepreneurship). This is where the practitioners who support entrepreneurship finally get the education they have always deserved and rarely received. It belongs in:

• Healthcare, so therapists and physicians actually understand these people and the mental-health burden they carry

• Political science again, but here producing the policymaker who comprehends the downstream consequences of their regulations, incentives, and allocations of public capital

• Research, where university researchers and the economists studying this sector need to grasp how profoundly it differs from the rest of the economy

• Teaching, where the people who will impart all of this to the next cohort are trained

This is public policy, economics, history, mental health, and psychology.

How to KISS and explain this to a dean or an EDO director:

• Understand it: work on it

• Apply it: work in it

• Study it: work for it, on behalf of entrepreneurship

That is how you teach entrepreneurship. Not as how to do business, which we already teach; not as how to launch a startup, which is a bootcamp, not a curriculum. You teach it as a field, so that we produce people who support entrepreneurship, people who can work inside startups, and professionals across law, finance, government, healthcare, and research who genuinely understand the sector they keep claiming to serve.

What To Actually Do with Entrepreneurship Education

If you run a school, an economic development office, or a policy shop, the first move is diagnostic, not constructive. Before you build anything:

1. Trace a few cohorts of your existing “entrepreneurship” program. Ask whether graduates ended up more capable of fixing problems under uncertainty, or merely more fluent in business-plan vocabulary. If it is the latter, you are running a business class in a costume created to sell the seats.

2. Decide which of the three tracks you can actually staff. Most institutions can credibly teach exactly one. Pick the one that matches your faculty and your region’s real economy.

3. Route the founders themselves to an incubator, not a syllabus. Stop asking a four-year degree to do a bootcamp’s job. Better? Funnel everyone through one of the three tracks of education and make students apply to a separate “startup” program just as we require in the private sector when founders pursue an accelerator.

4. Change the metric you judge yourself by. A good program produces very few founders; judge it instead on whether it produced lawyers, financiers, civil servants, doctors, researchers, and operators who understand entrepreneurship well enough to stop actively getting in its way.

That last one costs institutions their favorite marketing line, which is exactly why it’s one of the right tracks. “We make entrepreneurs” is easy to put on a brochure and impossible to deliver. Literally impossible. You might disagree with me, and by all means, flood the comments with how wrong I am; but then take a moment and ask yourself, are we teaching how to do startups or how to start a business? That’s not entrepreneurship, call it what it is. “We produce professionals who finally understand entrepreneurship” or “we teach what it means to work with entrepreneurs” is harder to sell and is the only version that would actually help the economy.

Schools, I beg you to trust when I point out that if you actually teach entrepreneurship such as the track of applying it, so people KNOW what it means to work that way, in those environments, and then have a professionally run startup incubator available to the students from within that who qualify because of good ideas and early traction, you will better serve students and their futures, while accelerating the economy for everyone.

Question this of anyone selling or buying a school entrepreneurship program; if your program produced zero founders but every single graduate finally understood how entrepreneurship actually works, did it succeed or did it fail?

Your answer tells you whether you were ever teaching entrepreneurship or just teaching business and hoping nobody checked the box. If the program is a startup incubator or startup accelerator, you’re ignoring the psychology, history, government and civics lessons, and economics study, that is a pre-requisite of actually understanding entrepreneurship. If the question itself makes you uncomfortable about the program you’re currently running or funding, that discomfort is the most useful thing I can give you; it’s worth a conversation before the next budget cycle locks the old model in for another year.

Explored in the Book: Startup Ecosystems

Most of what we call “hiring” is glorified pattern-matching against a résumé or familiarity thanks to conversations. We look at where someone went to school, how long they stayed at their last three jobs, and whether a reference will say something nice on a phone call they’ve been coached to expect. Then we act surprised when the person we hired to move fast turns out to be someone who waits to be told what to do. The signals we trust most are the ones that predict the least and nowhere is that gap more expensive than in startup team building, where a single wrong hire in a five-person team isn’t a personnel problem; it is likely to kill the venture.

It took me more time than it should have, to see this. With most of my work being in the economy of startups or through the lens of marketing, exploring team dynamics is something I do only occasionally, despite the team being the leading cause of success or failure of a startup. Surprising myself even more, I was recently reminded that when I moved to Austin, Texas, many years ago, I scratched my head when people who seemed to be working in startups, would say of my resume, “wow, you’ve been in a lot of different roles,” thinking, yeah, because I work with startups. Why would so many people think working in startups means a 5-year tenure of employment when the stages through which startups move last 3-18 months??

Then it clicked… the confusion of startups and businesses permeates into teams and employment, with people thinking that skill and experience in business is as meaningful in a startup as it is an established company. What too many are ignoring is that in startups, “execution” (get it done) is cited as a most important quality, while the needs of a venture at an idea stage vs. growth stage, are completely different skills and experiences. We’re not measuring agency in building startup teams; most failed teams are because we’re building them exactly as we would for a traditional role at a 1000 person company.

There’s a simple illustration making the rounds that captures the thing traditional hiring keeps missing. I’m recreating it here because it’s the simplest visual I can think of that separates a person who makes an organization faster from a person who makes it slower.

The framework says there are five levels at which any team member can show up to their daily work.

• Level 1: “There is a problem.” Then they walk away and leave it for someone else.

• Level 2: “There is a problem, and I’ve found some causes.”

• Level 3: “Here’s the problem, here are some possible causes, and here are some possible solutions.”

• Level 4: “Here’s the problem, here’s what I think caused it, here are some possible solutions, and here’s the one I think we should pick.”

• Level 5: “I identified a problem, figured out what caused it, researched how to fix it, and I fixed it. Just wanted to keep you in the loop.”

Read those again and notice what’s changing. It isn’t intelligence, and it isn’t effort.

Everyone from Level 1 to Level 5 noticed the same problem.

What’s changing is how much of the problem the person is willing to own before it lands on someone else’s desk. Level 1 hands you a burden. Level 5 hands you a resolved issue and a courtesy update. The entire ladder is a measure of ownership, and ownership is the thing you cannot see on a résumé.

Agency is Not a Personality Quirk; it’s a management concept with a paper trail

Back in 1974, William Oncken and Donald Wass published “Management Time: Who’s Got the Monkey?” in the Harvard Business Review, one of the most reprinted articles in the magazine’s history, and they defined what they called “levels of initiative.” The guidance that followed is where I find myself writing today: an employee can (1) wait to be told what to do, (2) ask what to do, (3) recommend an action then implement it with approval, (4) take independent action and advise you at once, or (5) take independent action and update you at an agreed time.

Oncken’s instruction to managers was blunt; when someone brings you a problem, “outlaw use of level 1 or 2.”

Fifty years, it would seem this should be conventional wisdom in our team building and hiring. That, a problem doesn’t belong to the person with the authority to solve it; it belongs to the person who noticed it.

A Level 1 employee is functionally handing you their monkey and walking off whistling. A Level 5 employee shot the monkey and buried it in the yard before you finished your coffee; a terrible analogy, I know, but I’m going off the HBR title and needed to refer to the monkey somehow.

Ed Catmull built Pixar on the same principle. In Creativity, Inc., he writes that if planning cannot prevent problems, “then our best method of response is to enable employees at every level to own the problems and have the confidence to fix them.” That’s Level 5 stated as a cultural design goal by a man who ran one of the most creatively productive companies in modern history.

Agency isn’t a soft skill; it’s the operating system of organizations that outrun their competitors.

In a Startup, Level 4 is the Floor

Here’s where my corporate framing and the startup distinction really matter. In a big company, you can carry Level 2 and Level 3 people for years. There’s slack in the system; there’s a manager whose entire job is to convert Level 2 reports into Level 4 decisions. A large organization is, in a sense, a machine for compensating for low agency. It has the headcount, the process, and the runway to absorb people who need to be told what to do.

A startup has none of that.

When there are six of you and eighteen months of cash, a Level 2 employee isn’t a developmental opportunity; they’re a liability you’re funding out of your own dwindling account. Everyone in a startup has to operate at Level 4 minimum, and ideally Level 5, because there is nobody standing behind them to convert observations into decisions. Or, rather, there shouldn’t be anyone standing behind them; as founders, your job is vision, cash flow, and demand gen, not management. If your operations lead brings you “there is a problem with fulfillment” and stops there, you don’t have an operations lead; you have an expensive smoke detector.

The team is the whole ballgame; postmortems on failed startups for years, and as I’ve written in why pivoting fails, point out that the failure of the team is consistently cited as the primary cause of collapse. The Kauffman Foundation has repeatedly confirmed the outsized role team dynamics play in whether a venture lives or dies. Capital problems and cash-flow problems are symptoms; the disease is almost always a team that couldn’t execute, couldn’t align, or couldn’t own its own problems. A team stacked with Level 2 and Level 3 people will burn every dollar you raise and blame everyone else on the way out.

Being high agency isn’t just about tackling problems in this specific way, it means your entire way of working is oriented toward Level 4 and Level 5 behavior as the default expectation, from day one, for everyone.

The signals we hire on are the signals that predict the least

So, if agency is the thing that determines startup outcomes, why do we still hire on education, tenure, skills, and references? Because those things are easy to measure and comforting to point at when a hire goes wrong. They are also, empirically, close to useless as predictors.

The single most cited body of research in personnel selection might be Schmidt and Hunter’s 1998 meta-analysis in Psychological Bulletin, which aggregated 85 years of hiring research.

Buried in that work, and expanded in McDaniel, Schmidt and Hunter’s analysis of job experience, is a finding that should end most hiring conversations before they start; years of experience barely moves the needle on predicting how well someone actually does the job, and whatever small edge it gives flattens out after about five years. Years of education predicts even less. Put concretely, if you sorted a stack of candidates purely by who had the most experience or the fanciest degree, you’d do only marginally better than sorting them at random; the two credentials we lean on hardest tell you almost nothing about how the person will actually perform. Filtering primarily on experience, as that research notes, “eliminates high-ability candidates who would outperform experienced but less capable applicants.”

Sit with that for a second.

The résumé filter, the thing every applicant tracking system is built to automate, the entire reason Indeed is stuffed with keyword optimizing job descriptions getting responses from AI generated resumes, is optimizing for the variables with the weakest link to performance. We’ve built an entire multi-billion-dollar recruiting industry around measuring the wrong thing accurately, and then we congratulate ourselves on the efficiency.

References are worse, because they’re not just weak; they’re actively distorted. A reference reflects how someone performed inside one specific environment, under one specific manager, with one specific set of constraints, filtered through a relationship where nobody wants to torpedo a former colleague’s livelihood. It tells you almost nothing about how that person will behave when the environment changes entirely, which in a startup it always does. Past performance in a structured corporate role is one of the most misleading signals you can carry into a hiring decision for a high-risk, unstructured startup role, because the two contexts reward opposite behaviors. The corporate high performer is often someone who mastered the process; the startup you’re building doesn’t have one yet.

High Agency and Self-direction Beat Credentials, especially at the Frontier

The research on what actually predicts entrepreneurial outcomes keeps pointing away from credentials and toward disposition. The study of distinct founder personality types tied to success, with the collective personality of the entire founding team significantly shaping the outcome, reinforces we need to be seeking agency, not skill. It isn’t the degree; it’s the wiring. In predicting startup success with personality data, a growing body of research confirms that personality traits, psychological wiring, and founding-team composition are among the most consistent predictors of whether a startup survives or collapses.

Startup programs and universities that recruit broadly and hope, without any lens on this, are effectively funding failure.

What does high agency look like in practice? It’s self-directed execution, the willingness to move without permission and update afterward, and the reflex to close a loop rather than escalate it. These are Level 4 and Level 5 traits, and they matter more in innovation work than anywhere else because innovation, by definition, has no playbook. You cannot hand someone a process to follow when the whole point of the role is to figure out a process that doesn’t exist yet. The person who thrives is the one who’s comfortable owning ambiguity, and that comfort is nearly invisible to a résumé screen.

This connects directly to something explored in the invisible talent problem. The people most likely to be Level 5 operators in a startup are frequently people who don’t look like obvious founders or ecosystem leaders yet. They don’t have the pedigree, the prior exits, or the warm-intro network, so they never surface in the channels where hiring and investment attention concentrate. That’s not a diversity slogan; it’s a discovery failure. We’re screening out the highest-agency people because agency doesn’t photograph well on LinkedIn, and then we wonder why the same twelve founders keep getting funded.

Could Predictive Intelligence Spot a Level 5 Before you Hire?

Here’s the question I’ve been circling, and it’s the reason for some of the I’m doing. If experience and references are weak predictors, and agency is a strong one, then the entire hiring problem reduces to this; can you predict agency before someone is on your payroll?

That’s not a rhetorical question.

The central problem in workforce intelligence right now is that turnover is the most expensive recurring failure in the economy, and turnover is overwhelmingly a matching problem. The wrong person landed in the wrong environment because the signals used to match them were historical, biased, and disconnected from how they’d actually behave in a new context. The behavioral question, how does this person operate when they hit friction, is the one that predicts both retention and performance, and it’s the one traditional hiring seemingly never asks. We’ve been putting real work into this with MustardHub, building toward workforce intelligence that reduces turnover by matching on behavior rather than on the misleading proxies we’ve inherited. The correlation is direct; if you can identify the people who consistently operate at Level 4 and Level 5, you reduce turnover in the general economy, because those people are more engaged, more autonomous, and better matched to roles that reward ownership.

Now run that same logic through a startup lens.

If predictive behavioral intelligence can identify high-agency operators in the general workforce to cut turnover, the same instrument, pointed at the startup sector, would do something more valuable still; it would help assemble the teams most likely to succeed, because in startups the team is the primary determinant of success or failure. We’ve had personality research and founder-type research telling us that team composition predicts outcomes. What we haven’t had is a way to operationalize it at the point of hiring, to move from “team matters” as a truism to “here is the behavioral profile of a Level 5 operator, and here is how to find one who doesn’t already look like a founder.”

I want to be careful about the claim I’m making, because the temptation is to overpromise. I’m not saying any platform can do this today, and I’m not saying MustardHub does it. I’m saying the problem is real, the research supports the approach, and the work of building predictive behavioral matching for the general workforce and the work of building higher-odds startup teams are the same work aimed at two markets; besides, this isn’t something AI or HRTech can tackle because it’s an infrastructure layer opportunity. If you can predict who reduces turnover, you can likely predict who executes under ambiguity. Are those not the same person?

Learn about Workforce Intelligence Here

The framework is a management tool long before it’s a hiring tool

Even if you never touch predictive analytics, the five levels are immediately useful as a management standard, and this is where the startup founder should start tonight. Oncken’s thoughts hold; when someone brings you a Level 1 or Level 2 problem, don’t accept it at that level. Send it back with the expectation that it returns at Level 4. Every time you solve a problem an employee could have solved, you’ve taught them that Level 2 is acceptable, and you’ve taken their monkey onto your own back. Do that enough times and you become the bottleneck for every decision in your own company, which is the specific way a great many founders unintentionally cap their own growth.

The practical tips seem embarrassingly simple.

• State the expectation explicitly on day one, that, “Everyone here operates at Level 4, and as we build trust, operate at Level 5.”

• Refuse to accept problems below Level 3

• Praise Level 5 loudly and publicly so the standard is visible to everyone

• And in interviews, stop asking about experience and start asking for a specific instance where the candidate found a problem nobody assigned them, diagnosed it, and fixed it without being asked. Their answer will tell you which level they live at faster than any reference check, and it’s the closest thing to a behavioral predictor you can run without any technology at all.

The organizations that win the next decade of innovation won’t be the ones with the best-credentialed teams. They’ll be the ones that learned to identify agency before they hired it, and to demand it after. The résumé told you where someone has been. It never once told you what they’ll do the moment something breaks and no one is watching.

What I’m realizing through MustardHub and my startup ecosystem work, is that whether you’re building a startup team or a workforce of thousands, if you screened every one of your last ten hires for the level they actually operate at, how many would clear Level 4, and what has it already cost you that you didn’t ask before they started?

The sequence almost every city follows when it decides to “do something” about entrepreneurship seems the same:

• Hire a director,

• Launch an accelerator

• Host a pitch competition

• Put out a press release about becoming “the next Silicon Valley”

• Wonder eighteen months later why founders are still struggling to

Rest assured, your community is doing something like this. I’ve seen the NSF grant announcements, I just heard that a nearby city hired a Head of Innovation (who used to work at McKinsey), and, with autumn around the corner, cities are announcing their Startup Week programming and asking for submissions for the awards and demo days.

Nobody ever asks what’s wrong. No one ever publishes a plan or a press release that isn’t a celebration of startups (or innovation) but is rather, a finding of what doesn’t work and what they plan to do about it. Oh, I know you THINK you did an assessment of what to fix; you probably even claim you did, have data to prove it, and can share the name of the consultant you hired that will agree with you.

In my 25 years of working with cities about startup programs, of one thing I can be certain, none of you have ever actually done so. I have never spoken with a city that knows what’s wrong or isn’t surprised (or grateful) to hear what they should be working on.

I get a kick of how many social media posts by cities celebrate from the Pitchbook / Carta / Government lists announcing the top 10 whatevers… but have you ever seen anything meaningful about the bottom 10 in venture capital? The 10 cities struggling with startups? The 10 sectors of the economy in desperate need of innovation?

Of course not. No one wants to expose what’s wrong. But, because of that, no one is working on what needs to be fixed; everyone is reinforcing the good while ignoring what needs work.

Audit your Startup Ecosystem

In the past few years, I’ve done dozens of regional startup ecosystem audits and despite what you think you have assessed, I know for a fact that none of you have what is needed.

In the sector of our economy in which almost everything fails, working on what isn’t going well is FAR more important than celebrating what is humming along. Especially if you work for the City / Government / Association / VC, founders need your support fixing what’s wrong, not congratulating everyone that you have an Angel Group.

Think of it this way, you have the equivalent of a surgeon opening you up before ordering an X-ray, except the patient is your regional economy and the anesthesia is putting everyone to sleep so they don’t complain that you’re not doing a very good job.

, an economic development and ecosystem builder whom ESHIP Alliance profiled as an influential connector, has become a driving force in the entrepreneurial landscape through his commitment to supporting founders and strengthening collaborative networks, has spent his career pointing out a version of this same problem; that, ecosystems get built by a handful of enthusiastic people before anyone has actually diagnosed what the ecosystem needs. As Chambers put it discussing how most ecosystems form, nearly every regional startup effort of the last two decades “started with a few charismatic connectors,” which is a nice way of saying most cities have been improvising for twenty years and calling it strategy, while under-supporting local leaders who are actually doing the work.

Chambers and I were talking recently and came up with a meaningful acronym for AUDIT, relevant to economic development in this context. This AUDIT reframes ecosystem development the way a competent internist reframes that patient consult; you don’t cut open until you assess, you don’t prescribe until you diagnose, and you don’t diagnose until you actually look. AUDIT stands for Awareness, Understanding, Diagnosis, Insight, and Transformation, and it’s a sequence, not a menu. Skip a step in your startup ecosystem assessment and the rest of the framework collapses, the same way a construction crew can’t pour a foundation without first knowing the soil composition, the water table, and whether the parcel next door is about to become a landfill.

I’m begging you to stop putting taxpayer dollars into that innovation hub that wins headlines while you have no grasp of whether or not the ecosystem can and will actually sustain it. Following AUDIT:

Awareness is the recognition, inside a city, region, or state, that entrepreneurship is an economic development strategy in its own right, distinct from small business support, workforce programs, or corporate recruitment. Most places never get past this stage; they fund a coworking space and call it an ecosystem strategy, which is a little like buying a treadmill and calling it a fitness plan. Awareness means the mayor’s office, the university, the chamber, and the capital sources all recognize that high-growth startups create a different kind of value than a new restaurant on Main Street, and that value requires different infrastructure.

Understanding is where you actually learn what your region’s entrepreneurial conditions look like today, not what you assume they look like based on the last press release from the state’s economic development office. This is where you’re honestly assessing whether you have the foundational conditions laid out in The 6 Considerations of the Economic Development of Startups: a culture that rewards competitive risk-taking, real wealth in the region, employers innovative enough to give founders somewhere to land if they fail, minimal government friction, people who have actually worked inside startups rather than small businesses, and a credible, honest story about what your region is actually good at. Understanding means you stop guessing and start measuring.

Diagnosis is the uncomfortable part, because diagnosis means naming what’s broken instead of what’s merely absent. A city can have plenty of activity, demo days, meetups, an “innovation district” with a nice sign, and still have a diagnosably sick ecosystem if none of that activity produces founders who reach revenue, capital that recirculates locally, or mentors who actually know the difference between advising a startup and advising a small business. Diagnosis is where you find out your ecosystem doesn’t have a marketing problem, it has a structural one, which is a distinction most consultants are financially incentivized never to make for you.

Insight is what you get once diagnosis has told you what’s actually broken and you translate that into a specific, prioritized set of interventions, in the right order, for your region’s particular conditions. Insight is not a generic playbook borrowed from Austin or Boulder; regions are not software, they don’t run on someone else’s code, and pretending otherwise is why so many “innovation strategies” read like they were written by consultants who’ve never met the region they’re advising.

Transformation is the fifth letter, and it is deliberately the only one of the five that isn’t research. Transformation is implementation, it’s the actual building, funding, and operating of the organizations, programs, and policy changes that Insight identified as necessary. It requires different skills than the first four letters entirely; it requires ecosystem builders, program operators, capital allocators, and policy staff who can execute over years, not analysts who can diagnose over months.

That last distinction is the entire point, and it’s why our regional ecosystem assessment work is built around exactly the first four letters identified in Chambers’ framework.

What we do as Startup Economists, having worked now with regions from Liège to Croatia to Colombia to Kansas City, are the Awareness, Understanding, Diagnosis, and Insight phases of a genuine audit; call it AUDI, which admittedly sounds like a car company, but the analogy actually holds up better than you’d expect. You wouldn’t ask the mechanic who diagnoses your transmission to also be the one who spends the next three years driving the delivery routes. Diagnosis and operation are different disciplines requiring different people and conflating them is why so many regional “strategic plans” get written, printed, celebrated at a press conference, and then shelved because the people who wrote the plan were never equipped, funded, or positioned to run the plan.

