Last week, the National Academies released its evaluation of ARPA-E’s mission and goals, an assessment of 17 years of one of the major energy innovation programs inside the federal government. ARPA-E was modeled on DARPA, built on the idea that capturing high-impact, high-reward, off-the-roadmap technologies required a more nimble way for the Department of Energy to pursue transformative ideas. And it has delivered. ARPA-E has seeded incredible startups like Antora and Smart Wires, funded areas the Department wasn’t otherwise covering, and used temporary hiring authority to bring in genuinely talented program directors to build programs around their expertise. Its annual Energy Innovation Summit remains one of the seminal gatherings of the entire energy innovation ecosystem.

The report contemplates both that positive impacts and the improvements still needed. A few months ago, I was invited to present to the Academies committee on how the Office of Energy Efficiency and Renewable Energy (EERE) now CMEI, my former office, approached program design, implementation, and impact tracking, and I served as a reviewer for the final report. The slides I presented are here, and I welcome any thoughts or opinions.

The argument I made to the committee is the same one I want to make here: transformative technologies do not come from a single funding decision or mechanism. They come from a portfolio. And that’s true not just based on the findings of the Academies. The argument flows through in two ways: (1) you need a portfolio of approaches beyond just funding early innovation to support commercialization of technologies from ideation through commercial demonstration; and (2) you need a portfolio of mechanisms to support this journey.

Follow the Fervo thread

Take Fervo, one of ARPA-E’s most cited success stories. Fervo received ARPA-E funding that supported the company’s exploration of enhanced geothermal systems, a novel way to extract geothermal energy where there is no surface expression. That funding mattered and certainly came at a critical time. But look at what else helped enable it:

• A Cyclotron Road fellowship incubated the idea inside the labs. Cyclotron Road is funded by EERE, not ARPA-E.

• FORGE, the enabling field laboratory infrastructure that made enhanced geothermal credible, took more than a decade of Geothermal Technologies Office funding, selection, and negotiation.

• Early ecosystem partners like Elemental Impact got some of their original funding from EERE and the Office of Naval Research.

• Demonstration support came through the Geothermal Technologies Office.

• And further catalyzing these technologies at true scale requires demonstration funding of the kind the Office of Clean Energy Demonstrations was positioned to provide, or a mechanism like a Drilling Appraisal Risk program like the one we’ve recommended be implemented.

And this is in addition to the support and funding from the private sector that has catalyzed and sustained momentum for Fervo, and the workforce built through both support in federal funding and workforce programs developed by DOE.

Moonshot funding, which ARPA-E styles its funding, was a critical nexus point.

It was never the whole story.

Why “ARPA is the solution” Fails to Recognize a Portfolio Approach

ARPA is the best branded “solution” to supporting innovation within government and so when the question comes up “how can we be more innovative in X problem area” everyone seems to think, “let’s create a DARPA.” The conclusion seems to be that the ARPA model is the critical unlock to all of the world’s hardest hard-tech or energy challenges. But this elicits a certain amount of eye-rolling among those of us who have technology run programs.

Not because the ARPA model is bad. But because it minimizes what commercialization of innovation at scale actually requires: infrastructure, foundational research, workforce development and fellowships, demonstration funding at scale, and capacity in the surrounding innovation ecosystem.

Take a look again at the Fervo story, so much more was needed than just an ARPA-E award. It is neither realistic nor pragmatic to expect ARPA-E to tackle all of that alone. That is exactly why well-matrixed, well-resourced institutions inside DOE, connected outward to agencies like the National Science Foundation and the Department of Commerce’s Economic Development Administration as examples, are necessary conditions for innovation to thrive.

When we assume or hope that a single intervention can solve the real challenge of commercialization of technologies, we are missing the other ingredients that are needed - and have been demonstrated as needs - in order to support the full lifecycle of bringing a technology (or suite of) to market. ARPA-E plays a critical role in catalyzing innovation and has experimented with continued funding through SCALEUP to provide additional capital at the valley of death of pilot demonstration. But those are only some of the pieces that are necessary scaffolding to support startups as they go from idea to market penetration.

Take as an example the way we thought about portfolio buildout within the Water Power Office (though this is broadly true for all technologies). To move technologies forward under a common commercialization framework, there are four critical elements: (1) research and development; (2) labs and testing facilities; (3) startups and entrepreneurs; (4) end users and adoption. All of these are interlocking requirements that are necessary to scale and commercialize technologies.

And then when we were considering our investment strategy, we would ask questions in each of these categories to understand how/if we would proceed with certain investments. But the reality is, along the commercialization continuum, all of these questions need constant redressing. And all four of these quadrants need different mechanisms to support what is considered the enabling, supporting infrastructure - or scaffolding - beyond even just funding in order for technologies to successfully commercialize.

If the ARPA models only invest via grants or cooperative agreements into part of this web of needed infrastructure, then we’re only solving the first part of a continuum - the beginning innovation funding. It leaves a lot of other necessary ingredients for the full commercialization of technologies either at a high cost to entrepreneurs, or more persistent gaps that can’t be addressed by the private sector alone.

We’ll get to portfolio design more in future posts, but this all comes down to why creating an entire series of investments with public sector funding, in partnership with the end users who need the solutions, is needed.

What Institutional Knowledge and Iterative Design Taught EERE

Here is another part that gets missed most often: a lot of the innovation in how to fund and structure innovation has come from outside the DARPA and ARPA models. And there is a structural reason for that. ARPA-E’s staffing model runs on term appointments (set multi-year employment agreements). Brilliant researchers come in from academia and industry, are told to stand up a program, and are handed a limited toolset along with a ticking clock. They essentially copy and paste from the playbook they are given, which is a playbook of grants and cooperative agreements. Term-limited staff also can’t easily translate institutional knowledge into new policy and program design, and they may not know who inside the agency to navigate with.

The report picks up on a telling symptom: ARPA-E has run exactly one prize in 17 years. Not because it can’t. ARPA-E can certainly do prizes. EERE and other offices have used them ubiquitously, as a seeding mechanism, a workforce mechanism, and a way to crowd in entirely new innovators.

One of the important distinctions I drew for the committee is that EERE, for the most part, does not hire on term appointments. Lessons learned in program design are retained in house and iterated on over time by the same staff. That is a big part of why EERE’s toolbox looks the way it does today.

Over two decades, EERE moved from grants-only to cooperative agreements with active program management, then layered in prizes, vouchers, technical assistance, partnership intermediary agreements, and other transactions authority, plus seeding and supporting new models like the Foundation for Energy Security and Innovation.

The Sunshot Incubator alone introduced tiered awards, concept papers, replies to reviewer comments, preselection interviews, and payable milestones negotiated around results rather than effort. The American-Made Network built a system of more than 500 support organizations that measurably increased first-time applicants and the diversity and quality of the applicant pool. In water power, we deliberately built a four-quadrant portfolio: foundational R&D, startups and entrepreneurs, testing infrastructure, and end users and adoption, each with the mechanism best fit for the job.

None of that came from a moonshot mandate. It came from institutional knowledge and regular input from awardees on what they needed to succeed compounding over time, and from people who stayed long enough to learn what worked, fix what didn’t, and hand the playbook to the next person.

This is the necessary other ingredients beyond just a technical challenge outlayed in a moonshot-like setting. The reality is as technologies move from good ideas on paper, or in a lab, into a bench-scale, then pilot-scale, or commercial-scale demonstration you will need a suite of programmatic solutions to support the entirety of the commercialization journey. And those solutions require intentional design - infrastructure buildout and access, as an example, has taken decades of iterative tries to ensure companies can access the expensive infrastructure necessary to validate their designs and systems. And many times these solutions require scaffolding in the form of alternative funding mechanisms, beyond just say access to cash or grants. But understanding how you structure these alternative mechanisms requires institutional design expertise in order to capture the lessons that have been learned on effective technology commercialization management. Simply put, the ARPA model alone is not enough.

Highlighting Alignment with What the Report Recommends

There are a number of recommendations - and very important accolades - for ARPA-E writ large. But I wanted to highlight where the committee saw the same pattern of a gap in program design and portfolio integration in particular. One of its conclusions puts it plainly: “Program design for technology innovation is itself a discipline requiring a specific combination of skill and experience.”

The recommendations I would highlight:

• Innovate on the agency itself. ARPA-E has focused on funding innovative technologies but not on innovation in its own operations, following essentially the same program development procedure for over 15 years. The committee recommends an ongoing focus on operational improvement, with emphasis on more flexible contracting mechanisms that allow different program structures.

• Rethink who becomes a program director, and how they hand off. The committee calls for a cohesive hiring strategy that targets a range of expertise and career stages, along with a real strategy for program director handoffs. I would add my own side note here (and a strongly held biased belief): program directors are usually limited to PhDs, and PhDs are not the only good people to lead technical program design. I’ve seen good people with Masters turned away.