Please hear me when I advise you that you have to do this, because you aren’t.

I sit in meetings with city officials who start by applauding their local accelerator only to then admit, “yeah… it really isn’t working.” I’m tired of hearing Chamber of Commerce tell me they have a funding problem while the local Angel Group has an office right next door. The next time I visit the celebrated Innovation Hub, to find it practically empty, I’m going to scream.

You all are launching a Startup Incubator because that’s what Bentonville, has, and no one stopped to ask if that’s what you ACTUALLY need, that is what will work, and that you have the other aspects of an ecosystem so that the incubator is even capable of working.

The AUDI work of a complete AUDIT isn’t guesswork dressed up as expertise, either; it’s built directly on the same two frameworks published and used to structure meaningful startup ecosystem development engagements. The Understanding phase runs directly against the Six Considerations; a region that thinks it has a “founder problem” often actually has a wealth problem, or a government-interference problem, or worse, no credible answer to what kind of startups it should even be attracting. And the Diagnosis and Insight phases map directly onto the ten dimensions of entrepreneurial capacity outlined after seeing the same pattern repeat across New Mexico, Alberta, Queensland, Lisbon, and Tulsa; silos that keep every organization operating in its own bubble, a missing middle where high-growth companies fall into a gap between accelerator graduation and Series A, ecosystem builders who are underfunded to the point of burnout, metrics that track activity instead of outcomes, collaboration that has to be negotiated instead of assumed, invisible talent nobody bothered to go find, environments that don’t actually let people perform, government and academia and the private sector that don’t talk to each other, regions chasing trendy sectors instead of their own comparative advantage, and global best practices copied instead of adapted.

Ten dimensions, and a proper AUDI tells you specifically where your region is struggling, in what order or priority, and why. That’s the map. Diagnosis without a map is just an opinion with a nicer font.

The sequence matters; most regions get this backward as cities routinely start at Transformation.

Don’t Start with the T! Don’t Even Keep Doing T if you are Now

You fund the accelerator, hire the director, and launch the innovation district, all without ever doing Awareness, Understanding, Diagnosis, or Insight first.

I’m just full of analogies today so here’s another one; you’re a gambler walking into a casino, skipping the odds entirely, and betting the mortgage on whichever table has the loudest crowd around it.

Then when by sheer luck, it works, you claim a win which then, thanks to an accident, causes the region to spend the next decade telling a survivorship-bias story about how their one lucky bet was actually a strategy. Most of the time it doesn’t work, and the region spends the subsequent decade wondering why the accelerator graduated forty companies and none of them are still headquartered locally (even they even still exist).

Do the AUDIT now, or again if you want to think of it that way; it’s not an argument about process for its own sake.

It’s an argument about capital efficiency. Every dollar a region spends on a program built on a wrong diagnosis is a dollar it can’t spend on the program that would have actually worked. Every year a well-meaning economic development office spends running events instead of fixing the structural gap in early-stage capital is a year founders spent leaving for a region that did the diagnostic work first. Transformation is genuinely exciting; it’s ribbon cuttings and headlines and the part everyone wants to be at the podium for. But Transformation without Diagnosis is theater, and theater doesn’t recirculate capital.

Has anyone working for your city, in the economic development office, through your university, at your chamber, or in your capital community ever formally answered what’s broken, specifically, in what order, before deciding what to build next? I would bet I can answer that question for you because I have never encountered a city that has done it, regardless of what they say. Your region, like most, simply started at Transformation because a consultant provided sufficient affirmation that they should and because Transformation is the part that photographs well.

Learn More and Request an AUDIT Here

If you’re an ecosystem builder, a policy professional, or an investor looking at a region and sensing that something structural is off, but you can’t quite name it, that instinct needs to be taken seriously rather than talking yourself out of it. That’s usually Diagnosis trying to happen without the tools to finish the job.

I’ve sat in enough state capitol conference rooms and federal agency briefings to know the exact moment it happens. Someone on my side of the table says “ecosystem;” then a policy maker, a legislative staffer, or a federal agency director flinches like I’d said something in Klingon, and a redirect comes, “we don’t want to get too conceptual.” Conceptual. As if an innovation ecosystem is a mood board and not the literal mechanism by which every economy on earth, including the one paying that person’s salary, actually functions.

So let me put it this way…

Have you heard of economies? Industries? Innovation?

“Yes.”

Congratulations, you’ve just described an ecosystem in three separate words and refused to say the one word that unifies them. Vizzini did something similar in The Princess Bride, quizzing Westley on Plato, Aristotle, and Socrates before landing on his verdict, and I’ll let you guess which word applies to the state and federal officials doing this with public capital right now.

This isn’t a semantic tantrum on my part. It’s an accusation with a paper trail, and I’m going to walk you through the history of the word, why government avoids it on purpose, and five separate examples of your tax dollars getting shredded because policymakers would rather fund a thing than fund a system.

Do We Even Know What “Ecosystem” Means?

The word is younger than you’d think.

British botanist Arthur Tansley coined it in 1935, in a paper for the journal Ecology called “The Use and Abuse of Vegetational Concepts and Terms.” Tansley built it as a contraction of “ecological system,” borrowing the word “system” directly from physics, because he needed a term for something bigger than a single organism or a single species. He wasn’t describing a nice forest. He was describing the whole interacting complex of organisms and the physical environment around them, insisting that you can’t understand a plant, a predator, or a nutrient cycle in isolation from the soil, the climate, and every other organism sharing that space. Pull one piece out and you don’t get a smaller version of the same system. You get a different, usually worse, system.

Move that concept into business and you land on James F. Moore, who published “Predators and Prey: A New Ecology of Competition” in the Harvard Business Review in 1993. Moore argued that competitive advantage doesn’t come from a single company’s performance but from the collective strength and coordination of an entire network of firms co-evolving around an innovation, using Apple’s sprawl across personal computers, consumer electronics, and communications as his example. That article won the McKinsey Award, got cited a few thousand times, and gave government offices in the country a word they now seem to dodge in public meetings. The irony writes itself.

Now, ask yourself, does a country want to support one company, and their innovation, in a globally competitive market? Or, do we want the collective strength and coordination of an entire network so that the work is competitive?

Break the word down and it gets even more obvious. “Eco” derives from the Greek “oikos,” meaning household or dwelling, which is also the root of “economy,” itself literally “household management.” “System” means an interconnected set of parts functioning as a whole. An ecosystem, in the economic sense that matters to you, is a system for managing the household of an economy: capital, labor, ideas, and institutions, all interdependent, all producing outcomes as a whole that no single part can produce alone.

Sounds like a primary purpose of government; which is to say, explicitly what governments should be supporting rather than avoiding being so involved in the public good of the economy. When I hear a federal agency hint that they’d rather avoid “the ecosystem language,” they are, etymologically, telling me they’d rather not think about the economy as a system.

Read that back to yourself slowly.

The Policy Instinct That’s Half Right and Entirely Mismanaged

Here’s where I’ll be less antagonistic and more precise, because the underlying policy instinct isn’t crazy. There’s a real, defensible argument that government shouldn’t try to pick winners, that markets discover value more efficiently than committees, and that the private sector should be left to sort out where capital and talent actually belong. I agree with that in principle. Hayek was right that dispersed knowledge beats centralized planning, and nothing about building startup ecosystems requires abandoning that.

The problem is what happens when that correct principle collides with a budget cycle.

As a matter of policy, “let the private sector work it out” is sound. As a matter of practice, inside a legislature that needs a ribbon to cut and a press release to send before the next election, “letting the private sector work it out” turns into funding the single most visible, most fundable, most photogenic piece of an ecosystem while ignoring every other condition that piece depends on to function. You don’t get restraint; you get a linear allocation of capital into a hub (something that is inherently part of a network), followed by confusion when the hub doesn’t produce the systemic outcome it was never structurally capable of producing alone.

I wrote an entire book about this, Startup Ecosystems: Understanding Why Startups Thrive and Ecosystems Fail, because I got tired of watching this exact pattern repeat itself in city after city, state after state. In it, I lay out two diagnostic frameworks that map what an actual ecosystem requires.

The Six Considerations of the Economic Development of Startups cover the foundational conditions: a culture that celebrates risk and ambition rather than penalizing it, reasonable private wealth that’s actually engaged in early-stage risk rather than parked in safer assets, innovative employers in relevant sectors who give founders a safety net if the startup fails, minimal government interference in the form of regulation and licensing that makes no sense for pre-revenue ventures, access to people with genuine startup experience rather than corporate consultants repackaged as mentors, and credible, specific regional promotion instead of generic “we’re the next Silicon Valley” branding.

Layered on top of that is the Ten Dimensions of Entrepreneurial Capacity, which map the operational infrastructure an ecosystem needs once the foundation exists: breaking down the silos between universities, accelerators, and government agencies; building support for the “missing middle” of high-growth firms past the seed stage; funding the ecosystem builders and connectors who do the unglamorous coordination work and are almost always broke; measuring outcomes instead of activity; building genuine collaboration culture; widening discovery so founders outside the usual visible circles actually get found; architecting high-performance environments; aligning government, academia, and the private sector around shared goals; building around real local competitive advantages instead of chasing whatever’s trendy; and adapting global best practices instead of copy-pasting them.

Every single one of those sixteen factors interacts with the others. Fund one in isolation, ignore the rest, and you get exactly what most regions have: a lot of activity and very little consequence. That’s not a moral failing on the part of policymakers. It’s a structural one, and it’s completely predictable once you understand what political incentives actually reward. Programs are visible, scoped, staffed, and photographed regardless of whether they produce anything a market would call valuable. Conditions, the boring regulatory and capital-formation work underneath the programs, produce no ribbon cuttings and rarely get funded first, which is backwards, because conditions are the foundation everything else sits on.

Dr. , Co-founder of Ecosystem Edge, built a scorecard to measure how connected the different players in a region truly are; not what programs exist, but whether the people running them work together. Run it across regions and the same pattern I talk with legislators incessantly about emerges, every time; it’s never a missing program, it’s a missing relationship, “Everyone funds the anchor institutions (the university, the accelerator, the workforce agency) and assumes that’s an ecosystem. It’s not.” An ecosystem isn’t even a collection of institutions, it’s a set of compounding relationships between the people inside them, whose work is interdependent but who under the current incentive structures, often act in silos.

Worth a read: Cities Measuring Outcomes

Two Failures You’ve Already Watched Happen

You’ve seen this play out in your own city, even if nobody called it by name. A region funds an Innovation Hub, complete with that modern industrial design, the standing desks, and the startup meetups, and then discovers that corporate partners and venture investors won’t show up because the hub was never connected to the deal flow, the anchor employers, or the capital density that would make participation worth their time. The building exists; the ecosystem around the building does not, and the building alone was never going to produce one. You don’t get network-level outcomes out of a hub-and-spoke budget. That’s never a funding problem. It’s a design problem.

Or a state pours money into university research grants under the theory that more research equals more commercialized technology equals more regional jobs, without fixing the fact that technology transfer offices are structurally incentivized to license IP for guaranteed royalty income rather than spin out founder-led companies that might actually scale. The research gets funded; the commercialization pathway stays broken. The grant produces papers, not companies, and the region gets to claim it, “invested in innovation!” while the actual mechanism for turning that investment into an economy remains untouched.

Both of those are linear allocations; which is to say, money goes into one node, the node does its narrow job competently, and the systemic outcome never arrives because the rest of the system was never funded, coordinated, or even acknowledged as necessary.

Now let’s go find five more, at the federal level, with real dollar signs attached.

Five More Times Washington Wrote a Big Check and Skipped the System

Opportunity Zones. The 2017 tax bill created this program to funnel capital gains into distressed census tracts, deferring and eventually forgiving tax on investments held long enough. The theory was straightforward: give investors a reason to put money into places capital normally avoids. In practice, the Treasury’s own Office of Tax Analysis found roughly two-thirds of Opportunity Zone investee businesses were concentrated in real estate, construction, or lodging, not the small businesses or startups the program was sold on, and the Urban Institute found that 93 percent of investment flowed to metropolitan areas already showing signs of economic improvement rather than the places the program was ostensibly designed to lift. This is a linear allocation of a tax incentive with zero attached infrastructure for mentorship, deal flow, or founder support; it assumed capital alone would build an ecosystem and instead subsidized apartment buildings in neighborhoods that were gentrifying anyway.

Manufacturing USA Institutes. Since 2014, this network of federally sponsored advanced manufacturing institutes has received hundreds of millions in annual funding across Commerce, Defense, and Energy. The Government Accountability Office’s (GAO) own evaluation, straightforwardly titled to make the point for me, found the institutes had shown initial accomplishments but faced ongoing challenges measuring performance and ensuring they could sustain operations once federal funding ran out. Each institute is a real, technically competent node. What was never built alongside them was the regional capital formation, workforce pipeline, and commercialization infrastructure to keep them alive without a permanent federal drip line, which is precisely the ecosystem-level work that “just fund the institute” was never going to do on its own.

The CHIPS and Science Act. Fifty-two billion dollars in federal incentives, catalyzing hundreds of billions more in private fab construction across the country, is about as linear an allocation of capital as you’ll find: build the factory, and the ecosystem will apparently assemble itself. It hasn’t, though you certainly probably think it has given all the semiconductor announcements. Industry projections put the semiconductor sector short roughly 67,000 of the 115,000 new jobs it needs filled by 2030, and the talent shortage is now tied with tariffs as the industry’s top concern among semiconductor executives. You cannot fund a fab and skip the ten-year regional workforce pipeline, the community college partnerships, and the housing and immigration policy that determine whether skilled workers can actually live near the plant and then act surprised when the plant sits half-staffed. That’s not a chip shortage; that’s an ecosystem nobody funded.

“And YOU! Friendless. Brainless. Helpless. Hopeless! Do you want me to send you back to where you were? Unemployed? In Greenland?! Morons!”

Dr. Beaird added, “Government treats translation and commercialization like a technology and legal problem: fix the IP terms, fix the capital stack, fix the tech transfer policy. But none of it moves until relationships of trust, reciprocity, and shared value turn into galvanized initiatives that compound. Scale without coherence is just expensive chaos; the technology is rarely the constraint, the relationships and experience around it.”

SBIR and STTR. America’s oldest federal seed fund for small business R&D has handed out billions since 1982, and it genuinely produces real technology. It also produces the single most on-the-nose metaphor; practitioners call the gap between a completed prototype and a viable commercial product the “valley of death,” and Congress created Commercialization Readiness Programs specifically to bridge that gap, though implementation has been uneven across agencies. Meanwhile, GAO found just 22 firms, less than 1 percent of participants, captured about 10 percent of all Phase II dollars between 2011 and 2020, a concentration critics have started calling “SBIR mills” because they’re optimized for proposal volume rather than getting technology into the market. You funded the invention; you did not fund the bridge to the customer, the follow-on capital, or the market discovery work that turns an invention into a company, and you’re mostly funding the same repeat winners because nobody built the discovery infrastructure to find anyone else. Sounds a bit like what universities need to fix, doesn’t it?

The EDA’s Build Back Better Regional Challenge. A billion dollars distributed to sixty regional coalitions in 2022 to build industry clusters, with a genuinely sophisticated selection process and a serious research partnership with Brookings and Purdue to study what worked. What actually got funded, in large part, was coordination; designated coalition leads and Regional Economic Competitiveness Officers to serve as the coordinating “CEO” role for each cluster, plus millions more in grants just to study and document the program’s own outcomes. That’s money spent describing the system rather than building the capital formation, anchor employer density, and private investor engagement that would let those clusters function without EDA’s hand on the wheel indefinitely.

Studying an ecosystem is not the same as funding one, and a billion dollars is an expensive way to discover that distinction. Trust me, this is literally my job description.

Every single one of these programs is staffed by competent, well-intentioned people.

None of them are corrupt; all of them are structurally guaranteed to underperform their stated goals, because each one took a systemic problem and applied a linear solution to it:

• Fund the visible node

• Skip the invisible infrastructure

• Measure activity because outcomes take a decade and nobody’s still in office by then

It’s not that government doesn’t understand ecosystems; it’s that admitting an ecosystem is what you’re managing means admitting that no single grant, tax credit, or ribbon-cutting can fix it, and that’s a much harder thing to put in a two-minute floor speech. Well, it is harder for some, ecosystem builders do this in their sleep, but who asks them??

What Happens When You Say Ecosystem Out Loud

If you run economic development for a city, a state, or a federal agency and you’ve been avoiding “ecosystem” in your own strategy documents because it sounds too academic, theoretical or too hard to measure, ask yourself honestly which of the sixteen structural factors above you’re actually funding, and which ones you’re assuming will just show up because you funded the shiny one.

Grab a copy of Startup Ecosystems here. A quick read, it gives you the direction.

“Ecosystem isn’t the fuzzy word here,” noted Dr. Beaird. “Funding the same linear logic that never produces outcomes, that’s fuzzy. Coordination without connection is what gets funded instead, because a hub is easy to put in a budget brief. Compounding relationships between people who’ve never had a reason to talk to each other are not. Every time we score a region on that, the same people and resources go undiscovered and disconnected; not from lack of funding, but because nobody built the relationships to reach them.”

The Sixteen Factors, and the One Nobody Budgets For

Go back through every example in this piece and a pattern emerges that has nothing to do with which agency wrote the check.

The Innovation Hub skipped the innovative employer’s consideration and the alignment dimension; it never connected government, academia, and the private sector around a shared goal, so corporate and venture participation had no reason to show up.

• The university grants skipped access to startup-experienced people and the missing middle; a technology transfer office staffed by licensing administrators is not the same as a founder-experienced mentor, and a grant that ends at the patent filing never reaches the stage where a company needs Series A support.

• Opportunity Zones skipped funding the local operators and ecosystem builders, and credible, specific promotion; capital gains deferrals reached developers, not the connectors and mentors who’d have routed that capital toward an actual founder.

• Manufacturing USA skipped the missing middle and regional capital formation.

• SBIR skipped the same missing middle from the other direction, funding invention with no bridge to commercialization.

• CHIPS skipped innovative employers as an ecosystem input rather than an output, treating the presence of a fab as sufficient instead of building the housing, training, and regional workforce pipeline a fab actually needs.

• The EDA’s Regional Challenge skipped overcoming silos in practice, funding coordinator job titles instead of the shared infrastructure that would let coalitions coordinate without a coordinator.

Read those seven failures side by side and the missing factor isn’t one line item on the list of sixteen. It’s the layer connecting all of them, and it’s the one governments consistently treat as if it doesn’t need a budget: infrastructure. Not roads and bridges; the twenty-first-century kind. The case is made more directly in States, This Is a Startup Ecosystem Model That Works, infrastructure is the foundational capacity that makes economic activity possible in the first place, the shared connective tissue, like a common system founders and investors can actually use, that sits beneath every program and makes the program’s output legible to the rest of the ecosystem instead of trapped inside it. Without that layer, overcoming silos stays a slogan, funding the actors stays an afterthought, and access to startup-experienced people stays with whatever consultant happened to answer the RFP. Every one of the sixteen factors this book maps depends on a coordinating infrastructure to connect it to the other fifteen; fund the fifteen individually and skip that layer, and you get seven federal programs, and probably a few in your own city, that spent real money doing real, competent, individually defensible work that never once added up to an ecosystem.

That’s the check to run before your next budget cycle.

Not, “which program should we fund next,” but “does anything in our region actually connect the programs we already have.” If the honest answer is no, you don’t have a funding gap. You have an infrastructure gap, and infrastructure, unlike a ribbon-cutting, doesn’t care whether you’re willing to say the word out loud.

In 1897, Nikola Tesla stood at a gala in downtown Buffalo called the “Power Banquet” and declared that the city, newly electrified by his alternating-current system transmitted 25 miles from Niagara Falls, was “sure to become one of the greatest industrial centers of the globe.” The New York Tribune called the Niagara-to-Buffalo power transmission “one of the triumphs of the century.” As the Tesla Science Center at Wardenclyffe records, Tesla had just pulled off the most important feat in electrical engineering history on the Niagara Frontier, and he had done it by ignoring every piece of conventional wisdom that said it couldn’t be done; most of it from Thomas Edison, who responded to the competition with a public campaign electrocuting animals to frighten people away from alternating current.

If there is a metaphor for what it takes to build something genuinely new in Buffalo, that is probably it; ignore the incumbent with the better PR, harness the force that everyone else considers too dangerous, and light up an entire city.

Buffalo has been doing exactly that, in fits and starts, for more than 130 years. The city that gave the world the electric grid, the air conditioner, the cardiac pacemaker, the cheerful chaos of the Buffalo wing, and the first Nickelodeon channel (yes, that started in Buffalo in 1977 as “Pinwheel” before going nationwide) has never lacked for inventive energy. What it has lacked, historically, is the structural architecture that converts inventive energy into scalable companies; that gap is closing faster than most people outside Western New York realize, and understanding why it is closing, and where the remaining gaps are, requires something more than a cheerleading press release from a state agency.

The frame for this analysis is the one developed in Startup Ecosystems: Understanding Why Startups Thrive and Ecosystems Fail, which examines not whether a region is doing things (because almost every region is doing things) but whether the things being done are structurally sound enough to produce compounding outcomes. Buffalo earns high marks in several dimensions of that framework; it earns some hard questions in a few others so let’s get to know it better.

A City Built on Audacious Engineering

You cannot understand Buffalo’s entrepreneurial potential without understanding what the city actually is, which most people who haven’t spent time there cannot tell you. Buffalo sits at the northeastern end of Lake Erie, 20 miles from Niagara Falls, positioned at the convergence of water, rail, and (thanks to Tesla) electrical power that made it one of the wealthiest and most industrially significant cities in the United States by the turn of the 20th century. When the Pan-American Exposition was held in Buffalo in 1901, the city had a population of approximately 350,000 and was the eighth-largest city in the United States, chosen in part because it was within a day’s journey by rail or steamship for more than 40 million people.

That Exposition (officially nicknamed “The Rainbow City” because of its spectacular electrical illumination) was itself a statement about what Buffalo represented to the world at the time. A major highlight was electric lighting powered by hydroelectric energy from nearby Niagara Falls, including the Electric Tower standing 410 feet tall, and in addition to electricity, the Exposition displayed other cutting-edge inventions such as incubators for infants and X-ray machines. Buffalo was not just an industrial city; it was a showcase city for the most advanced technology on the planet.

On November 16, 1896, Nikola Tesla flipped the switch that made Buffalo the first electrified city in the world, and this first long-distance transmission of alternating current electricity helped power Buffalo’s street cars and light up its streets, before the entire world converted to Tesla’s innovations. In elementary school, Tesla had seen a picture of Niagara Falls and dreamed of using water to generate power, and in 1893 he turned that dream into reality, and in 1896 the flip of a switch sent the first power surge from Niagara Falls to Buffalo.

What happened next is the part that most contemporary startup ecosystem analyses skip over because it is uncomfortable when it happens – Buffalo’s extraordinary position did not last.

The 20th century’s shift away from heavy manufacturing, the decline of the steel and grain industries, and the broader deindustrialization of the Rust Belt reduced a city that once rivaled Chicago in commercial significance to one that economists were still describing as below its pre-pandemic employment peak as recently as Canisius University’s 2024 economic analysis. The Federal Reserve Bank of New York notes that the Buffalo MSA is home to more than 1.1 million people and historically built its economy on manufacturing, with long-running wage disparities relative to national averages reflecting the ongoing transition away from that industrial base.

That context matters in understanding what the startup ecosystem is actually working with, and against. A city recovering from decades of deindustrialization has real assets (deep industrial knowledge, affordable infrastructure, institutional resilience, a workforce accustomed to solving hard physical problems) and real liabilities; such as, a thin layer of private capital concentrated in sectors that don’t necessarily circulate into startup investing, a modest density of high-income earners who write angel checks, and a cultural memory of risk that is shaped more by what was lost than by what was gained.

The Buffalo Niagara region GDP sits within the Buffalo-Rochester combined market at approximately $239 billion, and the area’s primary employment sectors are healthcare and life sciences, financial services (anchored by M&T Bank), manufacturing and industrial, education, and government.

The startup ecosystem’s best long-term strategy runs directly through them.

The Invention Tradition: A Narrative of Creativity the Region Barely Credits Itself For

Every good startup ecosystem has what the Startup Ecosystems framework calls a culture of competition, potential, and creativity; the shared belief that building something ambitious is a rational and celebrated thing to do. Buffalo’s problem has never been a shortage of invention; it has a shortage of institutional credit for the invention tradition it already possesses.

Consider the list.

In 1855, the first railway suspension bridge in the world opened over the Niagara Gorge, built by John A. Roebling as the prototype for future bridges including his Brooklyn Bridge; in 1902, Willis Carrier, a new employee at the Buffalo Forge Co., invented the air conditioner; in 1943, Curtiss-Wright Corporation developed a pioneer high-velocity wind tunnel at the Buffalo airport that later became Calspan and invented crash test dummies; in 1959, Wilson Greatbatch invented the internal pacemaker; and in 1964, Teressa Bellissimo at the Anchor Bar concocted the world’s first chicken wings recipe.

Add to that list that Herman Hollerith, born in Buffalo, was a statistician who developed a mechanical tabulator based on punched cards whose company was eventually merged into others to form IBM; Robert E. Rich, Sr. of Buffalo, in 1945 invented the first non-dairy whipped topping that could be frozen; and of course, Pinwheel, which became Nickelodeon. Willis Carrier’s air conditioning invention came specifically from his work at the Buffalo Forge Company, making Buffalo the birthplace of an industry that now generates over $150 billion annually in global revenue.

None of these inventions happened in a vacuum of individual genius; they happened because Buffalo’s industrial infrastructure, its manufacturing base, its proximity to Niagara’s energy, and its position as a commercial crossroads created the conditions under which people with ambitious ideas could find the resources and the customers to turn those ideas into something real. That is, in a nutshell, what startup ecosystems are supposed to do. Buffalo has been doing it, on and off, for 150 years; what has changed in the last decade is the intentional effort to build the connective tissue that makes it repeatable.

The Buffalo Economy as a Macroeconomic Operating Environment for Startups

Healthcare and life sciences represent the most significant anchor sector in the Buffalo Niagara economy today.

Kaleida Health, Roswell Park Comprehensive Cancer Center, and the Buffalo Niagara Medical Campus, Inc. together constitute a multi-billion-dollar institutional cluster that generates research, employs scientists, and creates the kind of patient institutional environment that life sciences startups need to survive long development cycles. The University at Buffalo’s medical school and research infrastructure contribute directly. This is not just a jobs story; it is a founder safety net story, which the Startup Ecosystems book frames as critical; founders who know that if their company fails, they can return to a meaningful sector-relevant job will take risks that founders without that option cannot afford to take.

Financial services, anchored most visibly by M&T Bank, represent both a source of institutional capital and a potential customer base for fintech and financial infrastructure startups. Delaware North, a hospitality and entertainment company headquartered in Buffalo, is one of the largest privately held companies in the United States and represents exactly the kind of anchor employer that can be a customer, a talent magnet, and an eventual acquirer for startups operating in hospitality technology, sports analytics, and food service innovation.

Manufacturing and industrial (the legacy of Buffalo’s industrial base) is undergoing the renovation the world is experiencing in manufacturing.

There are problems that industrial companies face in Western New York such as workforce planning, equipment maintenance, supply chain coordination, quality management, and compliance. These are not glamorous problems by the standards of a San Francisco pitch deck, but they are problems that have enormous economic scale and that the incumbents solving them are, by and large, doing so with software from the 1990s. , co-founder of ACV Auctions, has argued that regional markets like Buffalo have distinct advantages that coastal investors miss, including deep industrial expertise, established customer bases, and “unsexy” problems that represent untapped opportunities. That is not a consolation prize for not being in Silicon Valley; that is a competitive advantage, provided the ecosystem learns to market it as one.

Government plays a more significant role in the Buffalo startup economy than in most coastal markets, for reasons that are both structural and political. New York State has made substantial bets on the region through Empire State Development, the NY Power Authority, and a series of programs that funnel capital, tax incentives, and institutional support into the startup ecosystem. The START-UP NY program allows companies located in designated zones on university campuses to operate tax-free for 10 years, which is a genuine advantage for early-stage companies burning cash and trying to extend their runway. The Buffalo Billion initiative, launched under Governor Andrew Cuomo in 2012, directed substantial state investment into the region’s economic revitalization and laid groundwork for much of the institutional infrastructure that followed. Governor Kathy Hochul has continued this orientation; Empire State Development President Hope Knight has noted that “ten years of hard work at 43North has helped turn Buffalo into a well-recognized epicenter of entrepreneurial ventures.” 43North is one the city’s largest startup development organizations.

The tension in this government involvement is the same tension that the Startup Ecosystems framework identifies as a structural failure mode across most public-capital-heavy ecosystems: government moves toward deployment visibility and political safety rather than toward the conditions that create durable value. More on that shortly.

What Buffalo Has Built

Before getting into the structural analysis, Buffalo’s startup ecosystem has produced real companies, not just press releases.

ACV Auctions is the headline. George Chamoun, the entrepreneur behind ACV Auctions, is the same one behind the earlier success of Synacor, representing the serial entrepreneurship pattern that signals a maturing ecosystem. ACV went public on the NASDAQ in March 2021 and is currently the nation’s leading wholesale automotive auction marketplace. 43North has invested in a diverse portfolio of 69 companies and those startups have gone on to raise over $1 billion in capital, with nearly 60% maintaining a material presence in Buffalo and creating over 1,000 local jobs.

“When 43North was started, the goal was simple but ambitious: attract, finance and grow world-class companies in Buffalo. Today, it’s done exactly that. But more importantly, it has fundamentally changed the belief in what’s possible here. This milestone reflects what’s been built over time and recognizes the founders, partners, and community who made it possible.” – Jordan Levy Founding Chairman, 43North

Among notable 43North alumni: Genetesis (cardiovascular diagnostics biotech), Kangarootime (childcare management software, which acquired an AI platform in 2024), HelixIntel (industrial asset management), Ognomy Sleep (sleep apnea telemedicine), Rarebird, Inc. (functional coffee, relocating from San Francisco to Buffalo), Circuit Clinical (clinical trial management, which was acquired in 2025 according to Buffalo Angels portfolio records), and Top Seedz (organic snack food, leveraging New York Power Authority support and expanding to a 35,000-square-foot facility downtown).

Beyond 43North alumni, established companies including Paychex (headquartered in Rochester but deeply embedded in the ecosystem), Delaware North, M&T Bank, Kaleida Health, Moog Inc. (aerospace and defense, one of the most significant industrial employers in the region), and Roswell Park represent the anchor employers that give founders a fallback and give startups potential early customers. Synacor, one of the earlier technology exits, contributed alumni who went on to found and build subsequent companies; the recycling of talent and capital that healthy ecosystems require.

The 43North Foundation: The Institutional Spine

If you want to understand Buffalo’s startup ecosystem at its center, 43North is the right place to start and the 43North Foundation is where the most interesting and consequential next chapter is being written.

43North launched in 2014 with state backing as what was marketed as “the world’s largest business plan competition,” awarding $5 million annually to five winning startups. The pitch competition format (live, in front of thousands of Buffalo community members at Shea’s Performing Arts Center) was theater in the best possible sense; it made entrepreneurship visible, celebrated, and exciting in a city that had not had much to celebrate economically for decades. The visibility function cannot be dismissed; ecosystems require narrative, and 43North gave Buffalo a narrative.

With a unicorn and publicly-traded company among 43North’s ranks and several others on the rise, 43North co-invests alongside VCs like Bessemer Venture Partners, ICONIQ Capital Group, and Rise of the Rest. 43North’s mission is supported by sponsors including New York State, Empire State Development, M&T Bank and others.

Now, after more than a decade of building that foundation, 43North is entering its next era, with the 2026 Finals serving as the celebration of a planned evolution rooted in Buffalo’s success, opening the door to the next chapter of innovation, company creation, and long-term ecosystem growth.

The 43North Foundation’s Lightning Strike initiative, a 10-year, $100 million commitment announced in 2024, represents a meaningful upgrade in ambition. Lightning Strike has four focus areas: the Radial Ventures venture studio, TechBuffalo (talent pipeline), corporate connectivity, and Series B(uffalo) (ecosystem storytelling). That architecture reflects genuine learning about and application of what ecosystems need beyond pitch competitions and accelerator cohorts.

Radial Ventures: The Most Important Bet in the Ecosystem Right Now

If 43North is the institutional spine of the Buffalo ecosystem, Dan Magnuszewski is one of the operators most worth watching. Magnuszewski is a University at Buffalo computer science graduate who co-founded ACV Auctions as CTO, helped scale it to unicorn status with investment from Bessemer, Bain, and SoftBank, saw it through its NASDAQ IPO in 2021, and then made a decision that would be unusual in any ecosystem: he stayed in Buffalo and bet on it again.

“Our sole purpose is to help Buffalo grow and succeed,” says Dan Magnuszewski, in this case, CEO of 43North Foundation‘s venture studio, and he adds, “The goal is to create successful and exciting startups that people want to work at. By having these attractive opportunities, it not only helps retain talent, but it can also help to bring back people who may have left Buffalo for a variety of economic reasons.”

After co-founding ACV Auctions and taking it public with $350 million in funding from major firms like Bessemer and Bain, Magnuszewski launched Radial Ventures in Buffalo to back “unsexy” ideas that serve local industrial markets most investors overlook, leveraging Buffalo’s decades of industrial expertise and real customer base to build companies specifically designed to solve problems that coastal investors typically ignore.

Magnuszewski and his co-founders sold their houses and downsized to reduce expenses, taking advantage of Buffalo’s low cost of living with $1,200 monthly mortgage payments on nice four-bedroom homes, raised $1 million locally, launched their marketplace within five to six months, and achieved 350% year-over-year growth for five years while building their engineering team in Buffalo and expanding to Toronto.

The first company launched by Radial is Makeoff, an AI platform that helps electrical contractors generate accurate bid estimates from blueprints in minutes rather than days. Radial Ventures is led by CEO Dan Magnuszewski, CTO Mike Canzoneri, and a team of experienced software engineers and designers, and debuted in early 2025 with plans to launch a steady stream of new companies every year, using a proprietary market validation and research tool that assesses ideas based on the perceived problem, marketability, and market size. That is not a startup incubator running a curriculum. That is a venture studio that identifies real problems in industries Buffalo understands deeply and builds companies around those problems from the ground up. It is one of the most operationally sophisticated interventions in the ecosystem.

Bill Maggio, board chair of the 43North Foundation, said of Magnuszewski: “Dan easily could have gone in a different direction and pursued other opportunities. Instead, he decided to double down and give back in the best way possible and help our community.”

Connecting Buffalo to the Capital Markets That Actually Write Checks

is one of those rare ecosystem figures who operates credibly in multiple cities simultaneously without becoming a satellite of anywhere. A Buffalo-area native who spent years as Managing Director of the Founder Institute in New York, Siskar built a portfolio of over 160 early-stage technology companies and was named Best Startup Ecosystem Developer in the entire global Founder Institute network in 2016. He also served as vice president of portfolio and selection at 43North, which gave him the unusual vantage point of being simultaneously embedded in both the New York capital market and the Buffalo founder community.

“Don’t go looking for ideas. Go have experiences. Typically, when entrepreneurship works well something is going to hit you that is inefficient and doesn’t make sense. And even that is not enough. You need to wait until it’s a thing that keeps you up at night. When you’re in the shower or walking to your car and you can not stop thinking about it and you get excited about it and need to get after it. These things are really hard to do and you need that, otherwise you’ll fizzle out after 6 months, 1 year, or 2 years. You won’t build anything great if you’re not giving it all your attention,” Siskar has said, describing the founder orientation that ecosystems need to cultivate.

Siskar is the CEO of Finta, the Capital Copilot transforming private markets by automating deal-driven workflows, tracking over half a billion dollars annually across tens of thousands of investor relationships, and as General Partner at Firehouse Ventures, a syndicate with over 500 members that invests in innovative pre-seed and seed-stage startups.

Siskar’s direct engagement with the Buffalo ecosystem through 43North’s investor education work (where he laid out tips for those looking to join Buffalo’s investor network during a presentation at Seneca One Tower) represents exactly the kind of bridge function that the Startup Ecosystems framework identifies as essential; experienced operators who connect regional founders to the knowledge and networks of more developed capital markets, without requiring those founders to physically relocate to access them.

Siskar also launched the inaugural Founder Institute Buffalo cohort in 2019, graduating nine companies across fashion, electric vehicles, childcare, and more; a signal that the pre-seed education layer in Buffalo could be activated with the right facilitation.

Buffalo Startup Development Organizations: The Full Map

The Buffalo startup development landscape is substantially more populated than I think anyone realizes. Here is what I can find that matters.

43North remains the most visible accelerator, deploying $5 million annually in five $1 million investments to companies that commit to building in Buffalo, with office space at the historic Seneca One Tower and access to a network of over 1,000 North American investors. The 2026 Finals mark the program’s transition into its next phase of evolution.

Launch NY is the most operationally important organization for early-stage companies that are not yet ready for 43North. As the only nonprofit venture development organization and U.S. Treasury-designated Community Development Financial Institution (CDFI) serving Upstate New York, Launch NY has served nearly 1,750 companies since 2012 that have gone on to raise more than $1.5 billion and support more than 5,300 jobs. Launch NY secured $5 million in new funding from Empire State Development in February 2025 through the NY Ventures Community and Regional Partner Fund, supported by SSBCI capital, to continue expanding its mentorship and investment programs.

More than 70% of Launch NY’s portfolio companies are located in low-income neighborhoods, which makes it the most significant tool for equitable ecosystem development in the region.

UB Cultivator is the University at Buffalo’s pre-seed accelerator, running two phases over nine months. Phase 2 companies can receive up to $100,000 in SAFE financing through the Buffalo Innovation Seed Fund. To date, Cultivator has worked with 66 companies and funded 28 of them, with more than 70% of recipients identifying as women or founders of color. The UB Innovation Seed Fund writes both Cultivator SAFEs and follow-on investments up to $250,000.

Radial Ventures is the 43North Foundation’s venture studio, building companies from the idea stage with experienced operators as co-founders. As described above, this is the most operationally sophisticated early-stage intervention in the ecosystem.

The Western New York Incubator Network (WIN), managed by UB’s Business and Entrepreneur Partnerships office, coordinates the region’s physical incubation infrastructure, including the Incubator at CBLS (downtown, adjacent to the Buffalo Niagara Medical Campus) and the Incubator at Baird (near UB’s North Campus). WIN operates under the NYS Innovation Hot Spot program, offering tax incentive pathways to incubator clients.

The Western New York Venture Association (WNYVA) and its Buffalo Angels division form the primary organized angel investor infrastructure in the region. Buffalo Angels is an accredited-investor network that holds pre-screened pitch meetings and operates member-managed seed investment funds. As executive director Jack McGowan has noted, local angel and venture capital activity passed $200 million in 2017 and companies in the region raised just under $525 million in 2021, though most of that capital came from outside Buffalo.

Additional capital sources include Excell Partners (seed and early-stage VC focused on Upstate NY), Armory Square Ventures (seed-stage, Upstate NY tech focus), Impellent Ventures (Rochester-based, Northeastern city focus with Buffalo involvement), and Varia Ventures (growth-stage support). The WNY Impact Investment Fund represents a collaborative model pairing corporate, private, and philanthropic capital for social-return outcomes.

Buffalo Open Coffee Club (BootSector) remains the lowest-friction entry point for new founders; a community gathering that functions as the informal connective tissue between all of the more formal programs. BootSector itself is a 501(c)(3) founded in 2021 focused on empowering and educating the next generation of local startup leaders.

Endeavor WNY connects high-growth entrepreneurs to a global network of mentors and resources. Golisano Institute for Business & Entrepreneurship, founded by Tom Golisano (founder of Paychex) in 2023, is adding a second campus in downtown Buffalo in fall 2026 with a two-year Professional Certificate in Business & Entrepreneurship; a meaningful new pipeline for practical entrepreneurial education that bypasses the traditional MBA pathway.

New York State’s START-UP NY program provides 10-year tax-free operating environments for companies located in designated university zones, and several Buffalo startups leverage this in combination with accelerator or incubator membership.

Buffalo Through the Six-Part Startup Ecosystems Framework

Part One: The Lie of Innovation. Most ecosystems confuse activity with value, celebrate innovation theater rather than market-driven consequence, and deploy metrics that reward the wrong behavior. Buffalo is, to its credit, becoming self-aware about this. The ten-year anniversary coverage of 43North in a University at Buffalo analysis explicitly noted that room for growth exists in connecting the established business community more closely with the startup ecosystem, with having more founder exits, and an aggregated blueprint for how to start and build businesses in Buffalo, along with a critical need to develop and retain talent from local colleges and universities to fill jobs for growing startups. That self-critique (coming from within the ecosystem itself) is actually a healthy sign. Ecosystems that cannot self-critique cannot self-correct.

Where Buffalo still partially falls into the innovation theater trap is in the persistent tendency to celebrate the existence of startups rather than the outcomes startups produce. Startup Blink ranks Buffalo’s ecosystem as having grown 18.8% in 2025, with 138 startups and $197.5 million in total funding at a global rank of #200. That growth rate is genuinely impressive. The rank of #200 globally suggests the ecosystem is still early-stage relative to its potential. Celebrating double-digit growth without honestly reckoning of size is innovation theater.

Part Two: The Things We Refuse to Distinguish. Ecosystems fail when they conflate startups with small businesses, development capital with risk capital, job creation as a goal rather than a lagging indicator, and talent programs with talent attraction that comes from opportunity. Buffalo’s government investment is most exposed here. The political incentives of the state programs that fund 43North, Launch NY, and the WIN network all push toward job-creation metrics. Those incentives produce programs that look like they are serving startups but are actually optimizing for metrics that don’t tell you whether any scalable value is being created. The right metrics (founder exits that recirculate into the ecosystem as investors and angels; follow-on capital ratios from private sources; company survival rates at 5 and 10 years) are harder to report in a press release. The ecosystem’s leadership is aware of this; whether the state funders are equally clear-eyed is a different question.

The talent distinction is particularly sharp in Buffalo. As of 2025, Buffalo’s startup scene has experienced a 21% job growth over the last decade, with the 43North Foundation launching a $100 million initiative. That is a real outcome; but the talent retention problem (losing UB graduates to New York, Boston, and Toronto immediately after graduation) is not solved by building more programs. It is solved by building more companies that want to hire those graduates, which is precisely what Radial Ventures and the Lightning Strike initiative are designed to address from the supply side. The demand side requires corporate connectivity, and the 43North Foundation explicitly names that in Lightning Strike.

Part Three: Capital Is a Signal, Not a Gift. Perhaps the most important for Buffalo specifically, because the region’s startup capital structure is genuinely unusual thanks to a very large proportion of its investment activity involving public or quasi-public money, and the private capital market remains thin relative to the deal flow the ecosystem is generating. Most of the capital raised by Buffalo companies comes from outside the region, not from local investors. That is not a failure (outside capital validating local companies is a healthy sign) but it signals that the local LP base is not yet fully engaged in startup risk.

The venture capital co-investment pattern that 43North has developed is a more sophisticated approach to the capital signal problem than most regional ecosystems achieve. When a company that won 43North later raises from a top-tier coastal VC, that is not just capital; it is a signal about the quality of what Buffalo is building. That signal needs to be amplified, not buried in aggregate dollar totals.