• Build the bridge to the rest of DOE. The committee recommends that Technology-to-Market advisors create targeted programs coordinated with the commercialization-focused offices at DOE, and that Congress grow the SCALEUP program from roughly $40 million to $120 million per year, additive to ARPA-E’s budget, to address the scaling challenges that would otherwise strand promising technologies. This is a good start, and there is a critical gap in what I’d call medium-sized demo ($10-50M) funding, that is inclusive of scaling up the ARPA-E portfolio but not exclusively, as other DOE funded technologies can languish in that same gap.

• Move at the speed the model promises. ARPA-E targets 100 days from selection to award. About 70 percent of projects have taken longer, and nearly 40 percent took more than 150 days. For fast-moving technology areas, that delay is a barrier to entry for exactly the startups ARPA-E most wants. There are lessons to be learned from other parts of DOE on how to speed to scale, including examining other funding mechanisms that would avoid the more time-delayed issues of cooperative agreements.

The handoffs and relationship building are also absolutely critical. Ideas that come out of ARPA-E’s open topics are a great example of why the handoff question matters. Some of those seeds of discovery could have homes in applied offices if we thought specifically about their transition, rather than it being ARPA-E and then nothing. I have worked with many wonderful ARPA-E program directors who made a cognizant reach-out when fields they were shaping moved into ours. But that shouldn’t depend on individual initiative. It requires better linkages and better handoff strategies, and it is a microcosm of the enduring, flexible structures the federal government needs writ large.

A Program Design Center of Excellence

One of the gaps that has been apparent to us is the lack of institutionally-embedded support to evaluate not just the portfolio of work needed, but also a base of understanding inside DOE to support the how work gets funded. So the question is, should DOE stand up a Program Design Center of Excellence? The notion is not that ARPA-E has to hold all of this knowledge in-house. In fact if you’re on a three year appointment, you won’t want to spend time fixing the funding process. It is that the Department should make a specific, structural effort to give really talented term-appointed program directors the tools and support to think differently about allocating funding, drawing on mechanisms that may be unfamiliar to them but can make their work more impactful. Pair the most innovative technology thinkers with people who understand how to get things done inside the federal government. Whether that is a mini office inside ARPA-E or one that connects up broadly across DOE, I’ll let ARPA-E decide. But it’s a persistent gap that we’ve seen.

The Takeaways

Two things can be true at once. First, ARPAs aren’t everything. Any technology that succeeds needs infrastructure, foundational research, workforce training, incubation and ecosystem support, offtakes, and demonstration funding. A whole portfolio approach, not a moonshot in isolation. Second, if we want our most innovative technology thinkers to succeed inside government, we need to pair them with the people and structures that know how to execute there.

ARPA-E has transformed energy innovation. With a broader set of program design tools and mechanisms, it can be more transformative still. The report highlights many of the really great things ARPA-E does, and I encourage everybody to read it.

The National Academies report, Evaluation of ARPA-E’s Mission and Goals, is available at nap.nationalacademies.org/29399. And I encourage everyone to think beyond the ARPA model when we think about the critical unlocks needed for technological innovation, and the need for valuing the how and not just the what for innovation.

AI NOTE

To help us in organizing and disseminating our thoughts and original ideas, we use AI tools. We transcribe our thoughts and do a lot of iterative editing for each and every post. We believe too that using the best new tools for us to continue to innovate is necessary. The reality is that AI is a part of our technological toolkit, and we want to be able to use these tools for maximum efficiency. All of the ideas presented here in our Substack are our own.

When I was a deputy director at the Solar Energy Technologies Office (SETO), one question followed me around more than any other. A company would come in, often after months of lobbying, and ask some version of: Congress set aside $25 million for perovskite solar cell, module, and manufacturing research and development. Why is your funding opportunity so much smaller? Where did the rest go?

The honest answer is that the rest never disappeared. It was already working, spread across lab research and prior commitments that also counted toward that same $25 million. But that answer only makes sense once you understand how a program office turns a topline number into spending decisions. So let me walk you through it, using a real year: FY2023, when SETO’s budget was $318 million.

Almost none of that $318 million was a blank check. By the time it reached the director and me for final calls, most of it was already spoken for, and understanding how it got spoken for tells you more about federal funding than any press release ever will.

Take Away: An office’s topline budget is not its discretionary budget. The gap between the two is where most of the confusion, and most of the disappointment, lives.

This overview is more about how an office thinks about the allocation of a budget based on congressional direction and resource needs. There are numerous other decisions that get made about what to fund – which solicitations, what mechanism, staffing through fellowships – that are not necessarily covered in this post but are additional critical activities undertaken as part of budget scenario planning.

Two People and a Number

Budget season inside the office started the same way every year, with the director, the deputy director (me), and our topline appropriation on the table. For FY2023 that number was $318 million.

That figure arrived carrying scars. The Administration had requested $534.6 million, an 84% increase built largely around a new push into domestic solar manufacturing. Congress enacted $318 million, about 10% over the prior year. Every plan built for the bigger number, and the staff spun up to imagine it, now had to fold back into a budget that looked a lot like the year before. (How an office gets to that number in the first place, and why it spends a year and a half planning against five different versions of it, is the subject of the previous post.)

For clarity, I’m setting aside prior-year carryover and the mechanics of operating under a continuing resolution. Assume a clean $318 million landing on the table on day one. Even under those friendly conditions, watch how quickly it gets claimed.

Subscribe now

Layer One: What Congress Already Decided

The first and largest claim on the budget belongs to Congress, and it arrives in the appropriations report language: the joint explanatory statement (also known as conference language) that accompanies the bill.

Here’s how Congress told SETO to allocate funds among subprograms in FY2023:

● Not less than $60,000,000 for Concentrating Solar Power R&D

● Not less than $77,000,000 for Photovoltaics R&D

● Not less than $45,000,000 for Balance of System Soft Costs

● Not less than $55,000,000 for Systems Integration

● Not less than $70,000,000 for Manufacturing and Competitiveness

Add those five floors together and you get $307 million, or 97% of the $318 million topline, carrying a “not less than” directive tied to a specific subprogram.

That number sounds suffocating, and it’s the single most misunderstood thing about appropriations, so let me be precise: a floor governs distribution, not use. Congress told us how much had to land in each of our five subprograms. It did not tell us specifically what to fund inside them that would roll up to the top lines of each subprogram. Within Concentrating Solar Power R&D’s $60 million, we still decided which programs to run on which topics. The floors shape the funding structure of the office; they don’t hand you a shopping list.

SIDEBAR: WHAT SETO’S SUBPROGRAMS ACTUALLY DO

Congress set a funding floor for each of the office’s five subprograms. Here is what sits behind each name.

Photovoltaics R&D. The solar cells and modules themselves. Work spans efficiency, long-term reliability and durability, and new materials, including silicon, cadmium telluride, and perovskites.

Concentrating Solar Power R&D. Technologies that use mirrors to concentrate sunlight as heat, including collectors, receivers, and the thermal energy storage that lets solar power be dispatched after sunset. DOE also calls this area concentrating solar-thermal power.

Balance of System Soft Costs. The non-hardware costs of a solar project: permitting, inspection, interconnection, siting, financing, customer acquisition, and installation labor. Together these account for a large share of what a customer actually pays.

Systems Integration. Getting solar onto the grid reliably, securely, and affordably. Power electronics, grid services, cybersecurity, and operating a system with high shares of solar generation.

Manufacturing and Competitiveness. Domestic manufacturing and the path from lab to market. Manufacturing processes, supply chain security, support for entrepreneurs and small businesses, and cost and data analysis across the solar value chain.

But then a second layer arrives, and this one does start naming names. Nested inside those subprogram floors, the same statement directs:

● Not less than $5,000,000 for the National Community Solar Partnership (NCSP)

● Not less than $25,000,000 for cadmium telluride (CdTe) R&D and commercialization

● Not less than $25,000,000 for perovskites, including pilot-scale manufacturing

● Up to $40,000,000 to lower barriers to solar adoption for low-income households

● Up to $10,000,000 for research on solar’s impacts on wildlife and ecosystems

One more wrinkle. This isn’t even the only binding document. Report language from the House report, the Senate report, and the conference language all stand unless the final conference language explicitly contravenes them. Senate language that never gets mirrored in the House or the conference still counts as congressional direction. Stack all three reports on top of each other and the total volume of prescription an office wades through can be larger than any single document suggests.

SIDEBAR: THE TWO TECHNOLOGIES CONGRESS NAMED

Both directives point at photovoltaic absorber materials, the layer in a solar cell that captures sunlight.

Cadmium telluride (CdTe). The second most common absorber material (part of a cell that turns light to electrons) in photovoltaics, behind silicon.

Perovskites. A set of next-generation absorber materials that could be cheaper than today’s photovoltaics.

A Primer on Translating Congressional Direction

As a part of the markup of reports and final languages, appropriations staff and congressional members choose a set of terms to indicate the weight and importance of their direction. Understanding how to read these reports requires a translation exercise as these words are carefully and they are not interchangeable:

● “Not less than” (and its twin, “at least”) is a floor. You must spend this much or more. It is a mandate.

● “Up to” is the opposite: a ceiling. It’s permission, not obligation. You may spend this much; you may also spend nothing. Advocates routinely misread “up to $40 million” as $40 million guaranteed. It isn’t. Technically, it could mean nothing (though that would not likely make Congress happy).