Part Four: Ecosystems Fail for Structural Reasons, Not Moral Ones. Ecosystems fail because they are designed (often perfectly) to produce outcomes other than value creation; generic accelerators expand enrollment to please sponsors, public capital deploys for visibility rather than signal, and soft landing programs treat geography as the binding constraint when capability is the actual gap.

Buffalo has been guilty of all of these, at various points. The Z80 Labs incubator, which preceded the current structure, lacked a hands-on approach to building the products and providing a team that could help get an idea off the ground according to Magnuszewski, and it eventually ran out of funding without leaving a durable institution behind. The lesson was applied, and Radial Ventures specifically addresses the gap Z80 Labs left by bringing experienced builders into the company formation process rather than just providing office space and curriculum.

In Startup Ecosystems, I make an important point about public capital that applies directly to the Buffalo situation; it can function effectively when deployed with clarity about its comparative advantages, but those advantages are not in replicating what private capital does. Public capital’s comparative advantage is in building the conditions (talent pipelines, physical infrastructure, research capacity, procurement pathways) that make private capital deployment rational. The Lightning Strike framework’s separation of venture studio funding (Radial), talent development (TechBuffalo), corporate connectivity, and storytelling (Series B(uffalo)) suggests that the 43North Foundation has internalized this.

Part Five: Marketing Is the Missing Discipline. The most consistently undervalued capability in startup ecosystems is market discipline; understanding the market (not just customers) deeply enough to price products correctly, build demand validation into the founding process, and develop the narrative coherence that attracts capital and attention at scale. Buffalo’s industrial and manufacturing heritage is a genuine asset here, because those sectors demand real customer relationships, real unit economics, and real feedback loops. Founders building AI tools for electrical contractors (Makeoff) or wholesale vehicle auction platforms (ACV) cannot hide bad product-market fit behind venture capital theater; the customer either shows up or they don’t.

Where Buffalo needs deliberate strengthening in this dimension is in the health sciences and biotech sector, where the long development cycles and regulatory pathways create conditions in which founders can spend years without genuine market feedback, sustained by grant capital and university affiliation. The University at Buffalo’s research output in bioinformatics and life sciences is significant; the commercialization pathway from that research to customer-validated company remains, as in most university ecosystems, more complicated than the institution’s promotional materials suggest.

Part Six: What Actually Works. Conditions precede programs is a turn of phrase that communities need to embrace because far too many launch programs and hope they’ll figure it out. Capital formation is a policy architecture question rather than a funding pipeline question, that density and optionality compound over time, and that the ecosystem KPIs that matter are the ones that measure real outcomes rather than activity. David Colligan has observed that “a key indicator of a successful entrepreneurial system is whether the entrepreneurs who successfully launch one startup become serial entrepreneurs and launch additional startups thereafter,” and he notes that examples like George Chamoun (Synacor to ACV) and Eric Reich (Campus Labs, then 43North board chair) prove this is happening in Buffalo.

That serial entrepreneurship pattern (talent and capital recycling through the ecosystem rather than extracting and leaving) is the most important leading indicator of ecosystem health, and it is accelerating in Buffalo. Magnuszewski’s return to build Radial rather than decamping to San Francisco is the most visible example, but the pattern shows up elsewhere in the ecosystem’s alumni and investor network.

Where Buffalo Excels and Where It Needs Work

The Startup Ecosystems book’s ten dimensions of entrepreneurial capacity provide a direct diagnostic framework for any regional ecosystem. Applied to Buffalo, here is the assessment.

1. Ecosystem Builders Must Be Funded and Supported. Buffalo does reasonably well here by regional standards. The 43North Foundation’s Lightning Strike initiative is one of the most substantial commitments to sustained ecosystem building infrastructure that any non-coastal region has made. Launch NY, UB’s Business and Entrepreneur Partnerships office, and BootSector all represent funded ecosystem-building functions. The gap is in the informal layer; the event hosts, community managers, and connectors who aren’t attached to institutions and who burnout without sustained support. Series B(uffalo), the storytelling arm of Lightning Strike, addresses part of this by creating a funded narrative function. The informal connector community needs more direct investment.

2. Remove Silos. Buffalo has notable silo problems, and most of its ecosystem leaders will tell you so directly if you ask. The major institutions (43North, Launch NY, UB BEP, the WNYVA, Endeavor WNY) operate largely on parallel tracks with coordination that is episodic rather than structural. The VMG Business Advisory 2026 Founder’s Guide provides the most useful current map of the ecosystem precisely because no single institution makes that map available; a founder navigating the landscape needs external synthesis to understand what exists. Removing silos requires a shared operating framework with defined KPIs across all partners, which the Lightning Strike initiative is positioned to become; if it is willing to hold its partners accountable to shared outcomes rather than individual program metrics.

3. Fund the Ecosystem Actors. Partially addressed, as noted above. The institutional layer is better funded than it was five years ago. The informal layer is not.

4. Focus on Outcomes, Not Activity. This is where the critique lands hardest. Buffalo’s public reporting on its startup ecosystem leans heavily on aggregate numbers (total capital raised, total jobs created, total companies funded) rather than on the rate metrics that actually indicate whether compounding is happening. What is the survival rate of 43North companies at year 5? What is the ratio of follow-on private capital to public capital invested across the ecosystem? What is the rate at which founders who have had one exit are becoming investors? Those numbers exist in some form; they are not the numbers that appear in press releases, and they should be.

5. A Culture of Collaboration. Buffalo’s culture is genuinely collaborative by the standards of most regional ecosystems, partly because it is small enough that the relevant people all know each other, and partly because the shared experience of the city’s economic decline created a civic solidarity that translates into ecosystem cooperation. The Buffalo Open Coffee Club has operated continuously for years precisely because it is organic and unbranded enough to serve as neutral connective tissue. The risk here is that collaboration culture becomes mutual validation culture; everyone supporta everyone’s program without anyone being willing to say that some programs are working and some are not.

6. Include the Invisible Talent. The Golisano Institute’s arrival in downtown Buffalo (2026) will add another non-university-affiliated pathway into practical entrepreneurial education. Buffalo’s invisible talent gap is most significant in the manufacturing and industrial workforce; people with deep process knowledge and real customer relationships in industrial sectors who do not self-identify as “founders” and who are not being systematically reached by the ecosystem’s current discovery mechanisms.

7. Architect High-Performance Environments. The physical environment is strong. Seneca One Tower provides 43North companies with world-class space in a landmark building that has become a symbol of the ecosystem’s ambition. The CBLS incubator on the Buffalo Niagara Medical Campus provides wet lab and life sciences infrastructure that is genuinely competitive. Co-working options have expanded substantially over the past decade. The psychological safety dimension, whether Buffalo’s cultural memory of economic decline creates a risk-aversion that makes founders more conservative than they should be, is harder to quantify but worth watching.

8. Align Government, Academia, and Private Sector. This is the dimension where Buffalo has the most structural work remaining. New York State has been a genuinely important funder and program enabler, but state funding comes with political incentive structures that are not always aligned with what early-stage startups actually need. The START-UP NY program’s tax incentives are genuinely valuable but disproportionately benefit companies that have already achieved enough traction to locate in a university zone, rather than the earliest-stage founders who need the most help. The University at Buffalo’s research output is enormous and largely under-commercialized in the sense that there is no structural mechanism to convert research insights into customer-validated startups at the rate the research base would support. A shared strategic operating framework with defined cross-institutional KPIs remains the missing piece.

9. Unlock Local Competitiveness. This is where Buffalo has the clearest strategic opportunity and the most room for growth. The region has genuine comparative advantages in healthcare and life sciences, financial services technology, manufacturing and industrial operations, and clean energy (courtesy of ongoing Niagara hydroelectric infrastructure). The ecosystem’s programs are still too industry-agnostic, which means they provide generalist support to founders regardless of whether those founders are building in sectors where Buffalo has any real advantage. 43North’s industry-agnostic pitch competition format is politically defensible but strategically suboptimal; the most valuable companies in the portfolio tend to be the ones that leverage something specific about the Buffalo Niagara industrial base, not the ones that happen to have founders who moved to Buffalo for $1 million.

10. Adapt Global Best Practices, Don’t Copy Them. Buffalo has done reasonably well here. 43North is not a copy of Y Combinator; it is a different intervention designed for a different context. Launch NY is not a copy of any coastal accelerator; it is a CDFI-structured mentor-capital hybrid that reflects the specific capital access challenges of Upstate New York. Radial Ventures draws on global venture studio models but applies them to industrial sectors that no coastal studio is focused on. The risk going forward is in the temptation to benchmark against larger ecosystems in ways that import metrics and structures designed for contexts that do not apply.

What Buffalo Gets Right, and the Work That Remains

Buffalo gets the narrative right, at least within the ecosystem. The Tesla story, the ACV story, the Magnuszewski-returns-to-Buffalo story, these are authentic and compelling. Narrative discipline is a capital formation tool, not just a PR function, and Buffalo has real narrative material to work with.

Buffalo gets the institutional architecture approximately right, with some important gaps. The combination of 43North, Launch NY, UB Cultivator, Radial Ventures, and the Buffalo Angels creates a reasonably complete stack from idea through growth-stage capital. The gaps are at the very earliest stage (the pre-ideation founder discovery problem) and at the later stage (the Series A and B funding desert that most regional ecosystems face when companies outgrow local capital availability).

Buffalo’s startup ecosystem grew 18.8% in 2025, ranks #200 globally, with 138 startups and total startup funding over $197.5 million, and real numbers that represent real progress. The question the Startup Ecosystems framework asks about those numbers is whether they represent compounding or activity. Compounding means each dollar of investment produces more deal flow, more founder experience, and more recycled capital than the one before it. Activity means each program cycle needs the same external inputs to produce the same outputs, indefinitely, without self-reinforcement.

The trajectory of the ecosystem (from a state-funded pitch competition to a venture studio to a $100 million foundation commitment with four distinct investment pillars) suggests compounding is beginning. ACV’s IPO recycled talent and credibility back into the ecosystem in the form of Magnuszewski’s commitment, and that commitment is now the most sophisticated company-building infrastructure in the region. That is the pattern the book describes as structural health: outcomes that create the conditions for better outcomes.

The work that remains is in three areas where the ecosystem’s current momentum has not yet fully addressed the structural gaps. Private capital density needs to grow; too much of the ecosystem’s capital still flows from public sources, and the private angel and VC infrastructure, while improving, is not yet deep enough to fund the companies that are ready for it. Sector focus needs to sharpen; the industrial, healthcare, and financial services strengths of the region are underutilized by a program ecosystem that is still largely industry-agnostic. And talent retention needs a demand-side solution; more companies that want to hire UB graduates in Buffalo, not just more programs trying to convince graduates to stay.

Tesla’s 1897 prediction that Buffalo was “sure to become one of the greatest industrial centers of the globe” did not age perfectly. But the prediction was premised on something real: the convergence of energy, infrastructure, intellectual ambition, and civic will that the Niagara Frontier genuinely possessed. Those assets are still here, in different forms. The energy is renewable power and hydroelectric capacity. The infrastructure is a repurposed industrial base, affordable and available. The intellectual ambition is University at Buffalo research, 43North alumni networks, and Magnuszewski’s venture studio. The civic will is evident in the people who stayed, who came back, and who are betting on it again.

The question is whether the ecosystem’s design is sophisticated enough to convert those assets into compounding outcomes, or whether it will optimize for the metrics that look good in press releases from Empire State Development. Based on the evidence of the past several years, the answer is trending in the right direction. The work isn’t done but then, it never is.

You interviewed 50 customers before you wrote a line of code. You ran surveys, you sat through awkward Zoom calls with strangers, you built a whole deck of assumptions and then systematically tried to prove yourself wrong, because someone, somewhere, told you that’s what disciplined founders do. Then you opened LinkedIn, found 200 venture capitalists whose names are unfamiliar, and blasted them all the same cold email asking for money.

That is not a fundraising strategy. That is the investor equivalent of walking up to strangers at a bar and proposing marriage, and then being confused when nobody says yes.

Every founder who has been through an accelerator, read a startup blog, or watched a YouTube video about Lean Startup can recite Customer Discovery and Customer Validation from memory. built an entire methodology, and an entire industry, on the premise that founders don’t actually know what customers want until they get outside the building and ask. As he puts it, “Customer Development is designed to solve the problems of the Product Development model,” and it works precisely because it forces founders to test hypotheses against reality instead of against their own optimism. It is, in Blank’s words, an iterative process, “the customer development model is a circular track with recursive arrows,” not a straight line from idea to launch.

So why does that discipline evaporate the moment the subject changes from customers to capital?

Customer Discovery Was Never Actually About Customers

Before we get into fundraising, let’s set a record straight since most of you suck as customer discovery too. The part everyone conveniently forgets; Customer Discovery and Customer Validation were never purely about the customer, they were about the market, and the market is not a synonym for “the people who buy your product.” The market includes the competitors already serving that customer badly, the partners who control distribution, the substitutes your customer is currently duct-taping together, and the broader trends that determine whether this is a five-year opportunity or a five-week fad.

Blank’s own model is explicit that founders need to be the ones getting outside the building to validate all the initial hypotheses of the business model and product, not just the sliver of the business model that says, “will someone buy this.” The business model includes how you win, who you’re up against, what it will cost, and who else has to say yes before a dollar changes hands.

Investors are part of that market. They are not a separate species you court after the “real” work of product-market fit is finished. They are, functionally, a customer segment with an unusual product; instead of buying your software, they are buying equity in the probability that your understanding of the market is correct. If you have not done Investor Discovery with the same rigor you applied to Customer Discovery, you have not actually finished validating your market. You have validated half of it and declared victory, which is a little like training for a marathon by only running the first thirteen miles and being surprised when you collapse at mile 14 because you thought it would be the easy part.

I’ve made this point before when I broke down why most startup pitches are bad; investors aren’t your customers, their LPs are, and a pitch that reads like a sales page for your product is a pitch that misunderstands who it’s actually talking to. The same failure of imagination that produces a bad pitch is the failure that skips Investor Discovery altogether. You cannot design a message for an audience you have never interviewed, and yet founders write cold investor emails as though the recipient’s preferences, thesis, check size, and decision process are irrelevant details.

The Cold Call You Swore You’d Never Make

Ask a founder if they’d build a product for a customer they’ve never spoken to, based entirely on guesses about what that customer wants, and they’ll tell you that’s insane; that’s exactly the mistake Customer Discovery exists to prevent. Then watch that same founder mass-email 300 venture capital firms with an identical pitch deck attached as a 14MB, 50 slide, PDF, no personalization, no understanding of stage or sector fit, just pure hope wearing a business-casual outfit.

I’m not kidding, some of you actually create 50 slide investor pitch decks. What the hell are you thinking!?

Warm introductions increase the likelihood of getting funded by roughly thirteen times compared with a cold email (and yes, I did write in my book, Startup Ecosystems, that warm introductions are actually a sign of a problem, but that’s a different topic – explained in the book); the pertinent point is that analyses of outreach campaigns put well-targeted, personalized cold email in the single-to-low-double-digit reply range while warm approaches routinely clear 20 to 34 percent. Founders cold-emailing investors makes the stakes plain when it warns against wasting time on funds that don’t invest in your type of business, because a proptech firm isn’t going to spend too much time on a beauty tech startup. That’s not investor snobbery; that’s the same principle behind not pitching a vegan protein bar to a customer who told you, in your Customer Discovery interviews, that they eat keto.

You wouldn’t waste the call. Stop wasting the email.

Investors, like everyone with limited time and a full inbox, use pattern recognition to survive the volume.

If your outreach doesn’t demonstrate that you’ve done the homework, understood their thesis, and know why this particular fund and this particular partner should care, you look like noise, and noise gets deleted in five seconds. This is not a mystery; it is the exact same behavior your prospective customers exhibit when a generic sales email lands in their inbox, and you already know how that story ends.

What Investor Discovery Actually Looks Like

“Discovery is just the work that we’re doing to decide what to build.” – Author, Teresa Torres; with Business of Software

has an exceptional book I want you to read, Continuous Discovery Habits: Discover Products that Create Customer Value and Business Value, and yes, I took some creative license with the book cover in that image on the article (that was me trying to make you think! Her book is here).

Customer Discovery is not a sales call. It is a listening exercise designed to surface facts you didn’t have, run through hypotheses you were prepared to be wrong about. Investor Discovery works the same way, and it requires you to walk in without an ask.

Start by building a real target list, not a scraped spreadsheet of “anyone with the word Capital in their LinkedIn title.” Segment by stage, check size, sector focus, and geography, the same way you segmented customers by need and willingness to pay. Use LinkedIn the way you’d use it for account-based sales; look at who a target investor has actually funded recently, what they’ve written or said publicly, and where the overlap is with your category. Use a CRM built for this, not a spreadsheet with fifteen broken formulas. I’ve pointed founders toward for exactly this reason in the past when I mapped out the funding flywheel; fundraising is a pipeline with stages, contacts, and follow-ups, and treating it like a to-do list instead of a CRM-managed relationship is why most founders lose track of who they’ve talked to and what was said.

Then go have conversations that are not pitches!

Ask investors what they’ve seen work and fail in your category. Ask what actually gets a deal to a partner meeting versus what gets a polite pass. Ask how they think about your specific market’s risks. You are not asking for money in this conversation; you are doing reconnaissance, the exact posture Blank prescribes when he insists that in a startup no facts exist inside the building, only opinions. Replace “inside the building” with “inside your own assumptions about what investors want” and the lesson transfers perfectly.

This is Investor Discovery; learning what a specific set of investors need to believe before they’ll act, in their own words, before you ever send a deck. It is uncomfortable in the same way customer interviews are uncomfortable, because you have to sit with answers you didn’t want, and you have to resist the urge to sell.

Investor Validation Is the Part Everyone Skips

Once you know what investors need to believe, Investor Validation is the process of proving it, systematically, the same way Customer Validation proves that people will actually pay rather than just say nice things in an interview.

This means building the specific traction data points a given class of investor has told you they weight most heavily, whether that’s retention curves, CAC-to-LTV ratios, letters of intent, or a pilot customer’s willingness to expand. It means testing your pitch against a handful of investors you don’t expect to close, refining based on the actual objections you hear rather than the objections you imagined, exactly the way you’d run a beta before a full launch.

It also means Investor Relations starting long before there is a check to justify it. Keep the investors who passed, the ones who said “too early,” and the ones who expressed even mild interest on a simple update cadence; a monthly or quarterly note with real numbers, real wins, and real setbacks. I’ve written before about how why founders pitch investors even when they don’t want VC comes down to discipline, not dilution; the same logic applies here. Consistent, honest updates turn a “no” into a warm relationship that becomes the very warm introduction worth thirteen cold emails. This is not a nice-to-have. It is the compounding mechanism that turns Investor Discovery from a one-time research project into a pipeline that gets easier every quarter, which is the entire premise behind what I called the funding flywheel.

It Was Never Customers vs. Investors. It Was Always the Market.

Step back and the distinction between Customer Discovery and Investor Discovery starts to look artificial, because it is.

Customers and investors are both trying to answer a version of the same question; does this founder understand a real, growing, defensible opportunity better than the alternatives available to me? The customer’s alternative is a competing product or doing nothing. The investor’s alternative is a competing deal or keeping the capital in reserve. Both are market participants evaluating your understanding of the same battlefield; the competition circling the same customer, the partners who control access to that customer, and the trends determining whether the window is opening or closing.

That’s why founders who treat Customer Discovery and Investor Discovery as two unrelated exercises keep producing pitch decks that read like product brochures instead of the 2-slide, market-and-team argument that actually gets a second meeting, and why so many pitch decks fail before an investor ever says no. The market research you did to build the product is the same market research an investor needs to underwrite the bet; if you segmented it in your head as “customer stuff” and “investor stuff,” you did twice the work to get half the insight.

Startups, Stop Cold-Calling People You Haven’t Interviewed

If you would never ship a product without talking to the people who are supposed to buy it, stop fundraising without talking to the people who are supposed to fund it.

Build the list, do the reconnaissance, listen more than you pitch, and treat every “no” as data instead of rejection, the exact posture that makes Customer Discovery useful in the first place instead of just an exercise founders perform because a mentor told them to.

Before you send the next cold email to a fund you found by searching “top VCs 2026,” ask yourself if you have interviewed a single person at this firm the way you interviewed your first ten customers. At this point, I shouldn’t have to reiterate what you can expect if your answer to your question is no.

Quick primer before the outrage: OMB stands for the Office of Management and Budget, the White House office that writes the rulebook every federal agency has to follow when handing out grant money. You’ve never had to know that acronym before because OMB’s job has historically been boring bureaucratic plumbing. That’s about to change, and the reason it should matter to you, whether you run an accelerator, sit on an economic development board, or manage a venture fund that partners with SBIR-funded startups, is that OMB just proposed giving itself the legal right to cancel any federal grant in the country whenever it decides to, with no requirement that the grantee did anything wrong. Comments close July 13, 2026. After that, your only recourse is maybe a lawsuit, and lawsuits are slow, expensive, and nowhere near as effective as showing up now.

Now, before I illuminate, I want OMB to know that I don’t want to be on their bad side because we REALLY need you to change the rules about grants. You hand out money to startup programs without regard for performance or outcomes; you mix Research, Small Business, and Startup funds as though they’re the same things; and you seemingly throw money around based on the merit of a submitted proposal, without the kind of due diligence investors rightly perform; and look, we love the support but it needs to be allocated to infrastructure, startup education, and curriculum, not accelerators and startups.

In May, OMB published a rewrite of the Uniform Guidance, the regulation at 2 CFR Part 200 that functions as the operating manual for how virtually every federal grant and cooperative agreement in America gets awarded and managed. As with all legislative issues, you don’t have to remember CFR Part 200 and frankly you don’t care, I’m including that in the article for the lobbyists and lawmakers who need that for clarity. The rewrite runs more than 400 pages (because, of course it does) and would touch nearly every organization receiving federal financial assistance, including nonprofits, universities, healthcare entities, and pass-through organizations administering awards for someone else. You can read the full text and submit a comment at Regulations.gov under Docket OMB-2026-0034.