● “Directed” means do it. When the statement directs SETO to support pilot manufacturing plants for perovskites, that’s not a suggestion.

● “Encouraged” is a strong hint with room to maneuver. Usefully, it’s often where next year’s funding topics are quietly born.

● “Notes support for” is the softest touch: Congress is signaling it’s watching a particular effort, without requiring anything.

There’s also language that binds without appearing anywhere on the page. Some direction arrives as a shared understanding; everyone involved simply knows what a phrase means. A directive for “competitively selected projects” can carry an unwritten subtext understood by all parties: this money is not for the national labs.

Where the language comes from, and the conversation behind it

Report language doesn’t materialize from nowhere. Before appropriators finalize it, there’s a feedback loop in which DOE is asked to comment on draft language (known as technical assistance), and offices get one or two windows a year to flag what’s workable and what isn’t. Informal relationships matter too: a well-timed conversation could clarify which directives the committee actually cared about, and which were there to satisfy a persistent stakeholder. Knowing the difference changed how much sleep you lost over any given sentence. More on this process can be found in the first post of this series, The DOE Budget - 20 Months Before the Money.

Why the claims don’t simply add up

Now for the part almost nobody outside the building understands. You cannot tally these directives by stacking them: $307 million in floors, plus $25 million CdTe, plus $25 million perovskite, plus $5 million NCSP, and conclude Congress over-programmed the budget. The directives nest and overlap.

A dollar of Photovoltaics R&D money spent on CdTe device performance satisfies both the $77 million floor and the $25 million CdTe directive, simultaneously. One dollar, two requirements met. Similarly, a single project on module durability could count toward CdTe and perovskites at once, and towards either the Photovoltaics R&D or the Manufacturing and Competitiveness budgets, but not both. Even SBIR dollars, the small-business set-aside we’ll get to below, could count toward these prescribed spending levels.

This is why I’ve come to think the right way to understand congressional direction isn’t as subtraction. It doesn’t simply remove dollars from your discretionary pile. It’s a reduction in degrees of freedom. Each directive removes some of your choices about what a dollar can do. Within Photovoltaics R&D’s $77 million, some portion now has to be CdTe-shaped. Within soft costs’ $45 million, some has to be permitting-shaped and some NCSP-shaped. The office’s real job becomes a constrained optimization problem: find the allocation where your dollars satisfy as many overlapping constraints as possible at once. An office that’s good at mapping those overlaps preserves more genuine flexibility than one that treats every directive as a separate, additive pile of money.

SIDEBAR: VIEW FROM ANOTHER OFFICE — When the directives exceed the budget

Solar’s FY2023 language was heavy but arithmetically workable. It can get worse. Take the Water Power Technologies Office, whose budget was $179 million in 2023. For the marine energy portion of the language alone, which had a ceiling of $120 million, the prescribed language stacked to $176 million if you read every directive as a separate pile: testing infrastructure, university marine energy centers (“up to $30 million”), standalone competitive solicitations (“up to $50 million”), a grid-connected test facility with its cost overruns, and a $24 million “Powering the Blue Economy” crosscut. This didn’t include other crosscutting language that would also draw on the marine energy portfolio, including a $5 million directive for aquaculture with wind and wave, and a $40 million “aquatic decarbonization” crosscut.

The only way to reconcile it was overlap-mapping the language and double counting investments. The marine energy program manager kept a running line-by-line ledger, color-coding every project against every directive it could satisfy; one SBIR topic might count toward the competitive-solicitation language, the marine energy centers language, and the Blue Economy crosscut simultaneously. Real mental gymnastics, and at the end of it, not a lot of genuine discretion left over.

Take Away: Congressional direction removes degrees of freedom, not just dollars. Read the report language (all three documents of it) and learn its verbs before you size any opportunity.

Layer Two: The Automatic Deductions

Some money comes off the top before we deliberate about anything, because the law requires it. Two set-asides matter most:

● SBIR and STTR, the Small Business Innovation Research and Small Business Technology Transfer programs, together claim 3.65% (3.2% + 0.45%).

● The Technology Commercialization Fund (TCF) claims 0.9%.

But here’s the nuance that trips people up: these percentages don’t apply to the $318 million topline. They apply to whatever the office declares as RD&D. Roughly, the way we thought about our R&D base was to start from the topline and strip out the things that aren’t research such as the ~$45 million in soft-costs deployment work. That left an R&D base in the neighborhood of ~$273 million.

On that base, SBIR/STTR runs about $10 million and TCF about $2.5 million, roughly $12.5 million off the top. Had we applied the same percentages to the full $318 million, the bill would have been closer to $14 million. The difference is real money, and it flows directly from where the office chooses to draw the RD&D line.

The office drew that line itself. SETO leadership made the R&D delineation call each year; the Office of Energy Efficiency and Renewable Energy (EERE) above us would double-check it but rarely had input. So, the set-aside isn’t a fixed tax handed down from on high. It’s partly a function of how the office categorizes its own portfolio.

Two more things about SBIR are worth knowing. First, as noted above, SBIR dollars can still count toward Congress’s prescribed spending levels; the set-aside and the directives overlap too. Second, SBIR is what my colleague calls a secret mortgage: Phase I awards create expectations of Phase II follow-on funding, so an office that isn’t careful about its Phase I/Phase II pipeline discovers it has quietly committed more projects than it can support in future fiscal year, without sacrificing the next year and so forth.

(These figures are my reconstruction of how we approached the calculation in simplified terms, not published line items. Treat the specific dollar amounts as illustrative of the method rather than audited numbers.)

Layer Three: Overhead and Promises Kept

Overhead: the cost of running the office

Offices pay to operate out of their own program dollars, and the honest frame for this is operational and implementation support: the contractors and fellows who support federal staff to keep programs running and any needed IT or system investments (like a new program management platform). At SETO this ran roughly $15 to $20 million a year. I want to be careful with this one, because it’s easy to read as fat. It isn’t. Federal staff salaries come from a separate, chronically undersized “program direction” line at the larger office (i.e. EERE) level, and offices are not given enough of it to hire the federal employees the workload demands. Contractor and fellow spending is a capacity expansion tool made necessary by that gap, not a preference. Nobody is arguing it should go to zero. Fellows in particular are a workforce pipeline, and some flex capacity is healthy. But the balance should sit much further toward federal program direction than it does.

A recurring irritant is IT and data tools. Offices end up buying bespoke portfolio dashboards and management databases out of discretionary program funds, tools that should be universal DOE platforms built once. Both my office and my colleagues paid a major consultancy for portfolio-dashboard work that delivered, essentially, a report and a request for more money. That’s a future post of its own.

On top of office overhead sits the parent organization. EERE could assess its offices for overarching needs, sometimes framed as directly as “I need six percent of your budget for….[whatever the hot crosscutting topics of the day were].” We’d hold funds against that. If they went unused by year-end, we’d redirect them into a program we already had running. This could be for priorities, or to satisfy crosscuts called out by Congress (although truthfully, we referred to most crosscuts as bean counting exercises, not truly funded and supported efforts).

SIDEBAR: VIEW FROM ANOTHER OFFICE — The internal office tax

What happens when an office needs to fund an activity Congress never gave it a line for? WPTO had no tech-to-market or communications line in its appropriation. That office’s solution as a matter of directorial policy was an internal 10% carve-out from both the hydropower and marine energy budgets, dedicated to commercialization, analysis, communications, and stakeholder engagement, imposed by directive rather than negotiation. Before the carve-out, the commercialization lead had to go program manager to program manager asking for money.

The broader argument, which I share: every office needs a protected set-aside (call it 10%) for cross-cutting activities that are essential to executing the mission but absent from the appropriations structure. If Congress doesn’t fund the connective tissue, offices must build it themselves.

Promises kept: the funding the community would miss if it vanished

The national labs sat first in this category. The preponderance of lab funding ran on long-term annual operating plans, on three-to-five-year cycles, because labs staff up against expected funding and short-cycle uncertainty drives attrition and hurts the ability to sustain subject-matter experts in areas of national interest. What makes this funding unique is that the multi-year commitments to labs are mortgaged, funded year by year against a standing promise, and mortgaging was generally only allowed for lab work. Everything else, university awards and industry awards alike, had to be funded in full, up front, in the year of selection. That asymmetry quietly shapes the whole portfolio: a $9 million, three-year industry project consumes $9 million of this year’s budget, while the equivalent lab commitment consumes $3 million a year. While some projects are just a couple of years in duration, DOE would also provide funds to sustain long term capabilities and knowledge. For instance, SETO would fund R&D and associated facilities at NREL (today NLR) to further solar cell and solar module research.

Program continuity came next. Rather than re-run the same solicitation annually, we rotated topics on a multi-year cycle. Our Photovoltaics R&D team might run cell efficiency R&D one year, module durability R&D the next, field performance and supply chain the third: one per year instead of all three every year. This made the work easier to message to the community and, crucially, gave applicants certainty, as the projects awarded would run 3 years in duration. When one round closed, they knew a related opportunity would come around again to carry their work forward. Breaking that continuity risks losing research momentum, especially for early-stage performers at labs and universities.