Four issues in this proposal matter specifically to entrepreneurial ecosystem builders, and I want to walk through each one, tell you what I actually think, and tell you what to do about it.

Executive Discretion Beyond What Congress Authorized

Right now, every federal agency writes its own grant rules into its own regulations, which requires independent public rulemaking and gives each agency some room to deviate based on its own mission. The proposed rule kills that.

Once an agency adopts OMB’s new framework, it generally cannot modify the rules on its own, and because close to 40 federal agencies are adopting this simultaneously, OMB is centralizing government-wide control over how every grant in the country gets decided into a single office. Picture a 40-location restaurant franchise where corporate suddenly writes every menu, every price, and every hour of operation for every location, and corporate changes its entire philosophy every four years depending on who wins an election. That’s not oversight; this is a hostile takeover of authority Congress originally spread across dozens of agencies for a reason, mainly so no single administration could weaponize the whole grant system at once.

The Termination Clause; Like, a Landlord’s Eviction Right Written into Federal Law

Every founder who has negotiated a term sheet knows what a material adverse change clause is; the provision that lets an investor walk if something significant breaks before closing. A well-negotiated MAC (material adverse change) clause is narrow and objective, because a vague one is a weapon any investor can swing whenever they get cold feet. OMB just wrote the vaguest possible MAC clause into every federal grant in America and handed the trigger to whichever political appointee is currently in the building. Hopefully you appreciate that that means it is a trigger determined by whichever political party is in office, and I don’t care which side of the aisle you find yourself upon, what’s good for you now (in this case) is bad for you next time.

Under proposed Section 200.340, an agency can terminate a grant whenever doing so is “in the interest of the Federal agency,” including if the award no longer serves “program goals, Federal agency priorities, or the national interest as they exist at the time of the termination.” Like a “termination for convenience,” a doctrine built for federal procurement contracts on things like trucks and office supplies, this is not for a three-year accelerator commitment or a workforce program’s staffing plan.

Imagine signing a three-year office lease and discovering, on page 380, a clause letting the landlord evict you the moment your décor stops matching his mood, with zero requirement that you violated a single term. That’s the provision. No wrongdoing required, no cure period guaranteed, just a gut check with legal force.

The Political Loyalty Test Before Money Even Moves

A second provision makes my point that we should always be concerned when oversight is at the whim of a party. The rule requires pre-issuance review of every discretionary award by senior political appointees, who must confirm the award “demonstrably advance[s] the President’s policy priorities” before it’s issued, while peer review, meaning evaluation by independent subject-matter experts on technical merit, is preserved only as advisory input the appointees are explicitly allowed to override. Take a look at section 200.205. So, a grant for a regional biosciences workforce partnership or a defense-tech venture studio needs to clear a loyalty test before anyone even scores whether the program produces results… and if it doesn’t tow the White House line, you’re likely not getting funding.

OMB’s own preamble, published directly in the Federal Register, states that taxpayer dollars must not be “wasted to promote divisive doctrines of the far left.” Which, granted, itself isn’t neutral fiscal stewardship language but it says what it says; it’s an ideological filter written directly into the regulation that decides which regional programs get funded next year, and which don’t.

Higher Risk Discourages Organizations From Even Trying

The last problem is just math, not politics.

Two of the new rules make it a lot riskier to accept government grant money; one lets political appointees reject your application before anyone even checks if your program is good, and another lets the government cancel your grant later for almost any reason, even if you didn’t do anything wrong. Imagine a video game where the prize for winning stays exactly the same, but suddenly you can lose all your points at any moment for no clear reason; fewer people will want to keep playing that game. That’s what happens here.

The grant money doesn’t get any bigger to make up for the new risk, so smaller organizations, the ones already running programs with barely any staff or budget, are the first ones who decide it’s not worth the gamble anymore.

Federal Grant Overreach Is Coming for Entrepreneur Funding; Here’s Where to File, Call, and Push Back

Opinions are cheap (trust me, I have a lot of them) and Washington has learned to ignore them; specific, sourced pressure applied through the actual channels that count is not. If you run or fund anything touching a startup ecosystem, here’s where your effort goes.

File a comment, and file it correctly. Go to Regulations.gov and search Docket OMB-2026-0034. Lead every comment with the specific section number in brackets, meaning [200.340] or [200.205], since agencies are required to respond to substantive comments but lets them lump together anything that reads like a form letter; a comment that cites a section and describes a specific, real consequence to your organization gets weighed differently than a paragraph of general outrage. If your organization has never taken federal money directly, file anyway on behalf of the partner organizations in your ecosystem that have; you aren’t required to have been a grant recipient, just an interested party, which every regional connector, scout, and economic development office in the country technically is.

Call your representative and senators, not their press or communications staff, call their appropriations or oversight staff, since those are the people who actually track rulemaking dockets tied to executive overreach fights. Tell them specifically which section threatens which program in their district; a staffer hearing “Section 200.340 could zero out our regional biotech workforce partnership before the next fiscal year” is more useful to them, and more likely to get escalated, than a general complaint about federal bureaucracy.

Call a reporter, and call the right one. National outlets aren’t going to parse a 400-page rulemaking before July 13; your local business journal, your regional trade press, or an ecosystem builder with a newsletter will, because obscure rulemakings that quietly gut regional programs are exactly the kind of story that gets a reporter a byline nobody else has. Pitch them the specific local stakes, not the national abstraction.

Write about it publicly, under your own name, tied to your own ecosystem’s specific exposure. A LinkedIn post that says “this affects nonprofits” gets scrolled past. A post that says “our region’s SBIR-adjacent workforce grant could be canceled under Section 200.340 with zero notice, here’s what that means for the accelerator you already know” gets shared, because it converts an abstraction into a threat to something the reader recognizes.

Build a coalition before you file, not after. A single comment from your organization is a data point; five regional partners citing the same sections with five different concrete examples is a pattern OMB’s own preamble obligates them to address. Coordinate with your state’s economic development association, your regional accelerator network, or whoever else in your ecosystem has skin in this game, and file separately, on the same day, citing the same sections, before July 13.

None of this fixes itself by July 14. The rule doesn’t need everyone in the ecosystem to respond, it just needs enough people to stay quiet, and quiet is the default outcome unless someone forces it not to be. This is how this works friends.

Every regional accelerator, workforce program, and economic development office reading this has thirty seconds of actual work standing between them and a filed comment; the excuse that it’s someone else’s fight is the same excuse that got startup programs into a position where a 400-page rule change nobody read is about to decide who gets funded next year.

Every founder has been told to “validate their idea with customers.” Every accelerator puts it in the curriculum. Every startup blog has a listicle about it. Nearly all of them are teaching you the wrong thing, framing it incorrectly, and in doing so, setting up first-time founders to confuse the motion of talking to people with the actual work of figuring out whether a startup belongs in a market.

Customer validation and market validation are not the same thing. Conflating them is the kind of mistake that gets you killed slowly; you think you’re making progress because you’re having conversations, but the conversations you’re having are answering the wrong question.

Customer validation, as a formal process, is what Steve Blank outlined in The Four Steps to the Epiphany as step two in the customer development process, following customer discovery. It is a “test sell” process, a series of “quantitative pass/fail tests” designed to confirm that a product you’ve already hypothesized can actually be sold. Customer validation answers the question: do real people, in real purchasing situations, actually want to buy this specific thing I’ve built or described? It presupposes that you have done the prior work of understanding the market and identifying the problem.

It presupposes that the opportunity has already been validated and that what remains to be confirmed is transaction behavior.

Most founders skip everything that comes before it and start asking potential customers if their idea is good. That is not validation; That is therapy. You are asking strangers to make you feel better about a decision you’ve already made, and strangers, who have no stake in your success and no incentive to be honest, are going to tell you it sounds interesting. “Sounds interesting” is the most useless signal in the history of entrepreneurship.

What You’re Actually Supposed to Be Doing: Market Validation

Market validation is not a subset of customer validation; it is the broader, prior, and more rigorous discipline, asking whether a market opportunity exists before it asks whether your specific product can be sold. Research consistently shows that capital running out is where startup stories end, but the root causes upstream of that are poor product-market fit (43%), bad timing (29%), and unsustainable unit economics (19%).

Running out of money is the mechanism of death; building something nobody wants is why death was coming.

Of 431 failed VC-backed companies, 43% failed due to poor product-market fit, while running out of capital affected 70% of failures; running out of capital is a final symptom, not the root cause. You are not doing something novel and tragic when your startup dies because nobody wanted it; you are participating in the statistical norm. You are one of the 43%.

Market validation is not about confirming that your idea is good; it is about understanding the conditions of the market you intend to enter well enough to know whether you have a real role to play in it.

This means studying the competitive landscape, not to find “whitespace” you can pitch to investors, but to understand what solutions customers are already paying for and what those solutions reveal about how the problem is prioritized. It means evaluating trends, because timing is 29% of startup failure and building for where a market was three years ago is how you end up with technically competent products that nobody buys. It means talking to potential partners, because distribution is frequently the thing that kills a startup that built a genuinely good product. It means engaging investors early, not to pitch them, but to understand what signals they’re looking for and whether the category is investable at all. And yes, it means talking to potential customers, but as part of a broader intelligence-gathering operation, not as the entirety of your validation process.

As has argued, the Lean Startup methodology asks innovators to interview potential customers within their “market” to discover unmet needs but skips the market definition step entirely; the consequence is that innovators define markets around products, technologies, or demographics, and then pivot the market they’re targeting while simultaneously trying to establish product-market fit. That is the founding confusion.

You cannot validate a market you haven’t defined. And you cannot define a market by asking your potential customers to define it for you.

The Real Goal Isn’t Validation. It’s Founder/Startup Fit.

Here is the part that almost nobody teaches and that separates experienced founders from first timers…

If you have relevant domain experience, you should not need to validate an idea the way a beginner does. You already know the market; you’ve been in it. You’ve watched it fail people, watched incumbents ignore obvious problems, watched capital flow toward suboptimal solutions because the right founders hadn’t shown up yet. You are not discovering the opportunity; you are confirming what you already have reason to believe and then determining whether you are the right person to execute it.

This is Founder/Startup Fit, and it is the actual question market validation is designed to answer. Not “is this a good idea?” but “is this the right opportunity for me, and do I have the specific capabilities that this market gap requires?”

, citing of Benchmark Capital, framed it that, “When a great team meets a lousy market, market wins. When a lousy team meets a great market, market wins. When a great team meets a great market, something special happens.” The implication that most founders miss is that the market is the dominant variable, and the team’s job is to figure out whether they are the right team for the specific market they’re entering, not whether the market exists. The market exists or it doesn’t, independent of your enthusiasm for it.

Founder/Startup Fit is the question of whether your specific background, network, capabilities, and credibility give you an advantage in this particular market at this particular moment. A healthcare founder who spent a decade running clinical trials has a different position in that market than an engineer who read some PubMed abstracts and decided healthcare is interesting. Both might build a product; only one of them is doing market validation from a position of earned insight, the other is doing what looks like research but is actually rationalization.

The frameworks and work that goes into figuring this out are laid out in detail here, and the short version is this: real validation comes from competence, market insight, and testing assumptions, not from collecting reassuring responses to leading questions. The accelerator-standard advice to “go talk to customers” is not wrong; it is incomplete to the point of being misleading.

The Five-Part Market Validation Process That Actually Works

Assessing competition is the first move, not an afterthought for the competitive landscape slide in your deck.

Who is already solving this problem?

Not just the direct competitors, but the adjacent solutions, the entrenched behaviors, the workflows people have jury-rigged out of spreadsheets and email threads because nobody built the right tool. The presence of competition is not a threat to your startup; it is confirmation that the market exists and that people are willing to pay for solutions to the problem.

The absence of obvious competition should make you nervous, not confident.

It usually means either that the market is smaller than you think, that the problem is not painful enough to generate economic demand, or that you don’t understand the space well enough to know who you’re competing with.

Studying trends means understanding why now is the right time for this solution to exist, and whether the conditions creating the opportunity are accelerating or decelerating.

As Andreessen put it in his foundational essay on product-market fit, “You can always feel when product-market fit isn’t happening. The customers aren’t quite getting value out of the product, word of mouth isn’t spreading, usage isn’t growing that fast, press reviews are kind of ‘blah,’ the sales cycle takes too long, and lots of deals never close.” What he’s describing is the output of a timing problem as much as a product problem. The founders who entered that market too early did the same work as the founders who entered it at the right time and built successful companies; the variable was when, not what.

Evaluating potential partners is overlooked almost universally.

Go-to-market strategy determines whether a good product reaches the customers who need it, and distribution frequently already belongs to companies who are not you. A channel partner, a platform ecosystem, a reseller network, an integration partner with an existing customer base in your target segment; these are not nice-to-haves you figure out after product-market fit, they are part of what determines whether product-market fit is achievable at all, given your resources and timeline.

Talking to investors before you have a pitch-ready product is something most founders avoid because it feels premature.

It is not.

Investors see hundreds of startups in your category and have a reasonably accurate read on what traction looks like, what the exit dynamics are, whether the category is fundable, and what comparable companies have achieved. An early conversation with an investor who has portfolio exposure to your space is more useful market intelligence than twenty customer interviews, because investors have already aggregated the signals from comparable situations and can tell you whether the dynamics, you’re observing match what they’ve seen work or fail.

And then, yes, you talk to potential customers, but with specific questions designed to reveal purchasing behavior, not opinion.

The distinction between “Would you use this?” and “Pay $X/month for this, now,” generates completely different answers; startups that survive get to willingness-to-pay data early, through pre-sales, deposits, or structured panel research, rather than treating enthusiasm as a signal of purchase intent.

Why Founders Keep Getting This Wrong

The customer validation framework became the default teaching in startup culture because it was operationalizable; you could put it in a slide, assign homework, and run a workshop around it.

“Go talk to 20 customers and come back with findings” is a tractable exercise. “Go develop genuine domain expertise, build a network, study your market’s structural dynamics, and then determine whether you are the right founder for this opportunity” is a semester-long process that doesn’t fit between Tuesday’s cohort session and Thursday’s demo prep.

Accelerators need frameworks that work at scale, and customer interviews are teachable in 90 minutes. The result is a generation of founders who have been trained to perform the motion of validation without doing the work of understanding their market. They talk to people, they collect quotes, they build decks with “customer insights” slides, and they convince themselves that enthusiastic responses from people who will never pay for the product constitute evidence of market demand.

They don’t.

All the tweets and LinkedIn posts say that it’s all about customers, need to stop, and everyone who celebrates them, need to be forced a reality check.

Talking to customers requires that they are unbiased; disregarding the full picture of market validation is why 90% of startups fail, a ridiculously bad rate of performance we should all be ashamed of, driven entirely by founders pushing to do something that clearly isn’t going to work.

Stefan Sarbu of Thinslices’ warned, “Every startup begins with a hypothesis, but you haven’t validated anything until someone outside your team cares enough to engage, sign up, pay, or even just respond.”

The standard customer interview does not produce that evidence; a letter of intent from a potential customer does, a pre-order does, and a pilot agreement with a paying organization does. Someone sitting in a coffee shop telling you that your app sounds cool is the founding story you tell at the launch party, not evidence of market demand.

The Signal Investors Are Actually Looking For

Experienced investors do not get excited about customer validation data. When I sit back and think of the number of times I’ve ever in my 30 years with startups, EVER valued a founder saying they had some customers who told them this was a good idea, I can’t even put a 1 on the board.

Investors get excited about market validation data, and the difference is evident the moment you start talking. Customer validation data looks like, “We talked to 47 potential customers and 38 of them said this is a problem they experience.” I don’t care. Market validation data looks like, “The market is currently spending $2.3 billion on workarounds that partially address this problem, the three dominant players haven’t changed their offering meaningfully in five years, we have three signed letters of intent from companies in the target segment at the price point we’ve modeled, and the regulatory trend in this space over the next 24 months is moving in our direction.”

One of those is a founder who has done homework; the other is a founder who has collected opinions.

NFX, the venture firm, makes the argument that, “the smartest startups earn the right to build” before building, by testing ideas against the market first; not only does this save time and money, but it accelerates and strengthens product-market fit discovery because founders are operating from a position of genuine market knowledge rather than validated enthusiasm.

Stop asking if you have validated this with customers. Instead, start now, “have I validated this market well enough to know that I am the right person to enter it at this moment, with the resources I have, and a realistic path to the traction that would justify continued investment of time and capital?”

That question is significantly harder to answer, which is precisely why most founders avoid it in favor of talking to people who tell them what they want to hear.

If you’re sitting on a startup idea right now and you’ve been told to validate it with customer interviews, ask yourself what the competition looks like, what the trend line in the market is doing, which partners control distribution, and whether the investors who fund this space see the signals you’re generating as evidence of real demand. If you can’t answer those questions, you haven’t done market validation yet; you’ve just been having conversations, and conversations are not a company.

Your city’s startup ecosystem is probably an elaborate performance; there is a co-working space with exposed brick, there is a demo day with a banner, there is a grant attached to metrics no one is held accountable for, and somewhere in a municipal budget, there is a line item called “innovation” that mostly buys PowerPoint presentations for city council. None of it is working the way you think it is, and there is a good chance it is actively crowding out the conditions that would actually work.

This is the conversation , Ayhan K. Isaacs, and I had during a recent Founder Institute on economic development and entrepreneurial community building. Kevin has invested in over 160 early-stage companies, built the Founder Institute’s New York City chapter into one of its most productive globally, and now runs Finta, a platform that automates capital-raising workflows across private markets. Between us, we have sat in enough ecosystem meetings, grant reviews, demo days, and mayor’s offices to have a fairly precise diagnosis of why most of this work doesn’t compound.

It looks like my ISP wanted to make me blurry on camera so while you can tune in here, let me also share what we covered, and what you should probably do with it.

Your City Is Not the Right Unit of Analysis

This is the first place most ecosystem conversations go wrong; Silicon Valley is not a city, it is a geography that spans San Jose, Santa Clara, Sunnyvale, Mountain View, Palo Alto, Menlo Park, and a dozen surrounding communities. Interestingly, it doesn’t technically include San Francisco and yet people throughout the world think they should go to San Francisco to raise money in Silicon Valley. These confusions are why I talk about this; most of you think locally when you need to think regionally. The density here emerged from that regional scale, not from any single municipality deciding to become a startup hub. I have said this more times than I can count, Silicon Valley is four cities, people. The entrepreneurial communities that actually compound are regional in their infrastructure, even when they are local in their identity.

The reality, so stop pretending otherwise, is that a single city cannot generate sufficient deal flow, enough angels, enough mentors with genuine startup experience, or enough optionality for founders to stay when their first company doesn’t work out. The Kauffman Foundation’s work on ecosystem momentum has consistently found that ecosystem performance scales with the density of the connections between participants, and density requires mass that most individual cities simply do not have on their own. You need the state, the region, or at minimum a multi-city coalition meaningfully involved, not in a ceremonial way with shared letterhead, but in a structural way with shared infrastructure, shared capital formation expectations, and shared accountability.

The way to operationalize this is to tie public funding to outcomes; if a state is writing a seven-figure check to accelerators and incubators, it should require outcome-based metrics in return, not how many events were hosted, not how many founders attended a cohort, but downstream results such as follow-on capital raised, companies still operating at year two, and, over the longer arc, jobs created in sectors that didn’t already dominate the local economy. This is not a radical idea, it is the logic behind any capital allocation that is supposed to function as investment rather than subsidy.

Measuring the Wrong Things Is Worse Than Measuring Nothing

The metric problem in ecosystem development is so entrenched that it has its own self-sustaining political logic; that demo days create events, events attract press, press attracts politicians, politicians want credit, credit requires visible activity. So, the ecosystem optimizes for visible activity, produces copious amounts of it, and ends up measuring itself against outputs that have virtually no relationship to whether founders are actually thriving.

The question that should anchor every ecosystem assessment is a brutally simple, “Are we producing companies that are getting funded, growing, and creating durable employment?” And the answer in the majority of cities is no, or not at a rate that justifies the public investment being made. Guidelines for government entrepreneurship promotion found that despite decades of attention to startup development, actual new firm formation rates in the U.S. dropped from roughly 15 percent of all businesses in the late 1970s to around 8 percent by the early 2010s. More programs, more events, worse outcomes; that is not a coincidence.

Kevin made a specific point here that the people writing grants to ecosystems care about jobs, ecosystems start optimizing for jobs. So, a startup that is still figuring out its business model, maybe twelve months in, starts hiring to hit a headcount number because the program funding them requires it. That hire, made before the company has product-market fit, is often the wrong hire made for the wrong reason, and it frequently costs the company the next twenty hires it would have made under better conditions. Measuring job creation at the wrong stage of a company’s development is not just unhelpful; it is actively destructive to the companies being measured.

The metrics that indicate a healthy ecosystem are lagging, patient, and structural; capital formation per capita, the ratio of startups that survive to year three compared to regional averages, the number of new angels who became angels because a local exit generated both the capital and the conviction to reinvest locally, and the rate at which that capital stays in the ecosystem rather than flowing to wherever the acquirer is headquartered. These are not the metrics that look good in a grant report submitted six months after a program cohort. They are the metrics that actually tell you whether anything meaningful is happening.

Scarcity Breeds Silos, and Silos Kill Ecosystems

The dynamics of emerging ecosystems produce a dysfunction that when capital is scarce, the organizations managing it compete rather than collaborate.

• The accelerator doesn’t share its mentor network with the angel group.

• The university tech transfer office guards its deal flow.

• The economic development agency funds programs that replicate rather than integrate with what already exists.

Everyone is protecting their slice of a very small pie, and the net effect is a community of organizations doing parallel work with the founders caught between them navigating a maze where every door opens onto a different map.

Kevin described observing this in ecosystems globally; in some cases, participants in the same city would actively undermine each other’s funding relationships to preserve competitive advantage. When there is not enough capital to go around, everyone behaves rationally in their own interest in ways that collectively make the situation worse. The fix is not a plea for collaboration at the annual summit; it is shared infrastructure that removes the competitive advantage of hoarding. A common CRM, a shared mentor pool, integrated communication channels, a single directory of investors that every organization in the ecosystem is obligated to contribute to and point founders toward; these are not glamorous interventions, but they systematically dismantle the silo logic by making collaboration the path of least resistance.

Kevin also noted a phenomenon that I have observed in every ecosystem that has made the leap from fragile to self-sustaining; there is a tipping point where the scarcity logic flips. In New York City, it happened incrementally as Google, Twitter, and then Facebook opened engineering offices and began employing enough people that the talent pool thickened dramatically. The math changed; suddenly there were enough engineers that losing one to a startup was not catastrophic for your company, and enough that a founder who lost a company could get a job while building the next one. That optionality shift changed the psychology of the whole ecosystem from protection to participation. Capital started moving toward sharing deal flow rather than guarding it and reputation started to matter more than control.

Your Ecosystem Is a Secret and That Is Your Fault

There is a communication failure in almost every startup ecosystem outside of the top ten cities, and the failure is peculiar because ecosystems are built by people who are supposedly in the communication business. The networking events work, the community-building happens, and the founders are convening and the investors are showing up and the panels are full, and then absolutely nobody outside of a fifty-mile radius hears a word about it.

The reason Austin and Silicon Valley have name recognition that generates inbound interest and talent migration is not that they have better startup ecosystems in every objective dimension; it is that people are exporting their story. Investors and founders outside your geography can only decide to participate in your market if they know your market exists and has something worth participating in. The entire point of building a community is so that the people outside it can figure out how to get in. If your local newsletter goes to the same two hundred people who are already at every event, you have not solved a communication problem; you have built a very pleasant internal chat group.

I shared with our audience, that people like me are in every major startup ecosystem building community out there; there are maybe 100 of us in our situation. And yet, the number of times we have heard about your regional summit BEFORE it takes place can be counted on one hand. I intentionally criticized on our panel, that the number of times I hear about your regional startup summit counts in the hundreds; how you’ve not fixed that with effective communication and social media, is beyond me, and should be an immediate concern in the your next team meeting.

Fix your external-facing communications aggressively and specifically.

Not with vague “innovation hub” language that every city from Tulsa to Toledo is using simultaneously; with specific wins such as the company that just raised a seed round, the founder who just signed her first enterprise contract, the mentor who made an intro that turned into a partnership. Specificity is credible while generalized innovation language is noise. Credible and distinct promotion of what is genuinely working in your market is one of the highest-leverage investments an ecosystem can make, and it is almost universally underfunded compared to the demo days and networking events that mostly create noise.

Generic Is Invisible

If your startup ecosystem’s positioning is “we support tech startups,” you are competing with every other city in America and most of Europe on the least compelling possible basis. Being number thirty on someone’s list of “tech hubs to consider” is not a marketing problem; it is a strategy problem, and the solution is not better design on your city’s innovation landing page.

Specialization is the only strategy that generates genuine optionality for founders, and optionality is the only thing that makes a founder commit to a city long enough to build something there.

Most startups do not survive past twelve to eighteen months. A founder considering relocating to or staying in your city is implicitly asking whether, if this company doesn’t work out, there are options here in the sector she cares about, including jobs at other companies operating in that space. If your ecosystem is generic, the answer is maybe; if your ecosystem specializes in energy or agriculture or defense or healthcare or manufacturing, the answer is yes. That yes is what changes a founder’s calculus from hedging to committing, and committed founders build better companies than hedging ones.

The deeper point, which I have argued consistently is that sector specialization is not a choice you make by putting “AgTech Hub” on your website. It is something you earn by having anchor employers in that sector employing people, creating the talent density that makes founders in that sector willing to bet that the fallback exists if they need it. The implication for economic development professionals is that the path to a startup ecosystem in advanced manufacturing runs through having advanced manufacturing companies large enough to employ the operators who will eventually leave to found startups in that sector. You cannot shortcut that by running a pitch competition for AgTech companies before you have agricultural technology companies of any meaningful scale.

The Startup Investor Problem Nobody Wants to Discuss in Polite Company

There is a piece of conventional wisdom in ecosystem development that I am going to dismantle, because it has been causing damage for about fifteen years. The conventional wisdom is that creating a few local exits will naturally produce the next generation of angel investors and fuel the capital recycling that sustains a healthy ecosystem. The logic sounds right, but it skips a step.

Creating wealth does not automatically create startup investors. Creating wealthy people who know how to be startup investors requires education, exposure, and a very deliberate process of expanding what those people understand about how venture returns are actually generated and why the risk profile of a startup is categorically different from the risk profile of a rental property or a private equity buyout. Kevin and I both made this point during the panel with more force than it usually gets in polite ecosystem company that the investor class needs education as much as the founder class does, and almost no ecosystem is delivering it seriously.

The specific failure is real estate and oil-and-gas investors entering startup ecosystems with the expectation structures of their previous asset class.

• They expect control provisions that make no sense for early-stage equity

• They expect timelines that have nothing to do with how venture portfolios develop

• They expect the comfort of collateral in a category where the whole point is that there is no collateral because nothing tangible exists yet

The result is not just that they get disappointed; it is that they actively distort the founders they fund by imposing the wrong metrics, creating pressure for premature revenue at the expense of market validation, and exiting mentally from the ecosystem when they don’t get the returns they were never going to get on the timeline they were never going to get them on.

Kevin’s analogy for what fixes this is worth stealing (because yes, in startups, we steal ideas; stop pretending they’re some sacred IP that you deserve something for); Major League Soccer became substantially more watchable and competitive when it brought in internationally experienced players, because the elevated quality of play raised the standard for everyone else in the league. The same mechanism applies to angel networks. Bringing in investors who have done this correctly, at scale, in competitive markets, and giving local angels direct exposure to how those investors evaluate deals, structure terms, and think about portfolio construction, changes what local investors expect from themselves and from the founders they back. That education is not a soft program or a relationship-building exercise; it is a structural intervention that changes capital behavior in ways that compound over time through the recycling mechanism.

Capital from exits needs to stay in the ecosystem and get redeployed into the next generation of companies if the ecosystem is going to compound. Silicon Valley’s compounding was not inevitable; it was the product of decades in which founders who had liquidity events chose to become angels, angels who generated returns chose to form funds, and those funds chose to back local companies because the deal flow was the best they had seen anywhere. That compounding is precisely what does not happen when an exit creates wealth that flows into real estate or disperses to wherever the acquiring company is headquartered. The ecosystem gets a press release and then returns to the state it was in before the exit occurred.

What the Mayor, the Economic Development Director, and the Founder Should Each Do Monday Morning

It would be easy to conclude a panel like this with a feel-good synthesis and a call for collaboration, but that is the kind of output that ecosystem conversations produce in abundance while producing very few actual changes to anything structural. So, Ayhan asked, what does everyone do, right now?

Mayors need to get educated on what a startup actually is, why it is categorically different from a small business, and what that difference means for the kinds of policy interventions that help versus hurt. They also need to become loud, specific, external advocates for what is happening in their markets; not in the vague “we’re becoming a startup hub” language, but in the specific “here is what our energy tech founders built this year and here is who is buying it” language. Mayors who understand and articulate what is happening in their ecosystem change what kind of investors, founders, and advisors take their cities seriously. Mayors who don’t, regardless of how many ribbon-cuttings they attend at co-working space openings, don’t move the needle on anything.

Economic development directors need to rebuild their measurement systems around outcomes rather than activity, and they need to start treating the recycling of capital as an explicit policy objective rather than a hoped-for side effect. That means understanding fund formation barriers in their jurisdiction, understanding what tax architecture is doing to investor behavior, and being honest about the difference between an ecosystem that is performing and an ecosystem that is performing the appearance of performance. It also means funding the connectors and ecosystem builders who hold the informal networks together, because as I have laid out in Startup Ecosystems, these people are doing infrastructure-level work on event-ticket budgets, and that is not sustainable in any city that wants to build something durable.

Founders have a specific obligation that rarely gets named; the ecosystem that exists when they are building their first company is the one that previous generations of founders built for them, and the ecosystem that exists for the founders coming after them will be the one they build now. Paying it forward is not sentiment; it is the mechanism by which the mentor networks, the angel pools, and the institutional knowledge of how to survive in early stage actually get transmitted. The founders who treat ecosystem participation as a distraction from their company are the ones whose ecosystems remain permanently thin, because nobody in those ecosystems has the density of experience to actually help them when they need it most.

The throughline of the entire conversation was intellectual honesty, applied consistently to the gap between what ecosystems claim to be doing and what they are actually producing. Nobody in this work has all the answers, including Ayhan, Kevin, and me, and the landscape shifts fast enough that what worked five years ago is weaker today. But the gap that most regions are losing the most ground to is not strategic; it is educational. Mayors who don’t understand venture, investors who don’t understand startups, founders who don’t understand ecosystems, and economic development professionals who don’t understand any of the above; that is not a resource problem, it is a knowledge problem, and knowledge problems are fixable by people willing to do the work.

I’m writing today, thinking not only is a Saturday a terrible day to post, but ‘what am I thinking, tomorrow is Father’s Day, no one is online!?‘ But I’m fixated on getting this published because I’ve now scoped it with a half dozen states throughout the United States, and every conversation has ended with, “Why is no one suggesting this??” So, let me chalk this up as a bit of a Father’s Day gift, that as we get this right for founders, we’ll see a few more startups born.

Some context, after two decades of “innovation economy” investment, most states outside a handful of coastal markets still struggle to retain venture-scale companies, attract serious capital, or develop the mentor density that experienced founders need. So, what, exactly, has all that public money been doing?

Funding activity rather than outcomes, mostly. The gap between those two things is the entire game. Every state thinks it has a startup problem; most of them have a design problem, and there is a meaningful difference between those two diagnoses.

What’s Actually Broken in Startup Ecosystems

“We need more startups” is not a diagnosis. It’s a press release. It’s an easy consequence of growth. In the same breath that we can celebrate more venture capital, because there are more startups, when the reason is because we simply have more, we’re celebrating volume, not quality.

If you believe the challenge is a shortage of startups, your solution is more accelerators, more pitch competitions, more ribbon-cuttings in front of co-working spaces that will, within three years, quietly become real estate plays. If you understand the challenge is structural, that the scaffolding beneath the ecosystem is absent or incompatible with how startups actually work, you can start building something that addresses what’s actually broken.

So let’s be honest about what’s broken.

The silo problem is the one every ecosystem builder will acknowledge in private; too many confront it publicly, and they need to. Across city analyses from Baltimore to Iowa, too many organizations operate their own mentor pools, their own demo days, their own investor networks, their own CRMs, without sharing data or pipelines. The ecosystem, as such, is organized to benefit the organizations within it rather than the founders passing through it.

Not surprisingly, most civic funding actually like it that way, because celebration the local support of a local program makes headlines people can be proud of; nevermind if it actually works, that’s an afterthought.

When organizations compete for relevance, funding, and credit instead of coordinating around founder outcomes, silos are the only possible result; they are a feature, not a bug, of the incentive structure.

The missing middle problem (evident in capital but actually not just about capital) is the gap that kills more scalable companies than any other single factor. In city after city, the pattern is consistent: reasonable support for very early-stage founders (grants, incubators, seed angels), some support for established operating businesses (chambers, SBDCs), and almost nothing in between for the company that has found product-market fit, is generating early revenue, and needs Series A capital plus operational expertise to scale.

• Texas biosciences has this problem

• Iowa has this problem

• Las Vegas has it

• Baltimore has it

• Nearly every non-coastal market has it

It isn’t a coincidence; it’s a structural outcome of designing ecosystems around early-stage programming resulting in great ideas funded but lacking the understanding of scale to substantiate more investment. Without the underlying capital architecture to carry companies forward, few survive.

Again, cities get away with disregarding it because they can simply blame inexperienced founders or claim the failures are bad ideas. No, the reality is that we’re not enabling meaningful ideas and capable founders to compete and grow. The missing middle isn’t the lack of capital; it’s the lack of what’s required in the middle stages to justify the capital. The lack of the capital is the signal.

The university tech transfer problem is the most predictable challenge. In every city I’m engaged, I’ll ask about the University’s commercialization and entrepreneurship work; I’ll at least get a chuckle... easily half will tell me it isn’t working. And yet, this is the one most regions are least willing to confront. Most universities behave like medieval guilds guarding trade secrets behind toll booths.

Professors are inventors, not founders; that’s a functional reality the tech transfer world needs to stop pretending away with entrepreneurship centers. The IP sits on shelves. The commercialization paths are opaque. The incentives point toward licensing revenue and grant capture, not founder success. The OECD has been blunt about the fact that IP ownership is becoming less correlated with where value is ultimately created, and yet most regional universities still manage their innovation assets as if the year were 1985.

Why isn’t it being confronted? Because no local leader wants to criticize the local university; particularly when they don’t understand the work, what’s wrong, or how to fix it.

The corporate engagement problem is the one nobody names because corporate sponsors pay for the events (sometimes. Fine, okay... barely). What large employers actually provide, in most regional ecosystems, is a logo on a banner and a speaker at a demo day who spent thirty years in corporate finance and has never once made a payroll decision with three weeks of runway.

So, add to our list that Universities produce research without commercialization pathways, governments offer incentives without deal flow, and investors wait until traction exists, long after the region has lost half its would-be founders, that companies say they’re innovative and supportive, but come on...

The experienced mentor scarcity problem is the one that compounds all the others. Regions point to their business advisor networks, their university entrepreneurship centers, their corporate mentorship programs, and claim they have this covered. They don’t. A mentor who has only worked in operating companies defaults to corporate-scale advice that startups cannot implement and should not try. The relevant experience is experience working inside startups, making decisions under resource constraints, navigating the specific psychological and operational challenges that distinguish an early-stage venture from every other kind of organization.

And look, let’s be honest about this one instead of my deliberate criticism; yes, most cities have some great mentors. What you don’t have, what you can’t have, are enough specialists experienced with everything being done. This isn’t a city criticism, this is an observation of the reality that you can’t accomplish this locally.

None of these are solved by another accelerator cohort, another pitch competition, or another delegation trip to Austin. That last one deserves some direct commentary so let’s go there next...

You Cannot Replicate Silicon Valley; Stop Trying

The most damaging idea in modern innovation policy is not that Silicon Valley succeeded; it clearly did, spectacularly, for reasons worth understanding carefully. The damaging idea is that its success can be reproduced through imitation rather than through the conditions that made it inevitable.

That belief persists across decades of visible failure in cities, states, and countries worldwide, not because the evidence supports it, but because it offers a convenient explanation for why outcomes haven’t materialized. In the book, Startup Ecosystems, I’ve detailed this as the intellectual equivalent of cargo cult thinking; you’ve observed the visible artifacts of a successful ecosystem, reproduced them in your market, and are now baffled that the cargo planes aren’t landing.

What cannot be manufactured or imported:

• The regulatory environment and decades of venture history

• The density of experienced operators who have actually failed and scaled companies

• The organic talent network from Stanford and Berkeley that compounds annually

• The access to federal defense and research contracts that seeded early capital

• The sector specializations that took forty years to develop

• The culture of disruptive innovation, risk taking, and global thinking, or however you want to characterize it, that isn’t the same as yours.

The accelerator that works in Palo Alto frequently doesn’t work in Peoria, not because Peoria lacks talented founders, but because the conditions that make the accelerator model effective are absent. What a region needs to study about successful ecosystems is the underlying conditions, not the surface-level programming. Borrow principles; never copy implementations.

But the answer to “don’t replicate Silicon Valley” is not “do what we’ve always been doing,” because what’s being done isn’t working either; the silos haven’t closed, the missing middle hasn’t been bridged, the university tech transfer offices haven’t opened up, and the mentor networks haven’t deepened. The capital hasn’t shown up in meaningful density. More of the same with slightly different branding is not a solution.

If you can’t win the game by copying the rules from somewhere else, and you can’t win by playing by the rules you’ve always used, you change the rules.

First Principles: Force Everything to Work Together

Step back from the goals and ask the honest first-principles question: how do you force everything to work together when it demonstrably doesn’t?

• How do you create conditions under which investors find more deal flow?

• How do you get founders in front of better mentors without knowing the right person?

• How do you move capital from seed through growth more efficiently?

• How do you get universities to actually open their IP?

• How do you raise the ecosystem’s collective profile enough that external capital and talent pay attention?

The answer isn’t another program. It’s architecture; specifically, the overarching institutional architecture that every functional ecosystem has sitting above the individual organizations and below the specific programming.

We know from trade associations and economic development zones that policy, funding, and resources allocated broadly throughout a state produce results. We also know, painfully, that those same mechanisms fail startups consistently. The reason isn’t that the model is wrong; it’s that the model gets captured by corporate technology company goals, workforce development priorities designed for operating businesses, and policy objectives shaped by the needs of companies that already exist.

The problem in most states is not a lack of innovation assets but a lack of structure, connection, and communication. “Tech” is not startup. “Business” is not startup. The existing associations, built to serve operating technology businesses, cannot effectively serve founders trying to discern and develop new business models because those two activities require entirely different things from policy, capital, infrastructure, and support.

They do work; just not for this. Which hopefully raises in you the question, “Can’t we design one that does?”

A State Innovation Association; and No, It Isn’t What You Think You Already Have

Right now, some of you are thinking, “we have something like that.”

You’re thinking of the Governor’s Innovation Advisory Council, or the State Technology Office, or maybe the thing the biggest local accelerator runs when it wants to feel official. That is not what this is.

What’s needed is a statewide, bipartisan, formal nonprofit association that:

• Has no direct affiliation with any organization already operating for startups

• Is not controlled by any single city or metro

• Is not structured around whoever happens to have the most political relationships in the state capital

• Is explicitly chartered for startups; not “tech,” not “business,” not “entrepreneurship broadly” (once again, cheers to Louisiana for taking the lead on this one)

The WTIA in Washington state is one of very few that I’ve found that is a close analog. Former WTIA CEO Michael Schutzler, speaking to GeekWire on WTIA’s 40th anniversary, explained the critical pivot that made the difference:

“We pivoted it to a mission-driven organization, instead of a member-driven organization.”

In 2024, WTIA provided resources for startups in 19 counties, far beyond any single metro, and assisted over 250 startups with coaching, connections, and member resources. That geographic breadth is the point; most state startup programs orbit one or two major metros and pretend the rest of the state doesn’t exist.

The difference between what most states have and what’s being proposed here is specificity of charter. A generic technology or business association serves operating companies and calls it innovation. A startup innovation association is designed, from its founding documents forward, to serve the specific needs of the temporary ventures trying to discern and develop new business models; the Steve Blank definition of a startup, not the popular conflation of startup with any new small business.

A food truck is not a startup. A new law firm is not a startup. The semiconductor company in town isn’t applicable here because “tech” has nothing to do with this.

A Series A SaaS company changing banking and navigating its first enterprise sales cycles absolutely is, and that distinction determines every policy, program, and resource decision that follows.

How to Govern a Startup Association Without Getting Hijacked

The governance design is where most attempts at this fail, because they get captured by whoever shows up first with money or a title.

Dual entity structure. The association should be formed as two complementary organizations: a 501(c)(6) trade association for advocacy and industry representation, and a 501(c)(3) educational foundation to manage grants, research, and workforce development partnerships.

This dual structure is the mechanism by which the association can both lobby effectively for startup-friendly policy and receive philanthropic and public grant funding, which carry separate legal requirements. This foundation must provide both operational sustainability and policy credibility.

Board qualification. Every Board seat must be occupied by someone who has actually worked inside a startup or venture capital firm, not just “advised” businesses. This is the line that gets negotiated away in most states because large companies, government offices, and university administrators all want Board representation. Their participation is not unwelcome; except on a startup-focused Board, they need to bring startup-fluent representatives.

A VP of Innovation at a Fortune 500 who has never pushed to release despite known bugs and the lack of Q&A, is not equipped to govern a startup-focused institution. That’s not an insult; it’s a qualification.

Sector-specific Advisory Board. An innovation association cannot meaningfully serve “all technology,” because there isn’t a place in the world that does that effectively. The association must identify the sectors of genuine strength in the state and orient its advisory structure toward those sectors specifically.

For example, a region is strong in energy technology, aerospace and defense, agriculture, or data infrastructure. A state trying to build an advisory board for “tech broadly” ends up with a room full of people who are experts in nothing relevant to the actual founders in the ecosystem.

Paid professional staff. The consistent failure mode of these efforts is over-reliance on volunteers who burn out when the association extracts their expertise without sustaining it. The ecosystem builder burnout problem is well documented; the writer, event host, and connector who holds the informal network together is already doing this work unpaid in most cities. That has to change at the institutional level.

A minimum effective staff:

• President (public affairs and ecosystem experience)

• Marketing lead (experienced; not an intern running the social accounts)

• Communications director

• Research function

• Public Affairs and Legislation lead

• Venture Capital Liaison

• Startup Development Organization Liaison

These aren’t part-time committee chairs; these are paid positions, because the work of connecting a state’s innovation economy requires the same professional commitment as any other serious institutional function.

Who Pays, Who Participates, and Why the Enforcement Mechanism Matters

The funding model is where the design either becomes a genuine ecosystem solution or collapses into another organization serving whoever pays the most.

Founders and advisors: free members. If the association charges founders to access what should be a foundational resource of their state’s innovation infrastructure, it has already failed. Founders are who the association exists to serve; charging them to be served is the equivalent of a city charging residents to call the fire department.

Investors: paid members. Angel investors and venture capitalists derive direct economic value from higher-quality deal flow, better-prepared founders, and more transparent information about what’s happening in the market. They should fund it. The association should also provide investor education, because angel and venture capital investing is not the same as business investing, and treating it as if it were produces the kind of risk-averse, debt-mentality angel groups that proliferate in most non-coastal markets. The Kauffman Foundation has documented this gap extensively; their research on access to capital consistently identifies the need to build critical market infrastructure well beyond simply moving dollars around.