SIDEBAR: VIEW FROM ANOTHER OFFICE — Mortgaging as strategy, not just constraint

WPTO was working to extend the mortgage logic beyond the labs before Jenn Garson left. Her wave-energy device program illustrates why. Budgeted one year at a time at ~$30 million, every stage-gate decision about which devices to advance had to fit inside a single year’s envelope. Framed instead as ~$90 million over three years, with the appropriators’ blessing (which she obtained), the office could run a genuine multi-year pipeline and make better advancement decisions. Same dollars, better portfolio management. But ultimately, the solicitation was caught in the transition and was never awarded, which tells you something about how fragile good budget practice is.

This is where the perovskite mystery from the opening resolves: the “missing” millions weren’t missing. They were here, in ongoing lab durability work, standing multi-year commitments, SBIR projects, and continuity programs that also counted toward the $25 million perovskite directive.

Take Away: Multi-year commitments are the load-bearing walls of an office budget. They’re why the community gets stability, and why so little of any year’s money is truly up for grabs.

The Decision: How the Rest Gets Spent

Only now, after all of that, do we get to the part everyone imagines budget season is about: deciding what to actually fund.

The mechanism was simpler and less dramatic than outsiders tend to picture. Each of our five program managers (Photovoltaics R&D, Concentrating Solar Power R&D, Systems Integration, Balance of System Soft Costs, and Manufacturing and Competitiveness) brought proposals for how to use their portion of the budget. They’d account for their directives, their standing lab commitments, and where they sat in the rotating cycle, then make their case for what to fund with what remained. The director and I listened, weighed the pitches, and made the final calls.

People expect budget allocation to be a knife fight, with teams clawing at each other for dollars, but fortunately for me it wasn’t. Program managers proposed; leadership decided. There were strong cases and there were disappointments, but the constraint on new funding was never internal politics. It was the stack of prior claims we’d already walked through: floors, directives, set-asides, overhead, commitments, each one narrowing what was left before anyone in that room made a single discretionary choice.

The Punchline: Follow One Directive All the Way Down

Let’s answer the question we started with by following the money. Congress said: not less than $25 million for perovskites, and not less than $25 million for CdTe. That’s $50 million in combined direction. And you don’t have to take my word for what happened next, because DOE documented it publicly.

In September 2023, SETO released the Advancing U.S. Thin-Film Solar Photovoltaics funding opportunity announcement, or FOA (DE-FOA-0003058): $36 million total, with up to $20 million for perovskite projects and up to $16 million for CdTe. The notice of intent said the quiet part out loud: the FOA would support SETO in meeting the $50 million of congressional direction. Support. Not constitute.

So where was the other $14 million? Exactly where this post has been pointing all along:

● Already committed: multi-year perovskite and CdTe work at the national labs, running on standing three-year plans that predate this budget year. Counts toward the directive.

● In adjacent ongoing programs: module durability research, for instance, that legitimately counts toward perovskites (and sometimes toward CdTe at the same time). Counts toward the directive.

● Through SBIR: small-business thin-film topics funded from the set-aside. Counts toward the directive.

Run the arithmetic from the applicant’s side. A company that lobbied for the perovskite direction saw $25 million become a $20 million competitive topic, and a CdTe company saw $25 million become $16 million. Every dollar of the $50 million was real, and every dollar went where Congress pointed it. But only 72% of it arrived through the open front door, and applicants competed for that share against everyone else who read the same press release. The advocate who fought for $25 million and the applicant who saw a smaller opportunity were both looking at true numbers. They were just measuring different things.

(One last mechanic hiding in the dates: that “FY2023” opportunity was released September 12, 2023, the final month of the fiscal year, and selections weren’t announced until May 2024. DOE’s money doesn’t expire at year-end, so a fiscal year’s budget keeps landing in the world long after the year is over. File that away for the timing discussion in a future post.)

Take Aways

● A topline is not a discretionary budget. Read the report language before you size an opportunity, and read all three documents: House, Senate, and conference language all stand unless explicitly contravened.

● Congressional direction removes degrees of freedom, not just dollars. One dollar can satisfy two overlapping directives, which is exactly why the “set-aside” and the “opportunity” are different numbers.

● Learn the verbs. “Not less than” is a floor; “up to” is a ceiling. They are not the same, and confusing them will mislead your whole strategy.

● Multi-year structure is the hidden hand. Lab mortgages, rotating cycles, and SBIR tails commit money before the year begins. If your topic isn’t in this year’s solicitation, it may be next year’s; plan your timing accordingly.

● The residual is decided by a few senior people. Proposal quality wins, not lobbying volume, and the director’s truly free reserve is about 2%, not 70%.

Innovation Waypoints is brought to you by Waypoint Strategy Group.

The way most people understand federal funding is simple enough. Congress passes a budget, the agency spends it. That description is not wrong, but it covers a slice of the process that can run for upwards of two or more years.

By the time the Solar Energy Technologies Office (SETO) knew what it had for fiscal year 2023, we had been planning against that year since the spring of 2021, across five mutually exclusive budget scenarios of what the number might be. Four of them were wrong. We did not find out which four until well into the first quarter of the FY23 fiscal year (December 23, 2022), which was an early year given that typically budgets these days pass well into March or halfway into a budget year.

But this process is a long, beleaguered process, and one that is filled with far more anticipatory planning and direction shifts than can appear on the outside. To illustrate how long and drawn out, and quite frankly complicated, the process is, this post walks through one real fiscal year from first guidance for budget preparation on the agency side of the house to signed bill. Our next post (Where does the money actually go?) picks up where this one ends, with what an office actually does with the money once it finally knows the number. Hopefully by shining a light on how long and arduous this process, you can understand why federal budgeting is not as easy (or straightforward) as our Constitution lays out.

This post is meant to help illustrate the budget office to people and organizations outside of the executive and legislative branches, but might also be helpful for people new into budget planning cycles within government.

The Timeline, Using FY2023

Fiscal year 2023 began on October 1, 2022 (the federal budget cycle is October 1 to September 30) but work on it began seventeen months earlier.

One detail makes this harder than the list suggests. These cycles overlap. In the summer of 2022, while waiting on FY2023 and executing FY2022, offices were already building the FY2024 request. At any given moment a program office is spending one year’s money, defending a second year’s request, and inventing a third year’s plan. That is the normal condition, not a bad year.

Five Budgets, One Office

By late summer, an office is planning against five outcomes at once: the President’s request, the House mark, the Senate mark, a notional House and Senate compromise, and a continuing resolution.

These are not five versions of one plan. Each carries a different dollar figure and a different set of directions about what to spend it on, so they describe five different programs. A colleague who ran another EERE office put the practical effect well: you are constantly moving your floor while never having any visibility into your ceiling. The floor is the lowest mark.

Take Away: If you want to know what an office’s funding capacity will be, watch the lowest congressional mark, not the President’s request.

How the President’s Request Stopped Being the Anchor

The request is still the formal starting point, and offices put real work into it. The trouble is what gets asked. When an administration wants to know what you would do with $500 million against a $290 million base, “more of what we already do” is not an acceptable answer, particularly if your office is aligned with the priorities of that current Administration. The idea is that the intent behind this document shows how a President prioritizes budgeting, often in line with any recent White House Office of Science and Technology Policy R&D priorities. What this means in practice is that the ask is for new programs, so staff get spun up to design them.

Then Congress declines to buy the vision, often because the detail behind it was never granular enough to be persuasive or the request was so high to be fiscally unviable, and the office goes back to roughly what it has always done with some changes to respond to the needs of the technology area they support. Everyone who spent months imagining the larger program absorbs that. It is worth naming plainly, because the morale cost of aspirational budgeting is real and rarely discussed. And often the visibility Congress gets is not an inaccurate reflection - most often the Request briefings to Congress is limited to maybe 10-20 minutes per office for these requests, so Congressional staff never really get enough insight into any seriousness or weight put behind these figures.

The opposite extreme demands a different exercise. When a request proposes deep cuts, an office has to tell its staff what would survive, and the honest answer usually amounts to keeping the lights on at the national laboratories and keeping researchers employed. Even if you’re a seasoned government executive, the numbers if they’re slashed can have a deeply demoralizing resonance of the staff because their perception is - the work you’re doing doesn’t matter. But again, these budget have now been reduced to a messaging exercise for the most part, so career staff will try to reassure their own colleagues that the dramatic slash will not come to pass. And ultimately, you run the exercise knowing the outcome will land somewhere in the middle of these extremes.

Both extremes have the same practical effect. You end up discarding them the way you would discard outliers in a statistical analysis, and over time the President’s request has become the outlier rather than the center of the distribution. That is a strange place for a formal budget request to occupy. The current administration has taken the opposite approach, treating the request as the true spending plan, which is a genuine departure from recent practice and changes what offices can plan against, and in contrast to how the budgetary powers were originally constructed in the Constitution.