Service providers: paid members. Law firms, accounting firms, recruiting firms, marketing agencies, and technology vendors all derive client revenue from the startup community. If they want to be promoted by, featured within, or referred through the association’s network, a civically supported organization, then participation and financial contribution are the price of entry.

Universities: mandatory affiliates. All publicly funded universities running commercialization or entrepreneurship programs must affiliate with the association through a capital allocation and provide full transparency to all IP available within their programs. This is the university tech transfer fix that everyone says they want and nobody has been willing to mandate. Professors are inventors, not founders, and the gap between research production and commercial development requires active bridging; not hoping it emerges organically from relationship-building between organizations with no structural incentive to share.

The enforcement mechanism. Any entity serving startups or founders within the state is obligated to participate if it wants to qualify for support from the state or cities. No accelerator, no startup law firm, no innovation hub, no specific startup seeking public or state support can be featured, promoted, or directly supported by the association without participation. Organizations that benefit from public ecosystem investment should contribute to the shared infrastructure that makes their success possible.

The Infrastructure Layer Nobody Is Building

This is the piece that separates the model from every other association proposal you’ve encountered, because what makes an association for startups genuinely different isn’t only the governance and membership structure. It’s what the association provides beneath the programming level.

Programs sit between the association and the founders: accelerators, pitch training, mentorship sessions, cohort experiences. Those need to be developed locally to be meaningful. A program that works in Oklahoma City for energy technology startups is not the same program that works in Tulsa for aerospace companies or in Arkansas for AgTech founders. What works in Silicon Valley or Austin isn’t replicable there.

The association doesn’t replace local programming; it makes local programming more effective by providing the foundation it currently lacks.

Beneath the programs, there is infrastructure that every founder and every investor in every region of the state needs regardless of sector, stage, or city.

In the 21st-century sense that economic development professionals have been slowly accepting, infrastructure is not only roads and bridges; it is the foundational capacity that makes economic activity possible. In a startup ecosystem, three forms of it are currently absent or badly inadequate in virtually every state.

The CRM as Shared Ecosystem Intelligence

One of the most destructive patterns in regional ecosystems is the for-profit accelerator or innovation hub that positions itself as the center of everything, requires payment to access its network, and excludes organizations it considers competitive. The result:

• Multiple fragmented directories

• Multiple disconnected investor lists

• Multiple uncoordinated mentor pools

• Founders wasting enormous amounts of time figuring out who’s who

The fix is a genuine CRM, not a startup directory website that pretends to list everyone and provides janky promotional data, but a real platform properly developed on Salesforce or HubSpot, maintained by the association, and available to all members. It should document the angels, the VCs, the advisors, the mentors, the relevant service providers, and the startups. Everyone working on innovation in the state knows who matters because it’s organized and maintained by an entity with no financial interest in keeping certain people off the list.

The Agile Strategy Lab at the University of North Alabama put together some research on entrepreneurial ecosystem building which has identified cross-ecosystem fragmentation as the central barrier to spread:

“Ecosystems may be connected internally, but the connections across ecosystems are weak.”

A shared CRM maintained by an unbiased association breaks that pattern structurally and breaks the for-profit silo model at the same time.

AI as Member Capability

An AI system built out for use by association members, trained on the full suite of economic development, policy, regulatory, service provider, investor, and startup-oriented information relevant to that state, provides founders with on-demand access to the kind of guidance that currently requires knowing the right person, paying for a consultation, or being lucky enough to be in the right accelerator cohort.

• Need market research? Founders can ask it.

• Need the right advisors for a CleanTech seed-stage company working on battery storage? Everyone can ask it.

• Trying to understand how a state’s technology finance program interacts with federal SBIR grants? Ask it.

Yes, there are SaaS products and apps providing some version of this for well-funded companies. But if the association’s mission is to serve founders specifically, this is infrastructure that should be available as a member service rather than a subscription. The state provides roads regardless of ability to pay for private transport; AI-powered ecosystem intelligence should work the same way for the founders supposedly driving economic growth.

Social and Communication Infrastructure

Here’s where diplomacy runs out; after doing this analysis across enough cities, the pattern is unmistakable: most regional ecosystems are genuinely bad at community building, newsletters, social media, and earned media. Not somewhat lacking; genuinely bad.

A handful of individuals in any given ecosystem are very good at this, and they’re doing it unpaid, without institutional support, and heading toward burnout because the ecosystem extracts their social capital without funding it. As detailed in analyses of what ecosystem building actually requires, the event host, social group admin, and connector doing the work without support is the ecosystem builder.

That has to change.

A state innovation association with a paid, experienced Marketing lead and Communications director can build the social infrastructure such as the groups, the newsletters, the event community, and the earned media relationships, that requires experience to understand how to do proprly. This matters not just for founders talking to each other, but because compelling social and media infrastructure is how a region reaches the investors, advisors, and talent outside its borders who need to know what’s happening there.

You cannot be nationally relevant as an innovation economy with a newsletter that reaches 400 people and a social account updated twice a month by a volunteer.

These three forms of infrastructure, CRM, AI, and social-and-communication, are interdependent. The association doesn’t work without the infrastructure, and the infrastructure doesn’t become compelling without the association’s governance making participation a condition of ecosystem support. They are available freely only to the organizations that contribute as they should: paid membership for service providers, companies, and investors; participation for founders and advisors; data transparency for and from everyone. Organizations that refuse to contribute don’t get listed, don’t get promoted, and don’t qualify for state support.

The Accountability Problem Nobody Wants to Solve

Every ecosystem organization currently counts; they count events attended, mentors connected, founders served, workshops delivered. They do not, as if a rule, report on:

• Companies that reached Series A

• Jobs created by cohort graduates three years out

• Startups that exited

• Capital deployed that originated from local investors

The design of accountability within the ecosystem makes learning optional and self-criticism costly. Activity provides institutional cover in ways that value creation does not, and organizations optimize accordingly.

The state innovation association should publish a quarterly or biannual report on the performance of startup development organizations and investors, based on outcomes rather than volume or events. Transparency of performance data is a contingency of association membership, benefits, and any public funding from the state or cities.

• An accelerator that wants state support demonstrates outcomes, not attendance figures

• An angel network that wants to be in the CRM provides data on investments made, not just meetings held

We don’t fund roads based on how many planning meetings the transportation department held; we fund them based on miles built and traffic moved. The startup ecosystem deserves at least the same rigor.

How This Startup Ecosystem Model Solves What Has Failed

Pull back to the problems from the beginning and check each one against this model.

Silos are addressed through mandatory participation. When being listed in the shared CRM, receiving association promotion, and qualifying for state support all require contributing data and cooperation to the shared infrastructure, the incentive to silo your mentor pool or hide your investor relationships disappears. You can still run your accelerator with your own programming and culture; you just can’t keep the rest of the ecosystem in the dark about what you’re doing.

The missing middle is addressed because of mentor access and social/communication infrastructure as well as the investor education and the VC Liaison function. When the association is specifically charged with training angel investors in how startup investing actually differs from business investing, and maintaining a transparent CRM that shows where every company in the state sits relative to its funding stage, the information asymmetry that produces the missing middle begins to close. When you fix the fact that founders aren’t learning how to scale and can’t find channels in which to do so with support, you close the growth gap. Access to capital research identifies building critical market infrastructure as essential to expanding capital access; the association is that infrastructure.

University tech transfer is addressed through the affiliation requirement. When publicly funded universities must affiliate with the association and provide IP transparency as a condition of that affiliation, the incentive structure for commercialization changes. Founders can see what’s available. Advisors can match research to market opportunities. The association’s AI infrastructure can surface relevant IP when founders ask for it. This doesn’t solve the cultural resistance inside universities overnight, but it creates the institutional conditions for the connection to happen; which is where every serious university-startup engagement effort must begin.

Corporate engagement is addressed through the Board qualification requirement and the membership model. Large employers participate through Board seats occupied by people with actual startup experience; their financial contribution funds infrastructure that attracts better founders, which gives those employers better talent pipelines, acquisition targets, and innovation partners. The value exchange is explicit rather than assumed.

Mentor scarcity is addressed through the CRM, the AI, and the VC and SDO Liaison functions, which together create systematic connection between experienced operators and founders that currently depends entirely on who happens to know whom. When every advisor is documented in a shared CRM with their sector experience and stage expertise, and founders can query an AI system to surface the right match for their specific challenge, the brilliant mentor who would have been invisible because they weren’t part of any particular accelerator’s informal network becomes findable.

The result, fully implemented, is an ecosystem operating as a coherent system rather than competing nodes, with legitimate policy influence, measurable outcomes, and the kind of national profile that attracts external capital and talent. The gap between what states have and what they need is not more organizations; it’s connective infrastructure between the organizations that already exist.

Stop Treating Startup Ecosystem Infrastructure as Optional

There isn’t a state in the country that couldn’t benefit from this model. The problems it addresses aren’t regional quirks; they’re structural patterns that appear in Baltimore and Iowa and Las Vegas and Oklahoma and everywhere else that has invested in startup programming without investing in startup infrastructure.

The hidden architecture of startup cities is not accidental in the places that work; it was designed, funded, and maintained. What this model requires is experience operating across all of it simultaneously: economic development policy, startup ecosystem dynamics, nonprofit governance, venture capital, and the specific political navigation required to build something bipartisan and durable in a state capital environment. Those things rarely exist inside the organizations trying to build ecosystems, which is precisely why this has remained unbuilt for so long.

This isn’t a framework you hand to a committee and check back on in six months. Building this requires someone who has already done it.

Most states don’t have a startup strategy.

I know they think they do and you might even say they do, but what they really have a vocabulary and PR strategy to make it seem so. Press releases get stuffed with “entrepreneurship,” “innovation,” and “ecosystem,” ribbon-cutting photos get posted, and somebody wins an award for being an “innovative entrepreneur of the year.” Meanwhile, the capital stays scarce, founders leave or have to go back to a job, and the economic flywheel that the state desperately needs never turns.

Louisiana, of all places, just broke ranks with that charade and it did so by doing something both obvious and radical but correct: it deleted a word.

Nexus Louisiana, the nonprofit charged with statewide startup ecosystem development, removed the word “entrepreneur” from its website; their focus is Louisiana startups. On purpose. , the organization’s new President and CEO, himself the founder of Skilltype, a B2B SaaS platform now deployed by hundreds of institutions across the US, UK, Canada, Australia, and Israel, took the role and rewrote the mission in a way that people in my line of work have been pleading of everyone.

As Christopher Wink reported for Technical.ly, Zanders has a simple test he runs on audiences, “Ask how many would be thrilled to make $1 million in revenue this year. Some raise their hands. Then ask how many would be ‘grossly disappointed’ if they only made a million. A different group raises their hands. That’s how I explain the difference between an entrepreneur that’s a small business owner versus a high-growth tech startup.”

That distinction is not pedantic; it is, in fact, the entire argument.

And Louisiana making it explicitly, in policy, in language, and in organizational strategy, puts it ahead of most states that spend far more money on positioning.

The Word “Entrepreneur” Is a Liability for Economic Developers

What too few are willing to say out loud is that in new firm creation approximately 1 in 100 ventures is a startup that commercializes science, attracts risk capital, hires at scale, and throws off the multiplier effects that actually drive an economy. The other 99 are small businesses (laudable, important to communities, the backbone of local commerce) but structurally incapable of producing the kind of compounding regional wealth that economists mean when they talk about economic engines.

When a state economic development office says it backs “entrepreneurship,” it is almost always funding both of those categories out of the same budget with the same programs; without even realizing that entrepreneurship doesn’t mean new venture creation, it’s a behavior that some people exhibit, found throughout employees, non-profits, and neighborhood sports teams, as well as in some business owners and some founders. The problem we’ve created in the world is that encouraging into entrepreneurship sounds inclusive; it is actually incoherent. Wink put it plainly in his reporting on the Nexus decision, “When a state says it backs ‘entrepreneurship,’ it usually backs small business and high-growth startups with the same breath and the same budget.”

I’d push that even further; it’s actually referring to a type of people while misrepresenting that what you’re actually saying is that you want and are supporting startups and new businesses.

The US has been experiencing record business formation, but that growth is K-shaped with more non-employer firms at the bottom, innovation-led startups at the top (of which few succeed), and the middle, where we traditionally have businesses that hire and grow modestly, has hollowed out. Pouring development capital into “entrepreneurship” without specifying which kind is how you get a lot of activity metrics and not much economic consequence.

Zanders was direct about this with Wink, “That is such a broad term that encompasses anyone who is working on their own business.” Small business matters, and Louisiana has abundant programs for it. Nexus Louisiana is after something else entirely.

That “something else” has an accurate name: high-growth, venture-backable startups building new models that will change things and have the capacity to scale. That is a startup. And the person building one is a “founder.” These are not aesthetic word choices; they are definitional commitments that determine which founders you attract, which investors show up, which metrics you track, and (critically) whether any of the money you spend produces the multiplier effect that pays for itself in tax revenue and job creation fifteen years later.

The decision to replace “entrepreneur” with “startups” and “founders” is Louisiana’s way of drawing that line in policy-grade ink. More, it’s Louisiana leading the way on a path every other state, country, and city needs to follow.

Louisiana Has Been a Risk-Taking Culture Longer Than Anyone Gives It Credit For

The irony of Louisiana needing to assert itself in the startup conversation is that the state has one of the richest histories of innovation and risk-taking in the country. People forget, or never knew, because the dominant narrative about Louisiana is bayous and Bourbon Street, which is a little like defining Texas by cowboys and forgetting about semiconductors.

Start with the binocular microscope. Tulane University scientist John Leonard Riddell invented it; a device that transformed scientific research and the entire medical community. Then there’s Tabasco: Edmund McIlhenny grew his first commercial pepper crop in 1868 on Avery Island, secured a patent in 1870, and built a company that now sells in over 195 countries and territories; still family-owned, still made on the same island. That is a 155-year-old CPG startup that never got acquired, never moved its headquarters, and never chased a trend. That’s founder DNA.

There is also Alden “Doc” Laborde, a former Navy engineer who designed the world’s first mobile offshore drilling rig after his own employer at Kerr-McGee told him the idea had “too many unknowns.” The ASME (The American Society of Mechanical Engineers) designated his platform, Mr. Charlie, a Historic Mechanical Engineering Landmark in 2012. The modern offshore oil industry, which generates billions in global revenue, traces directly to a Louisiana founder who couldn’t get corporate buy-in so he went and built it anyway. Madam C.J. Walker, born Sarah Breedlove in Delta, Louisiana, overcame poverty and Jim Crow to become America’s first self-made female millionaire.

Jazz, Zydeco, and Cajun music are not just cultural exports; they are proof of a creative economy built on improvisation, risk, and fusion; exactly the cultural substrate that healthy startup ecosystems require. Louisiana Inventors & Innovators profiles six visionaries, including Laborde, Walker, J.M. Lapeyre (who mechanized the shrimp industry), and Ruth Fertel (who turned a single restaurant into the Ruth’s Chris Steak House empire), and describes them precisely the way you’d describe a great startup portfolio, “businesses and ideas which in many cases revolutionized industry and provided livelihoods for hundreds and thousands of people.” Ignoring the fact that something like Zydeco music is an invention, requiring risk-tolerant, creative, and technically inclined “founders,” who develop not just the code (musical score) but also the invention of tools that produce it (instruments) is why most startup ecosystems plod along – thinking startups are only “tech” in some coding language or hardware engineered sense.

The culture is what leads the way and entrepreneurial DNA matters, and Louisiana has both; what matters is then guiding and supporting that properly in distinguishing who and what is actually being developed in the economy. This is where Louisiana just took a leadership position everyone else needs to follow.

What has been missing (and why everyone is confused), are the institutional frameworks that converts that culture into a deliberate, measurable startup economy, until now.

Louisiana’s Economic Reality Makes Clarity Non-Optional

The state’s macroeconomic picture is the kind that forces hard choices. The state’s real GDP reached $248.6 billion in 2023, a healthy 5% increase from the prior year; but the growth story is complicated because Louisiana ranks among the bottom of states for population growth and has been losing residents while the broader south booms. The petrochemical industry, which once represented roughly a third of state GDP, now accounts for less than one-fifth and continues to shrink as demand shifts and the energy transition reshapes the sector.

(aside: Louisiana, we should talk about the fact that you should be in my Energy Belt work)

What the economic reality means is that Louisiana cannot afford to be vague.

A state with commodity-dependent revenues, population pressure, and a lean social safety net cannot dump its economic development budget into “entrepreneurship” programs implying it’s for everyone while actually not even serving actual entrepreneurs; meanwhile, producing no defensible data on high-growth firm creation. The energy industry still generates 25% of Louisiana’s economy and employs over 306,000 people directly and indirectly. That’s a massive asset base but it’s also a concentration risk that makes the startup diversification argument existential.

Zanders, from what I have been looking into, understands this in terms that most ecosystem builders never articulate. He has even reframed the way Louisiana’s leaders think about big-ticket wins like the $30 billion Meta data center that every state chases with tax incentives; you don’t get the data center without a tiny startup twenty years earlier building the infrastructure, the talent pipeline, and the technical culture that makes a location plausible for hyperscale investment. As Wink reported Zanders’ position, “Homegrown beats lured;” tax incentives attract facilities while startup ecosystems create the conditions that make a state worth attracting.

Louisiana’s government has been a mixed actor in this regard. Louisiana Economic Development‘s FastStart is genuinely well-regarded as a workforce development mechanism, and the state’s participation in the State Small Business Credit Initiative (SSBCI), through the American Rescue Plan, has seeded a new generation of venture funds. But governments tend to default to visible activity over structural change; cutting ribbons, announcing programs, and counting total jobs created or companies served rather than asking whether any of those companies will generate multiplier effects at scale. Such systems, as always, behave exactly as designed, and Louisiana’s systems (as most do) have historically been designed to measure activity, not consequence.

Nexus Louisiana’s 10-year goal of 1,000 net-new high-growth startups is not an activity metric; it is an outcome metric, and that distinction is the whole game.

Startup Ecosystems Create the Conditions that Make a State Worth Attracting – Here’s Louisiana Startups…

Louisiana’s startup support infrastructure is more developed than the national narrative suggests, concentrated in New Orleans and Baton Rouge but increasingly operating statewide.

Here is what I see when I dig in:

Statewide Leadership

• Nexus Louisiana: The Baton Rouge-based statewide nonprofit now under Zanders’ leadership, operating the Louisiana Technology Park (a business incubator in Baton Rouge), and pursuing its Stealthmode Startup School, which inverts the standard accelerator model (another YAY!): rather than coaching self-styled entrepreneurs, it converts domain experts (energy engineers, healthcare professionals, agtech specialists) into venture-backable founders. That inversion matters not just because the typical accelerator model isn’t actually working in most cases but because domain expertise in industries where Louisiana already has scale is the raw material for genuine, defensible startups.

• Louisiana Economic Development (LED): The state agency overseeing incentive programs, LED FastStart workforce development, and the coordination of the State Small Business Credit Initiative (SSBCI) equity fund program. LED’s technology sector data shows over 2,900 tech businesses operating statewide, with 30,470 IT professionals in the workforce.

New Orleans

• The Idea Village: One of Louisiana’s oldest and most established startup support organizations, now operating the Idea Village Momentum Fund, providing both programming and capital to high-potential ventures.

• Propeller: Working at the intersection of health, food, education, and water; Propeller specifically supports founders from underrepresented communities; operating a coworking space in New Orleans alongside programming.

• New Orleans BioInnovation Center and New Orleans BioFund: Originally seeded by HUD and the Office of Community Development Disaster Recovery, has invested in 17 local businesses; to date, portfolio companies have created or retained over 500 full-time equivalent positions. The fund is transitioning toward a micro-venture capital model.

• Tulane University / Lepage Center: The Albert Lepage Center for Entrepreneurship and Innovation publishes the Greater New Orleans Startup Report, launched in 2019 as the first comprehensive regional benchmarking tool for the startup economy. Tulane was prominently featured in the Nasdaq ARIE report as a key driver of New Orleans’ innovation performance.

• 4.0 Schools: An education-focused accelerator and fellowship program for education innovators and edtech founders, operating out of New Orleans with a national reach.

• Camelback Ventures: A fellowship and funding model specifically backing Black and Latinx founders in education and social impact sectors.

• Future Factory: New Orleans-based venture studio and fund.

Baton Rouge and North Louisiana

• LSU Innovation Park: A 200-acre business incubator operated by LSU Innovation (the university’s commercialization office), home to early-stage tech companies and connected to federal SBIR/STTR grant pathways, legal resources, and patent support.

• New Louisiana Angel Funds (NLAF): Managed by BRF’s Entrepreneurial Accelerator Program in Shreveport, focused on North Louisiana-based startups at angel and pre-seed stages.

• Pax Momentum: PAX is the first pre-seed investor to combine funding with structured B2B sales foundation

University-Linked Programs

• 1834 Ventures: A fund focused on early-stage ventures emerging from Louisiana’s academic institutions.

• Tulane Ventures and Ochsner Health‘s Ventures: Tulane being obvious, Ochsner is a nonprofit integrated, academic healthcare system.

Capital Formation: From Angels to Venture

Louisiana’s venture capital ecosystem was, for a long time, nearly nonexistent in the national conversation. The Nasdaq ARIE report ranked New Orleans 19th by recent VC totals among top-20 ecosystems, with less than $400 million in recent annual venture funding across fewer than 60 tracked startups. That is a ceiling in need of breaking, but the infrastructure to break it is now more materially present than it has been at any point in Louisiana’s history.

The federal SSBCI program, seeded by the 2021 American Rescue Plan, created nine locally run venture funds now actively investing in Louisiana startups:

• Boot64 Ventures which closed its Magnolia Fund I at $10 million; named Equity Fund of the Year and run by John Roberts who is as much an ecosystem builder as he is investor.