When Only One Chamber Speaks

As Congress goes through its budgetary process, two documents are produced. The bill text, which generally has the numbers associated for each office. And then an accompanying report. The House and Senate each have their own report, and then if a conference passes, another report is released alongside the bill text. And report language is where Congress tells an office what to do with the money, and offices are generally told to treat all of it as binding. Unless the final conference language contravenes something in the House or Senate report, it stands. Silence in the other documents does not void a directive. If the Senate report says to spend $5 million on something and the House and the final statement never mention it, the office still owns that instruction.

Which raises a question with no clean answer. FY2023 is a useful case. The House passed its Energy and Water bill on the floor in July 2022 as part of a six-bill package, adopted 220 to 207. The Senate never marked up its version at all; the Appropriations chair released Senate bills without committee action. So the two sets of direction reached the finish line through very different levels of legislative process, and neither the enacted omnibus nor the practice inside agencies draws much distinction between them.

It gets thinner still. Report language that clears a committee but never reaches a floor vote is, on the standard reading, direction an office is expected to follow. Whether it should be is a fair question to put to appropriators, and there is a practical remedy that more offices should use: through the Office of Congressional Affairs call the committee staff and ask whether they meant it. A short conversation resolves a surprising amount of internal hand-wringing.

Take Away: Report language binds even when only one chamber wrote it. Read the House and Senate reports, not just the final statement.

Where the Language Comes From

None of this language materializes on its own. Before appropriators finalize it, DOE is asked to comment on drafts, a process known internally as technical assistance, and offices get one or two windows a year to flag what is workable and what is not. Informal relationships fill the gaps. A well-timed conversation could tell you which directives the committee genuinely cared about and which were there to satisfy a persistent stakeholder, and knowing the difference changed how much attention any given sentence deserved.

VIEW FROM ANOTHER OFFICE — How language gets planted

In the Water Power Technologies Office, language that reappeared year after year: roughly $5 million for a university-led project on floating wind and wave power for aquaculture. Nobody in the office could make technical sense of it. You generally do not want floating wind and wave devices co-located, and $5 million does not demonstrate anything in the water. The mystery was resolved when the office traced it to a lobbyist whose clients had a prospective test site and an interest in aquaculture.

The lesson cuts two ways. Lobbyists do plant language, which can be helpful and important prompts for direction that offices can/should take based on stakeholder input and equities. But the deeper dysfunction is that stakeholders often feel they have no other meaningful channel for telling a program what it should work on, so they route ideas through appropriations staff instead of talking to the office, which would happily have taken the meeting.

Take Away: If you want a program to work on something very specific, talk to the program. Routing an idea through appropriations is slower, blunter, and often produces direction nobody can execute well. The exception would be Congressionally Directed Projects - or CDPs - which are discrete carveouts for individual projects.

What This Buys You

The point of walking through all of this is not to complain about the process. It is to explain why an office’s public number tells you so little. By the time SETO had $318 million in hand, that figure had survived seventeen months, one OMB negotiation, two chambers, five planning scenarios, and a three-month delay past the start of the fiscal year. It arrived with instructions attached and most of its flexibility already spent.

What happens to it next, and why a $25 million congressional directive does not produce a $25 million funding opportunity, is the subject of the follow on post (Where does the money actually go?).

Take Aways

● The number is the end of a process, not the start of one. Twenty months of planning precede the figure you read in a press release.

● Offices plan against five budgets at once. Watch the lowest congressional mark if you want to know what an office can actually fund.

● Report language binds even when only one chamber wrote it. Read the House and Senate reports alongside the final statement.

● Talk to the program, not the appropriator. Language planted through lobbying often produces directives that are difficult to execute well.

Innovation Waypoints is brought to you by Waypoint Strategy Group.

Subscribe now

If you make a breakthrough in the lab tomorrow that could revolutionize a hard-tech industry—metals, power generation, transmission, industrial processes—how long before it starts changing the energy system at scale?

The honest answer, based on historical evidence and firsthand experience with DOE program planning, company roadmaps, and venture and incubator claims, is probably somewhere between 20 and 40 years. This is not a popular number. It is, however, a well-supported one. Even with an accelerating pace of innovation due to AI and other approaches, there are inherent characteristics of physical infrastructure that limit how fast change can happen. And the gap between this reality and the timelines that dominate public conversation about energy innovation is large enough to be genuinely damaging.

This post is about that gap. The time axis is routinely missing from or misrepresented in conversations about breakthrough energy technologies, which distorts investment decisions, warps public expectations, and crowds out groups with impactful technology who are open about how long the work could actually take.

Not a New Issue or Observation

The observation that transformative technologies take a long time to reach widespread impact is not new. Several frameworks have tried to capture and communicate it.

Bill Buxton, the computer scientist and longtime researcher at Microsoft, coined the term “the Long Nose of Innovation” in a 2008 BusinessWeek article. The concept is a mirror image of Chris Anderson’s “long tail”: the bulk of innovation behind any breakthrough moment is low-amplitude work that takes place over a long period before the technology reaches a tipping point. Buxton’s canonical example is the computer mouse, which took roughly 30 years from first demonstration to mainstream adoption. Buxton argues that innovation “is not about invention,” and that the real work lies in an idea’s “augmentation and refinement” over decades that makes something actually usable at scale.1

The Gartner Hype Cycle, probably the most widely recognized framework in this space, maps a technology’s trajectory through five phases: an innovation trigger, a peak of inflated expectations, a trough of disillusionment, a slope of enlightenment, and a plateau of productivity. Gartner now publishes hype cycles specific to energy, including for disruptive energy technologies and low-carbon power generation.2 The framework is useful for illustrating the emotional and financial arc that accompanies technology development—the surge of enthusiasm followed by the crash of reality followed by the slow grind of actual progress. But the hype cycle’s x-axis is intentionally vague about time. Gartner notes that it “often takes between three and five years for an innovation to move through the Hype Cycle,” but that estimate is calibrated to information technology and consumer products. For energy infrastructure, the actual durations are far longer.

The most rigorous empirical treatment of this question comes from a 2018 study by Gross et al. in Energy Policy, which examined historical case studies of energy supply and end-use technologies from invention to widespread commercialization. The findings are striking: the timescales ranged from 20 to almost 70 years. The authors were direct about the implications: “the time needed for new technologies to achieve widespread deployment should not be overlooked, and that innovation policy should focus on accelerating the deployment of existing technologies as well as research into new ones.” The study also found that social and regulatory factors were at least as important as technical factors in determining the pace of deployment, and that long periods of stagnation could precede rapid scale-up when conditions aligned.3

These frameworks were developed primarily in the contexts of consumer products, information technology, and energy systems broadly. But the dynamics they describe are, if anything, more pronounced for the kinds of large-scale, capital-intensive, infrastructure-dependent technologies that dominate the current energy innovation landscape.

A More Realistic Progression

Based on experience designing and overseeing DOE programs, advising companies, and working with the investment community, here is a more granular view of what the progression from lab breakthrough to system-level change actually looks like for hard energy technologies. This is not a description of what it takes to reach viability and profitability as a company (which is a different goal on a different timescale), but of what it takes to change the energy system at scale.

A few caveats. This is rarely a fully linear process. Cross-pollination happens, speed bumps happen, learning cycles are needed and exogenous factors (changes to policies, geopolitics, global events, etc.) impact progress and trajectory. Some technologies and solutions will need to repeat steps, some few will be able to skip steps. This is a simplified representation that assumes things work at each step, which they often do not.

Building confidence in the breakthrough (2–3 years). The initial work involves generating enough data to feel confident spinning up a company or licensing the IP for commercialization. Funding typically comes from angels, friends and family, and DOE applied research offices. The work involves national labs and academic collaborators, and the goal is to move from scientific observation to reproducible results.

Proof of concept and company formation (2–6 years). The next stage requires generating a proof-of-concept piece of hardware or process and demonstrating that the team can operate as an organization outside of academia or a lab. This is Series A and B territory, often supplemented by DOE applied office funding. The wide range reflects the learning cycles inherent in moving from bench-scale results to something that begins to look and function like a product. Building confidence in the case for a larger demonstration is the primary objective.

Pilot facility (3–5 years). Designing, building, and operating a representative pilot facility generates the operational performance and cost data needed to justify larger investment. The timeline breaks down roughly as: one year for design and project development, one to two years for construction and commissioning, and one to two years of operation to generate representative data. Series C and D funding and DOE applied offices are the typical capital sources.

First-of-a-kind demonstration (5–8 years). The FOAK demonstration is where institutional investor confidence is either built or broken. The purpose is to prove that subsequent deployments can be financed using established processes. Funding mechanisms at this stage include DOE’s Office of Clean Energy Demonstrations, the Office of Energy Dominance Financing (EDF), tax policy, and an evolving set of other tools. The extended timeline reflects the reality of large-scale project development, permitting, construction, commissioning, and operation.

First generation of commercial deployments (5–8 years). Even after a successful demonstration, the first generation of commercial projects takes years to develop, finance, build, and operate. EDF, tax policy, and institutional finance are the primary capital sources. These projects generate the track record that underpins the next wave.

Second generation of deployments (5–8 years). This is where you start to actually see the system change and the new solution begin to displace incumbent technologies. The capital stack shifts toward institutional finance and tax policy, with less need for direct government support.

Total: 22–38 years.

This highlights the criticality of patient and non-dilutive capital. No venture fund operating on a 10-year close-end cycle and no institutional lender applying standard risk frameworks is going to accept this combination of time horizon and risk across the full progression. The Gaddy, Sivaram, and O’Sullivan study at MIT documented this mismatch definitively: of $25 billion that venture capital firms invested in cleantech startups from 2006 to 2011, more than half was lost. The companies developing new materials, hardware, chemicals, or processes were the worst performers precisely because they “required significant capital, had long development timelines, were uncompetitive in commodity markets, and were unable to attract corporate acquirers.” The VC model, built for software’s capital-light, fast-iteration dynamics, was structurally mismatched to the realities of hard energy technology.4

Some of the investment ecosystem has adapted since then. Breakthrough Energy Ventures operates on 20-year cycles and MIT’s The Engine fund expects 12- to 18-year return horizons.5 Corporate venture capital programs are stepping in with longer time horizons and strategic rationale beyond pure financial return. But the structural mismatch between the pace of hard-tech energy innovation and the patience of most available capital remains a defining challenge.

Why This Matters

There are several reasons the missing time axis is more than an academic concern.

It crowds out honest actors. When companies and programs claim timelines that are unrealistic given the actual progression required, they attract capital and attention at the expense of groups with potentially more impactful technology who are realistic about what the work entails. A company that says it will go from pilot to commercial scale in three years will look more attractive to an impatient investor than one that says it will take eight, even if the latter’s assessment is far more credible.

It risks hype bubbles that crash funding to valuable sectors. The cleantech 1.0 experience is instructive. The influx of venture capital based on unrealistic timeline expectations, followed by the inevitable reckoning, didn’t just hurt individual investors. It poisoned the well for an entire sector for nearly a decade. Funding dried up for companies that genuinely needed it and might have been on viable paths with more patient capital. Domestic forward progress slowed, and other nations with a more patient and strategic outlook gained competitive advantage. The risk of a similar dynamic playing out in fusion, direct air capture, or other sectors currently attracting significant investment is real.

It disillusions the public and policymakers about the value of government investment. When technologies fail to deliver on compressed timelines that were never realistic, the conclusion drawn is often that the investment was wasted or that the technology doesn’t work, rather than that the timeline was wrong. This erodes support for the kind of sustained, patient public investment that is most critical for technologies at this stage.

A Reality Check on Common Target Dates

A quick example - many groups, both domestically and internationally, have set targets around emissions and sector transitions to avoid and/or adapt to the worst impacts of climate change. These are commonly placed at 2030 and 2050. The dates themselves are artifacts—round numbers that are easy to remember and build policy around. But the relative time horizons matter.

None of the technologies currently in the lab-to-pilot-to-demonstration pipeline will start materially changing the energy landscape by 2030. That is four years away, and even the most advanced of these technologies are still working through steps two through four of the progression above. If pursuing 2030 goals, they must be achievable with established technologies with at least moderate deployed capacities. Even the 2050 horizon is tighter than it appears: achieving substantive, system-level change driven by technologies that are currently at the proof-of-concept or pilot stage by 2050 requires maintaining momentum continuously from now through then. Every year of stalled funding, every policy reversal, every hype-driven collapse in investor confidence pushes the actual impact date further out.

Side note: there is another, separate reason these horizons are longer than they appear, and it has nothing to do with the pace of innovation itself. Energy assets are among the longest-lived capital in the economy—refineries, pipelines, and transmission infrastructure routinely operate for fifty years or more, nuclear plants are now licensed to sixty and eighty, and the civil works behind large hydro can run past a century. An incumbent facility built the year before a breakthrough reaches commercial readiness represents decades of remaining useful life and, just as important, decades of expected return that its owners and lenders have already underwritten. Retiring it early means stranding capital, and capital that has been committed does not strand quietly. The result is that even a technically superior and fully financeable replacement can find the pathway effectively locked in from an investment standpoint—not because the new solution is unready, but because the old one is not yet done paying for itself. This adds a displacement lag on top of the development progression, and it is uneven: technologies that can slot into existing infrastructure or replace assets already near end-of-life move faster, while those meant to replace recently built, long-lived plants face a queue set by depreciation schedules rather than by engineering. When judging whether a 2050 target is reachable, the question is not only how long the technology takes to mature, but how long the thing it must replace still has left to run.

This is not an argument against investing in these technologies. The opposite. The progression described above makes the case that sustained, patient investment starting now is the only way to have these solutions available when they are needed at scale. The argument is for honesty about timelines and for structuring investment, policy, and expectations accordingly.

The Accountability Question

Anyone who claims they are going to bring a new, large-scale technology (not a drop-in component, sensor, or incremental process improvement) from lab to large-scale market impact in five years needs to explain, specifically, why the progression described above does not apply to them. Which steps are they skipping? What is different about their technology, their market, their regulatory environment, or their capital structure that compresses the timeline by 75% or more? There are occasionally legitimate answers to these questions. A technology that can leverage existing manufacturing infrastructure, supply chains, and regulatory frameworks can move faster than one that requires all of these to be built from scratch. But these cases are the exception, and the burden of proof should be on the claimant.

For program designers, investors, and policymakers, the practical implications are clear. Design programs and investment vehicles around realistic timelines, not aspirational ones. Build in the flexibility to sustain support across political cycles and market fluctuations. Structure evaluation and success metrics around the actual progression, not around what you wish the progression were. And when you encounter a timeline claim that seems too good to be true, ask hard questions about which steps are being assumed away.

A realistic time axis belongs in every conversation about energy innovation. Putting it there will make some conversations more uncomfortable but all of them more productive.

Innovation Waypoints is brought to you by Waypoint Strategy Group.

The U.S. Department of Energy funds the seventeen national laboratories to the tune of several billion dollars every year. Some are single-purpose focused on key areas like particle accelerators and nuclear waste remediation, while many are multi-purpose research institutions — NLR, PNNL, ORNL, Sandia, Argonne, Idaho National Lab, to name a few.

Across the seventeen, they sit at the nexus of decades of American scientific and energy research. By the standards of any other government’s research apparatus you’d care to compare them to, the labs are one of the country’s longtime competitive advantages, going back to the Manhattan Project and spurring transformational technologies from batteries to solar to satellites and even the Human Genome Project.

But the way we manage them has been a pendulum swing. Every administration that comes in has a view on how the labs should operate, and each view, even when reasonable, leaves a mark.

Take one data point.

In the first Obama Administration, there was a push to make the labs compete more like universities. Prior to the Administration, the labs were not expected to compete for funding alongside other potential recipients; scopes of work were developed with the labs and were not in comparison (broadly) to other research institutions.

There is a real argument for competition, but there are complexities specific to the labs and how they operate that do not make for a clean comparison between universities and the labs, like how complicated it actually is to hire people at the labs, how their funding structures work, and how the work that happens there is shaped by institutional context. One downstream consequence, intended or not, was that there was more of a push to do less multi-year, flexible funding for lab researchers to explore concepts on their own.

That trend has compounded.

Layered on top of it, a broader trend in how we over-manage R&D — quarterly reports, milestone reviews, annual renegotiation, top-down direction from agency headquarters — has migrated across to how we manage the national labs. Lab work is increasingly directed work. It’s not that it shouldn’t be; directed work is most of what the labs should do. But directed work has become the default, and the share of lab funding that is genuinely flexible has shrunk to something close to a rounding error.

Meanwhile, a body of literature on innovation has been telling us that this is the wrong direction. Bell Labs is the canonical reference with its physical co-location, generous unstructured time, and interdisciplinary collisions.

Steven Johnson’s Where Good Ideas Come From makes the case for serendipity as an engine of breakthroughs. Hunches that take a decade to ripen, accidents, useful collisions between fields that don’t ordinarily talk. None of which is a great match for a quarterly milestone reporting regime.

And Ezra Klein and Derek Thompson’s book Abundance captured a second emerging theme - our researchers are spending way too much time on applying for funding and monitoring their funding. Their example was NIH researchers estimated more than 40% of their time on this paperwork. And it’s a good bet that the DOE labs and DOE awardees in general spend about the same percentage of their time on the same things.

This is the first post in a series about reducing barriers and bureaucratic burdens on researchers.

But what this details isn’t just staring at the problem; this is also a walk through for an example program we built at the Water Power Technologies Office (WPTO) called Seedlings, with a follow-on tier called Saplings.

And to make a case that I’ll keep returning to in subsequent posts: there has to be at least some share of funding inside every R&D office, including funding that flows to the national labs, that gets out of the researchers’ way. And get back to what we should be funding these groups to do - to let researchers be researchers.

Operating Regime Challenges

When I joined the water power technologies office (WPTO), two things were quickly apparent.