• Callais Capital, an active early-stage investor in Louisiana-based startups across software, energy, and emerging tech.

• The Idea Village Momentum Fund, the operational arm of The Idea Village providing equity capital to portfolio companies.

• Propeller Ventures, Capital aligned with Propeller’s mission focus on underrepresented founders.

For angels and early check-writers:

• Gulf South Angels has surpassed $20 million in investments since its inception, making it one of the most consistent early-stage capital providers in the region.

• Flamingo Funders, a collective of accredited women angel investors in the New Orleans five-parish area, making individual early-stage investments and mentoring women-owned startups.

• Revelry Venture Partners, founded and led by Gerard Ramos, a 20-year angel investor now operating a structured fund.

For later capital:

• LongueVue Capital, one of the biggest Louisiana-based private equity/venture firms, having raised more than $360 million through its Partners IV fund.

• Bluehenge Capital Partners, a Gulf South-focused investment firm.

• Corridor Ventures, Co-founded by Kwamena Aidoo, serving as the U.S. Southeast region’s “corridor.”

• Camelback Ventures and Future Factory already introduced.

Mark Graffagnini, managing partner at Cara Stone law firm in New Orleans, confirmed that Louisiana software companies historically attract the most investment, followed by energy, agtech, and food and beverage startups, with health and biotech deals expected to increase. While that framing is helpful, what’s to be celebrated is when an ecosystem moves forward in sophistication of their language and expectations so that it’s clear where and why things are working. “Software companies” isn’t a meaningful framing but looking to energy, agriculture, food, and beverage, we see the sectors in which startups thrive there, be that through software, hardware, or just new models and recipes that change the sector. Louisiana is finding its defensible sectors, which is precisely the right move; chasing whatever’s trending or blanketly saying ‘our strength is in tech,’ is how you build a generic ecosystem that no serious investor respects.

Notable Louisiana Startups Proving the Thesis

Louisiana’s portfolio of notable startups is better than we might think; and each of them embodies the founder logic Zanders is institutionalizing.

• Natrx | Adaptive Infrastructure is a climate-tech startup with artificial reef and breakwater technology, having raised roughly $7 million; 85+ completed deployments worldwide from a manufacturing facility in Amelia, Louisiana. This is domain expertise (coastal Louisiana’s existential erosion problem) converted into a scalable technology business; exactly what Nexus’s conversion model should produce more of.

• Skilltype, which is Zanders’ own platform for workforce development in libraries, deployed by hundreds of institutions across multiple countries. A Louisiana-born product company with international reach.

• Bifrost which became part of Techneaux Technology Services, in a Lafayette-based software company specializing in bridging complex industrial data systems for oil and gas companies. Its CRO, Geoff Daily, referred to “Cajun-uity” to describe the Louisiana culture of problem-solving regardless of conventional constraints, “There is something about the Louisiana culture that has a willingness to not just go the same way people have been doing things, and a willingness to solve a problem regardless of the rest of the stuff around it.”

• RCOAST provides drone data collection for coastal risk assessment and protection planning.

• Glass Half Full is recycling glass into sand for land rebuilding; solving Louisiana’s coastal erosion and waste problems simultaneously.

• Codegig which is software services but integrated with existing workflows in an organization.

• Lumen Technologies, is now one of the largest legacy tech employers in the state, providing anchor employer infrastructure that makes risk rational for potential founders in networking and telecom.

• SpaceX is Building drones at Port Fourchon, which matters for the ecosystem not just as a brand name but as the kind of advanced manufacturing presence that creates adjacent talent and supplier networks.

The WYES documentary rightfully adds the Ruth’s Chris Steak House origin story to this list: Ruth Fertel, a divorced mother who mortgaged her house to buy a failing restaurant in 1965 and built an international franchise empire. That is a founder story, and clearly an entrepreneur, regardless of how I’m distinguishing startups.

Statewide Alignment Is the Unlock Here

Christopher Wink noted that the Nasdaq Entrepreneurial Center’s challenge to Louisiana was not “get more startups” but “organize statewide.” This adds to why I want to celebrate their discernment in language but also real leadership. Startup Ecosystems pushes clearly that silos are bad for startups and discouraging of investment but very few cities and states are actively breaking down barriers and putting the infrastructure in place that connects everything in the ecosystem.

New Orleans is a surprise high-performer for startup efficiency relative to its capital base; which is a signal that the fundamental conditions exist, and that the ceiling is structural rather than intrinsic.

“We don’t have the luxury of New Orleans competing with Baton Rouge, or Lafayette competing with Shreveport. That’s a luxury Texas has, where Houston and Dallas and Austin can have these really large, self-sufficient ecosystems unto themselves. We have to think as a state.”

I love that quote from Tony Zanders, not only because he’s exactly right about Louisiana but because as a resident of Austin, Texas, he’s actually pointing out what holds Texas back as well; it’s idiotic to establish ecosystems based on city boundaries and then compete with your neighbors. Startups don’t serve a specific city. And while many investors are passionate about where they live and might favor investment as such, it’s a pretty ignorant investment thesis to limit a portfolio to a couple dozen square miles of location. Texas doesn’t have that luxury, it’s done well despite regional collaboration and specialization, because the cities are so large.

Capital is not missing from Louisiana because investors haven’t heard of it; capital is scarce relative to the state’s ambitions because the signal that capital responds to (density of high-growth companies in defensible sectors, with experienced founders and clear market validation) has not yet reached the threshold that triggers compounding investment. Capital access, education pipelines, talent movement, and pro-founder policy, all require statewide coherence to activate the flywheel.

You cannot get the flywheel from one city operating independently. Anywhere.

Statewide, sector oriented, or at least regional storytelling, the kind that tracks founders, reports on funding rounds, and builds a narrative that investors elsewhere want to follow, is one of the least-appreciated infrastructure requirements for a startup ecosystem.

Because I’ll beat this drum till everyone catches on, marketing is not promotion; it is market discovery; the process of communicating genuine value signals to audiences that will respond because you’ve figured out what matters. Louisiana telling a precise, credible story about what it produces, in which sectors, and with what outcomes is more important to its capital formation trajectory than any single fund or accelerator.

Grab a copy of Startup Ecosystems

What Louisiana Does Well

Louisiana’s strengths are considerable and under-leveraged:

• Domain depth in sectors that produce genuine startups. Energy, coastal resilience, agriculture, healthcare, and advanced manufacturing are not generic tech plays. They are sectors with massive market sizes, deep technical complexity, and the kind of IP-generating R&D that produces defensible companies. The energy sector alone generates 25% of Louisiana’s economy; the talent and technical knowledge embedded in that workforce is exactly what converts into founders. Oil and gas companies here are actively partnering with Louisiana software firms for data solutions; a market dynamic most startup ecosystems would kill for.

• Cultural risk tolerance. The “Cajun-uity” concept is not a cute slogan. Well, let’s be frank, it is, and it’s brilliant. The culture that produced jazz improvisation, offshore drilling innovation, and a hot sauce empire that outlasted Reconstruction, two world wars, and five generations of management is a culture that understands optionality and tolerates failure. Startup ecosystems require a culture that celebrates risk-taking, and Louisiana’s creative, improvisational, multi-lingual heritage is genuinely rare.

• New institutional clarity. Nexus Louisiana’s language pivot needs to be established as institutional policy (everywhere). The goal of 1,000 net-new high-growth startups over a decade, combined with the industry-expert-to-founder conversion model, is the most structurally coherent startup strategy articulated by any statewide organization in recent memory.

• A new generation of capital. Nine SSBCI-participating funds did not exist five years ago. As one Louisiana investor told the Times-Picayune: “We’re tired of watching talent, money and decisions flowing out of state. When we invest in our startups, we’re saying we want to own our future.” That is the sentiment of a capital ecosystem beginning to compound.

• Statewide storytelling infrastructure. Independent journalism about founders, funding, and policy, without bias but with discernment to matter to audiences, is a non-obvious but essential ecosystem component that most mid-sized startup regions still lack.

Where Louisiana Needs to Be Honest with Itself

The gaps are structural, not moral, and being honest about this stuff is how regions work meaningfully with others to improve:

• Capital volume is thin. Less than $400 million in recent annual VC investment, with 2025 seeing fewer than two dozen tracked deals below $50 million, means the pipeline is functioning but there is a ceiling. Louisiana needs Series A and growth-stage capital to prevent its best companies from relocating for follow-on rounds.

• Population decline is an ecosystem threat. Talent follows opportunity, not programs. A state losing residents to the surrounding Sun Belt is signaling that its opportunity density is lower than its neighbors, regardless of how many accelerator cohorts it runs. Nexus’s 1,000-startup goal is also, implicitly, a talent retention strategy. What helps? The narrative design of more meaningful storytelling so people KNOW why to stay.

• University commercialization needs to mature. LSU Innovation Park is a real asset, and Tulane’s Lepage Center is producing useful benchmarking data. But the gap between research production and company formation at Louisiana’s universities remains significant. IP-to-startup commercialization pipelines require different models, patient capital, experienced technology transfer staff, and a culture that values exits over patents; most of Louisiana’s academic institutions seem to still be optimizing for patent counts rather than company creation. It’s known how to fix this.

• Ecosystem coordination remains fragmented. Multiple cities, multiple anchor organizations, and limited cross-regional collaboration is the historical pattern. Zanders’ “think as a state” mandate is correct; executing it requires governance structures, underlying infrastructure, shared KPIs, and the trust between New Orleans, Baton Rouge, Lafayette, and Shreveport.

• The accelerator-to-outcome pipeline is under-validated. Louisiana has accelerators; it needs data on which ones are producing companies that raise follow-on capital, generate revenue, and stay in state. Without that data, program funding continues on the basis of activity metrics rather than consequence, which is how ecosystems stay busy without getting better.

Why Every State Should Copy What Louisiana Just Did

The decision by Nexus Louisiana to delete “entrepreneur” from its website is not a branding exercise. It is an accurate and meaningful governance choice that signals to founders, investors, program partners, and government funders that this organization understands what they’re doing and has a defined target, a defined methodology, and a willingness to disappoint the wrong audience in service of the right one.

Most states cannot do this. Too many governments won’t do it. The political cost of saying “we are not a small business program” is too high when small business owners vote and write checks to campaigns. The institutional cost of saying “our accelerator is for founders, not self-styled entrepreneurs” is too high when accelerators need enrollment numbers to justify their budgets. The communications cost of explaining that “job creation is a lagging indicator, not a goal” is too high when every economic development press release leads with jobs.

Louisiana just paid those costs, willingly, by removing the wrong word.

Zanders’s inversion of the accelerator model (converting industry experts into founders rather than coaching self-styled entrepreneurs into industries) is the embodiment of everything competent startup ecosystem design calls for. Industry experts have the domain knowledge, the professional networks, the credibility with early customers, and the sector-specific insight to build defensible companies. They are not the people who usually show up to pitch competitions; they are the people Nexus is going to work with because they more likely create enduring and valuable startups.

Homegrown beats lured, precision beats breadth, and startups beat entrepreneurship, not because small businesses don’t matter, not at all! But because they aren’t the same thing; they require different strategies, different capital, different metrics, and different program designs. Conflating them has been the single most expensive mistake in American economic development policy for the last two decades.

Louisiana just stopped making it.

Full disclosure: this article is a little outside my usual lane. I write about startup ecosystems, capital markets, and why most economic development strategy is wrong. I do not typically wade into opportunity territory, but the Annoyance Economy is such a nakedly obvious sector opportunity for founders that ignoring it would be journalistic malpractice, and I cannot help myself.

Let’s start where everyone is starting right now: Alamo Drafthouse.

If you have been anywhere near a film community, an Austin Twitter circle, or, frankly, any group of humans who go to movies, you already know what happened. Sony Pictures Entertainment acquired Alamo Drafthouse Cinema in 2024 for an estimated $200 million. The Alamo, for the uninitiated, was not just a movie theater; it was the movie theater that made its entire identity out of throwing people out. Literally. The chain built its national reputation on ejecting patrons for talking or using their phones, and posting the audio of their furious voicemails as they were tossed during pre-show entertainment. That was the product. Their policy was the product. The culture of radical commitment to the cinematic experience was the thing people were paying for.

Then Sony arrived, as major studios tend to do, with a fresh set of priorities. The company launched a QR code ordering system, asking customers to pull out their cellphones in the theater to scan a code and place an order online, replacing the traditional pen-and-paper system that servers would collect at the seat. To recap: the chain that banned phones is now requiring phones. That is not a pivot; that is a hostage situation where the hostage is your entire value proposition.

Lord of the Rings actor and Austin resident Elijah Wood called it “a move completely antithetical to the ethos of the Alamo and to those who love the theatrical experience they provide,” adding, “This is a profound and upsetting mistake.” Wood, who is not a random internet commenter but someone who actually went back and tried the new system before rendering judgment, was not done. Adding, “I can tell you it’s truly awful. Rather than making ordering food and drink more efficient, it actually adds steps to the process. AND if you want to order additional items during the film you HAVE to open your phone,” he elevated himself to hero-of-the-people. The Austin Film Critics Association piled on, stating that requiring a QR code for ordering “is the antithesis of what made the Drafthouse what it is.”

A Change.org petition to reinstate the no-phones policy collected more than 10,000 signatures, and employees at a Denver location went on strike. Film editor Andy Young told The New York Times, in words that Fast Company quoted directly, “This hurt more than, like, most of the breakups I’ve had.” That quote would be funny if it were not such an accurate description of what it feels like when a company acquires something people love and immediately begins optimizing the love out of it.

This is not a one-off corporate blunder. It has a name.

Where Love is Lost, Find Opportunity

Writer and activist coined the term “enshittification” in 2022, and it has since become the most accurate name in the consumer vocabulary. CloudFest documented in a keynote breakdown of Doctorow’s framework; the decline in quality and value of services follows a predictable pattern: platforms start out great, lock users in, shift value to business customers, then drain the value from everyone to feed shareholders. Australia’s Macquarie Dictionary named enshittification the word of the year for 2024, which tells you something about how universal the feeling has become. When a dictionary committee’s word of the year is a profanity wrapped in academic suffixes, the committee is trying to tell us something.

The examples are everywhere. Google search is loaded with ads and sponsored content; Amazon promotes cheap, badly made products regardless of what a customer actually searches for; Facebook, once a way to keep up with people you actually know, has evolved into an ever-more frustrating feed of junk. Uber launched by subsidizing rides to build market share, then raised prices once drivers and riders had no viable alternatives. Wikipedia’s entry on enshittification is worth a read, “initially, vendors create high-quality offerings to attract users, then degrade those offerings to better serve business customers, and finally degrade their services to both users and business customers to maximize short-term profits for shareholders.” Doctorow, writing for the Financial Times, warned that enshittification is coming for absolutely everything.

Alamo Drafthouse is simply enshittification at its most emotionally legible.

Why I start with that is not just that it’s a shining example, the venture is (or was) also the perfect example of a non-tech startup. In a world oriented to thinking startup means tech, we have (or had), exactly what we need to say to people, “no, that’s not true, look at how Alamo Drafthouse changed the business model and changed how theaters are loved.” Then, a beloved, distinctive, irreplaceable experience got acquired by a corporation that does not have “direct contact with customers” by its own CEO’s admission, and so it defaulted to optimizing for metrics that made the experience operationally cheaper at the cost of what made people care about it in the first place. This is the story of every platform, every acquisition, every corporate rollup, playing out in a movie theater in Austin, Texas. You can see it in real time, with Elijah Wood live-tweeting from the audience.

Now zoom out.

The Annoyance Opportunity

The , working alongside Stanford economist Neale Mahoney, released a report quantifying what they call the Annoyance Economy: the total financial and psychological toll of navigating everyday corporate and bureaucratic frictions.

Their number is $165 billion annually in wasted time and lost money. This is an amount greater than the GDP of 14 U.S. states. And we have here not one problem to solve but several, each of which is a startup category waiting for someone to care.

Junk fees alone account for $90 billion per year; phone scams extract $25.4 billion; time spent with health insurance administrators costs consumers $21.6 billion in wasted hours alone; waiting for medical services costs another $19.4 billion; robocalls cost $8 billion; waiting for government services costs $1.6 billion. These are the costs of what Groundwork Collaborative’s Chief of Policy and Advocacy Alex Jacquez called a, “vibes-based analysis of the economy” in which “every consumer interaction is just harder than it used to be.”

The mechanism behind most of these costs is not incompetence, it is intentional design. Making cancellations more arduous for customers can boost corporate revenues by anywhere from 14% to over 200%. Friction is a revenue strategy.

It’s a questionable and hated revenue strategy, but it is a revenue strategy; the hold music is not an accident.

The chatbot that hangs up on you after forty-five minutes is not a bug; it is a feature of the attrition model, where the goal is to make you give up on your refund. The Groundwork report’s authors describe these as “designed friction,” deliberate architectural choices that extract value from consumers through exhaustion rather than exchange.

The time Americans spend on the phone with customer service has spiked by 60% over the last 20 years as companies have systematically reduced the human infrastructure for resolving problems while simultaneously expanding the complexity of products and billing. You are spending more time on hold because spending time on hold costs you money and saves them money. That is the transaction; you just did not consent to it.

Why Startups (and investors) Might Step it Up

Here is where my work starts to intersect with what might seem like a consumer frustration column; I spend most of my time thinking and writing about what makes startup ecosystems actually produce value rather than just activity and how venture capital avoids ecosystems that mistake signals of busyness for signals of opportunity. One of the most reliable signals in any market is a gap between what people are paying and what they are getting. Another is the presence of locked-in users who cannot easily exit. Another is a product category where incumbents have stopped competing on quality and started competing on switching costs.

The Annoyance Economy scores a hat trick on all three.

When a $165 billion annual toll exists and the entities producing it are profitable specifically because they have made it hard for you to stop paying them, that is not a market in equilibrium. That is a market waiting for a founder who is angry enough to fix it and clever enough to monetize the fix.

The gym equipment industry was built on people who got tired of paying for gym memberships they could not figure out how to cancel. Think about what happens when you try to cancel your internet service provider for an entirely valid reason such as, oh, I don’t know, you’re moving. Password managers exist because enterprise software made logins so complicated that a third-party product could build a business on simplifying one interaction. Anti-spam services are an entire sector premised on cleaning up the mess that corporate marketing automation made.

Every sub-sector of the Annoyance Economy follows the same logic: a monopoly or oligopoly creates friction, consumers get frustrated, a founder spots the gap, and a new market category gets created.

The Alamo Drafthouse story is particularly instructive for founders because it illustrates how fast a brand can be destroyed and how intensely consumers will respond to someone who offers them the opposite experience. Alamo built its national brand specifically on the promise that it would enforce standards nobody else would; it was the anti-AMC. It was the theater that took cinema seriously enough to finally put a stop to phone distractions. People loved it for that. When Sony removed that differentiator, the reaction was not mild disappointment; it was loud and public outcry. The Alamo Drafthouse subreddit exploded with backlash, with IndieWire critic David Ehrlich publishing a piece titled “The Absolute Hell of Watching a Movie at the Alamo Drafthouse in 2026”. The word “betrayal” appeared repeatedly in consumer commentary.

That anger is a market signal. Wherever consumers grieve the loss of a product that respected them, a startup can be built on giving that respect back. It is not complicated. It is just hard, because it requires a founder to care about quality more than they care about extracting value from a captive audience, which is increasingly rare in a corporate landscape trained to optimize for short-term earnings per share.

It’s Said Disruption is a Dirty Word Thanks to Startups, here it is Not. Disrupt This.

The most interesting founder opportunity in the Annoyance Economy is not in fighting the incumbents directly.

Regulation is one tool, and more than two-thirds of voters want Congress to prioritize addressing annoying business practices like spam calls, while 75% say they are frustrated by surprise junk fees, so the political appetite is clearly there. But regulation moves slowly and gets captured by the industries it regulates; I advocate that startups and investors need to be involved in policy work, but that is distinctly different path to chase. The faster path is building alternatives that make the friction irrelevant; you do not need to pass a law banning subscription traps if you build a product that monitors subscription spending and cancels dormant ones automatically. You do not need Congress to ban resort fees if you build a travel platform that shows total price at the first click and guarantees no hidden charges at checkout.

The annoyance exists because incumbents have no competitive incentive to remove it. The moment a competitor makes the frictionless version of any category available, incumbents lose the tool they were using to extract value passively.

The economics of disruption work precisely here. Disruption, in Clayton Christensen’s original framing, does not mean “cool new technology;” it means a product that starts by serving the customers incumbents have stopped caring about, at a price point or experience level that the incumbent refuses to match because matching it would cannibalize their existing margins. Every incumbent extracting $165 billion annually in designed friction is making a margin bet that the Annoyance Economy is permanent.

Founders who disagree have a $165 billion thesis to test.

This is admittedly a bit of a detour from the startup ecosystem analysis and venture capital strategy content I usually publish here, but the Annoyance Economy is exactly the kind of sector-level market signal that any serious analysis of where ecosystems fail and where markets open cannot responsibly ignore. The work of identifying what separates ecosystems that compound from ecosystems that stall is ultimately the work of reading these signals at scale. Enshittification is not just a clever internet word; it is a structural description of how market power concentrates and then decays. Understanding why it happens is the same intellectual exercise as understanding why startup ecosystems stall: Incumbents stop competing on value.

Friction replaces innovation and switching costs replace product quality; eventually someone builds the thing people actually wanted, and the incumbent discovers that captive users are only captive until they are not.

The Alamo Drafthouse situation is, in miniature, the entire story. A startup (a non-tech startup) with a differentiated product and a loyal customer base gets acquired by a company optimizing for throughput rather than experience, the brand value gets dismantled in the name of operational efficiency, and the customers who defined the brand’s success start asking out loud whether anyone is going to build the thing Sony just destroyed. Someone should; historically, when that question gets asked loudly enough, someone does… why not you?