The first was that WPTO used the national labs differently than several other offices. The Solar Energy Technologies Office and others were funding technology development at the labs, letting lab researchers build and refine actual hardware. WPTO leaned much further toward software, modeling, and analysis. There was solid annual work being done, but not a whole lot of building things.

The second was endemic across all offices; the structure of the annual operating plan, aka the contract between DOE and the labs on individual projects. Each lab task was prescribed in considerable detail. Each task came with monthly reports, quarterly reports, and a renegotiation cycle. While our staff is and was incredibly diligent on managing all forms of awards, I was never quite sure who was reading all of those reports. The honest answer, I think, is “almost nobody at the level that would justify the cost of producing them.”

There is a related problem here. When the labs are mostly working on software and analysis rather than building things, and when their work is heavily prescribed by the program office, you end up with a structural mismatch. In one line of funding to the labs we were asking the labs to support startups we’d partnered them with — to improve power electronics for a wave energy converter, to help work out mooring and anchoring for a tidal device — at the same time as we weren’t really letting them develop technology of their own through other funding. The implicit logic was a little strange.

We were asking the labs to be expert advisors on hardware they hadn’t been given the chance to build themselves because of concern of creating competition between the labs and the private sector.

Shifting the Incentive Structure

The phrase that sits in the title of this post came from a researcher who helped me rethink how we were structuring - or rather not - opportunities for serendipity and discovery.

I was on a work trip to Hawaii (yes, being in water power has considerable travel advantages to beautiful places on the ocean), sitting with a group from NREL (now referred to as the National Lab of the Rockies or NLR) and PNNL. We were discussing some of the lab work, including supporting prizes (we were there to launch a new prize with NOAA).

In the middle of the conversation, one of the researchers put it plainly: we weren’t letting researchers be researchers.

That phrase covered three specific failures.

1. Over-direction — the work was so tightly prescribed that the discovery piece had been written out of it.

2. Lab researchers were being asked to advise on technology they themselves hadn’t been allowed to develop, which limited what they could plausibly contribute.

3. DOE was putting more emphasis on technology commercialization over discovery so we were also asking researchers to act or think like entrepreneurs when that is not their core competency and nor should it be, unless they explicitly want to.

A fourth failure surfaced in the same conversation, and it deserves its own example.

WPTO had recently launched a successful multi-year wave-powered desalination prize, called Waves to Water. It turned into one of the more visible programs the office produced. And it was the first big investment into the Powering the Blue Economy initiative, which was evaluating alternative markets outside of grid-scale power for marine energy.

But the origin of the idea — the original spark that “wave power plus desalination might be a thing” — wasn’t a strategic decision made at headquarters. It came from a former DOE staffer who had given a lab researcher about $35,000 to explore whether the combination was interesting.

If we weren’t seeding hunches like that anymore we were going to keep missing the offshoots that turn into programs. There were no mechanisms in use by the office that allowed funding the kind of small bets that could help us continue to look forward and innovate in new ways.

And there was one other less obvious but also important missing piece for the labs.

A separate, well-documented problem in the lab system, covered in several pieces of research over the years, is that project ownership has been structurally biased toward senior principal investigators - or PIs.

Postdocs and early-career researchers come into the labs full of ideas, but if they can’t own a project — be the PI, hold the budget, set the direction — they aren’t building the skills they will need to run larger projects later in their career. We were producing a pipeline problem for talent, including incentivizing them to stay at the National Labs.

So I asked the team in that conversation: how much money would it take to design a discovery fund of some kind for the labs? I couldn’t carve out a lot. The implied risk was that if we weren’t overseeing it and were specifically funding off-roadmap work, something might go sideways and the lack of oversight would become an issue from a resource risk allocation perspective. But could we carve out a few percentage points of the office’s budget?

The answer was: probably fifty to a hundred thousand dollars per project. Which is, almost exactly, how we ended up sizing the Seedlings program.

What We Actually Built

I’d already spent a few years experimenting with different ways to fund the labs that opened the labs to more collaboration with the private sector - through vouchers and prizes - but hadn’t done as much on program design thinking on how we could create ways to administer funding that didn’t require us to either direct everything or burden the work with paperwork nobody read.

But putting the same philosophy to work - reducing bureaucratic barriers and letting new innovation thrive - was how we approached this new program. The design that came out of that conversation, and the months of internal negotiation that followed, was the Seedling program.

A few aspects are worth highlighting:

Award size: $50,000 to $100,000, to a PI at one of six national labs (NREL, PNNL, ORNL, Sandia, Argonne, INL).

Period of performance: nine to twelve months.

Application: three pages.

Scoring: a +1/0/-1 scale, evaluated by the WPTO program team with a discussion of the proposal in the room, often accompanied by a 15-minute live pitch.

Topic Areas: open topic permitted — half the calls go out under specific programmatic priorities; the other half permit researchers to propose under any topic with no restriction on research area.

Reporting: a single final deliverable. That’s it. No monthly reports. No quarterly reports. No headquarters check-ins.

Failure expectation: we wrote into the design that we expected up to 70% of Seedling projects would not move forward into a follow-on award. Failure here is not a verdict — it is the point. Let’s learn early that new approaches wouldn’t work, with low dollars instead of high millions+ dollar projects coming to that same conclusion.

To make this possible without inventing a new authority, we leaned on a mechanism that we had only used once in support of lab commercialization programs called the Request for Innovation (RFIn).

The RFI had been on the books for years and used only occasionally. It is what makes the open topic and the lighter application legally acceptable but without triggering bigger reviews and more bureaucracy, like a lab call or a notice of funding opportunity. There was no new authority required, no new appropriations language to push for — just the willingness to pilot a new way of using something already in the toolkit.

The 15-minute pitch and the +1/0/-1 scoring rubric were chosen specifically to lower the social cost for younger researchers to pitch their own ideas. Junior staff and postdocs at the labs were able to walk into a WPTO room (virtually), on the strength of their own delivery, and pitch an idea without needing a senior name on the front of the application.

When we launched, in August 2019, the inaugural Request for Innovation funded 37 projects at $2.6M.

Some of my colleagues were reasonably sure the labs would run out of ideas by year one or year two — not as a discount of the researchers, but because nobody believed this was how the labs operated.

Seven years in, Seedlings is still one of the most popular programs in the water office.

What Came Out of It

By the end of FY2023, the program had funded roughly 150 Seedling projects for about $10M across the six labs, plus its follow-on Saplings tier for the projects that warrant continuation, plus partnerships with NSF and the Minority-Serving Institutions STEM R&D Consortium to expand the model to universities, plus a Hydropower Pilot Program that copied the model with seven projects at $350,000.

Some specific outcomes worth naming:

Thermomagnetic generators (NREL) — harvesting ocean thermal gradients via phase-change magnetic materials. Reached Sapling at $250K. The kind of architecture that doesn’t show up on any roadmap written in 2018.

Triboelectric nanogenerator wave energy converters (NREL) — a wildly different harvesting principle than anything in the existing WEC architectures pipeline.

Self-powered acoustic fish telemetry (PNNL and Penn State) — environmental monitoring with no external power requirement.

Machine-learning streamflow forecasting (ORNL) — early leveraging of AI to support advanced hydropower operations.

Wave-powered ocean sensing buoy (Sandia and Woods Hole) — atmospheric and oceanic data collection augmenting existing buoys through the power of waves.

None of these were on the WPTO roadmap when they showed up. All of them are now. Several have resulted in hardware that has been in the water, with patents filed and awards won in recognition of their novelty. And many more continue to be developed with public and/or private funding.

And as important as any of the individual technologies is that an increasing share of the program’s PIs are early-career researchers and postdocs who, under the old structure, would not have been able to own a project of their own. Seven years in, that pipeline effect is visible, and it is the part of the program I am most quietly proud of.

Saplings and the Advancement of Light-Touch Funding

Seedlings were more popular than we had assumed, but we didn’t want to stop there. There was such a proliferation of ideas that we didn’t want to leave new ideas at the proverbial front door, but wanted to change structures to continue to fund these new ideas.

The concept we created was to add Saplings as a follow-on tier.

Sapling A funds foundational research between $250,000 and $500,000 and integrates into an existing lab Annual Operating Plan or stands up a new one. Above $250,000, that triggers external merit review — which restores some of the orthodox accountability.

Sapling B is a commercialization track: a discovery phase up to $75,000, plus an optional advancement phase up to $150,000 that requires at least ten customer interviews and a real commercialization plan before the money moves.

Carrying Forward Flexibility

I am not arguing that all research should be structured this way.

Directed research, with milestones and external review and reporting, is what the labs should be doing for a good percentage of their work. In addition, some research cannot be done for relatively small amounts of funding, especially when larger tools or expensive materials are needed.

But a share of lab funding that is genuinely flexible, that lets researchers chase a hunch, build a thing they haven’t been told to build, propose under an open topic, own a project as a postdoc, has shrunk over two decades to something close to invisible. That share needs to be defended, named, and renewed. Not because every dollar inside it produces a breakthrough. Most don’t. But because the breakthroughs we will most regret missing are the ones that never got proposed because there was nowhere to propose them.

If we want strong, enduring research programs that go beyond a one-year horizon, the job in front of us is to disseminate the idea that it is okay to let researchers be researchers. To reduce the paperwork burden on people we have hired specifically to do the work that the paperwork is supposed to be about. To create mechanisms — especially at the labs, where we have unusual ownership over what does and does not get funded — that protect serendipity and encourage unencumbered risk taking. And to make space for the next generation of PIs to own their own projects before we ask them to run someone else’s.

Seedlings is one example of what that can look like inside a single office. It is not the only possible example. But it is the most honest case I know that the fix is not necessarily a new authority, or a new budget line, or a new appropriation. It is, in this instance, sitting with researchers in Hawaii, listening to them tell you what is in the way, and being willing to take a few percentage points of the office’s budget and use it differently.

Letting researchers be researchers is mostly an act of subtraction and letting innovation thrive in different ways.

One of the underappreciated facets of designing a solicitation is the review criteria. Often when you’re a sponsor of research, you might know what you want to try to solicit – from software to hardware to programs – and you can identify the metrics you’re trying to meet or exceed. But where things get complicated is how you try to develop a scoring methodology that allows for any applicants to understand what they’ll be scored against.

The evaluation criteria drive both what you’re looking for as a sponsor, as well as the way applicants navigate the solicitation. The first thing you’re going to do when looking at an award is: 1. See you if you’re eligible to apply; and then 2. Understand whether your solution or idea will score against the criteria.

But the downside of federal solicitations is that there is no standardization of how you set up the review criteria to score a solicitation. This is one of the things we tried to work through in DOE. This post is more funder-focused, but companies applying in may find the way we evaluate scoring methodologies, particularly within the government, interesting as well.

Subscribe now

DOE Example - Standard Criteria and Weighted Criteria in the Office of Energy Efficiency and Renewable Energy

In the 2010 timeframe, a working group was established to create a common set of review criteria in solicitations in both the Concept Paper Stage and the Full Application Stage. If you wanted to deviate from these standard criteria, you would have to get an exemption to develop your own criteria. The purpose for doing this was to create a good first step for program designers and solicitation managers to have a template to begin with.

The thought was also that the standard criteria are great to allow some predictability for applicants to understand the weight in which we review solicitations.

But they’re still vague and don’t perfectly fit all cases. They also leave room for some subjectivity in review.

One example for Concept Papers is the alignment to the full solicitation. Which makes sense, but at the same time is pretty arbitrary. Here is an example of what you’d have the following criteria for a concept paper:

Concept Paper Criterion: Overall FOA Responsiveness and Viability of the Project (Weight: 100%)

This criterion involves consideration of the following factors:

• The applicant clearly describes the proposed work, describes how the proposed work is unique and innovative, and how the proposed work will advance the current state-of-the-art;

• The applicant has identified risks and challenges, including possible mitigation strategies, and has shown the impact that EERE funding and the proposed project would have on the relevant field and application;

• The applicant has the qualifications, experience, capabilities and other resources necessary to complete the proposed project; and

• The proposed work, if successfully accomplished, would clearly meet the objectives as stated in the FOA.

And then here’s one example of what you have to review against in a full application:

Criterion 1: Technical Merit, Innovation, and Impact (40%)

This criterion involves consideration of the following factors:

Technical Merit and Innovation

• Extent to which the application responds to requirements established in the respective AOI;

• Extent to which the proposed work is innovative;

• Degree to which the current state of the art and the proposed advancement are clearly described;

• Extent to which the application specifically and convincingly demonstrates how the applicant will move the state-of-the-art to the proposed advancement; and

• Sufficiency of technical detail in the application to assess whether the proposed work is scientifically meritorious and revolutionary, including relevant data, calculations and discussion of prior work in the literature with analyses that support the viability of the proposed work.

Impact of Advancement

• How the project supports individual area of interest objectives and target specifications and metrics; and

• The potential impact of the project on advancing the state-of-the-art.

Criterion 2: Project Research and Market Transformation Plan (30%)

This criterion involves consideration of the following factors:

Research Approach, Workplan and SOPO

• Degree to which the approach and critical path have been clearly described and thoughtfully considered; and

• Degree to which the task descriptions are clear, detailed, timely, and reasonable, resulting in a high likelihood that the proposed workplan and SOPO will succeed in meeting the project goals.

Identification of Technical Risks

• Discussion and demonstrated understanding of the key technical risk areas involved in the proposed work and the quality of the mitigation strategies to address them.

Baseline, Metrics, and Deliverables

• The level of clarity in the definition of the baseline, metrics, and milestones; and

• Relative to a clearly defined experimental baseline, the strength of the quantifiable metrics, milestones, and a mid-point deliverables defined in the application, such that meaningful interim progress will be made.

Market Transformation Plan

• Identification of target market, competitors, and distribution channels for proposed work along with known or perceived barriers to market penetration, including mitigation plan; and

• Comprehensiveness of market transformation plan including but not limited to product development and/or service plan, commercialization timeline, financing, product marketing, legal/regulatory considerations including intellectual property, infrastructure requirements, and product distribution.

Criterion 3: Team and Resources (30%)

This criterion involves consideration of the following factors:

• The capability of the Principal Investigator(s) and the proposed team to address all aspects of the proposed work with a high probability of success. The qualifications, relevant expertise, and time commitment of the individuals on the team;

• The sufficiency of the facilities to support the work;

• The degree to which the proposed consortia/team demonstrates the ability to facilitate and expedite further development and commercial deployment of the proposed work;

• The level of participation by project participants as evidenced by letter(s) of commitment and how well they are integrated into the workplan; and

• The reasonableness of the budget and spend plan for the proposed project and objectives.

And a lot of thoughts and effort goes into these solicitations including metrics, benchmarks, etc. But at the end of the day, these are the necessary criteria an applicant must write against in order to get a successful outcome.

But one key thing is, how do you know whether each of the subcategories (bullets in the criteria description) of the review criteria is weighted equally or if one matters more than the other? Are all elements weighed equally to reach that one number to score that one category worth 40% of the score? Which should you concentrate on?

The solicitation authors and program designers may know which is the most important, but it isn’t abundantly obvious within the breakout. It’s a good methodology to assess projects – and honestly better than some that I was asked to review against in other agencies.

But it still can be difficult for applicants to understand how to meet the percentage criteria. Because that percentage might give you an understanding of the total weight of how much innovation matters versus team, but it’s still subjective to an extent. This isn’t a criticism of the review criteria – it’s been used time and again for weeding out the good from the bad. And one of the frustrations that we had as program designers is that the standard approach can still lack the ability to effectively communicate what you want from the applicant.

And it’s not just the applicants. The other important variable in this scoring is the reviewers themselves that DOE brings in to support merit reviews. Based on their expertise or experience they may anchor to a subset of the subcategories and base their scoring for the whole criteria on that. This is not a variable applicants can discern from reading anything and can be a challenge for program staff to understand unless they talk to the reviewer themselves.

Enter in a new methodology we developed and honed over some years, review criteria where there are specific statements graded on a specific scale to cumulate a score (and not caring as much about getting to match up to 100%). Make the scoring more quantitative and keep both the applicants and the reviewer more on top of what they are doing.

A Case Study in Doing Something Differently - Scoring and Prizes

Starting in 2017, there was a significant ramp up of using prizes to distribute funding. With the American Made Challenge platform launched, there was also an effort to rehaul how we might evaluate prize applications. Those of us in the early day trenches working on the multitude of prizes had a key objective – create a more rigorous but also empathetic approach to laying out what we wanted to see achieved as an objective.

Prizes are pay-for-performance models (versus grants which compensates you via invoicing for work you’ve done but outlays the total budget upfront), so it does beg the need to clearly and effectively communicate what exactly you’re looking for in terms of a delivered performance. So, prize solicitations need to be very specific. If not, and you’re not exact on what you want to deliver and evaluate the delivery of the good, then you’re not really being fair to applicants who are taking the risk of performance on their own balance sheet.

So, what did we develop – a more transparent scoring methodology that assessed different statements of criteria we wanted to have met. That would mean not relying on categories that you’re describing and a percentage of how that score would add up to 100%

Here is an example from the Waves to Water prize from a few years ago, which had several criteria to add up to a total score:

What you’ll notice here is that rather than having arbitrary percentages, you have a sum total of statements. Below you’ll see what we mean by statements:

Each one of those bullet points on the right side are equally rated, with a total potential score of six (more on that in a second). But what makes this different, is that on the left side is the expected information or work that a performer would need to present to meet the review statement on the right side. It seems obvious but is not common in solicitations – be explicit with what you want to see to have the confidence to review each statement. And so, a reviewer both knows what to look for and an applicant knows what to write towards. A lot of time and effort goes into thinking through both the left and right side of the ledger.

And rather than having an aggregate numerical score - which is more typical - per criteria (usually on a 1-10 scale) that’s more of a general judgement call, we had a specific scale for each one of those right hand bullets that would sum up to a total score:

The other thing that you avoid with this scale is that you’ll notice….there’s no “neutral” position to take. By forcing either to even slightly agree or slightly disagree, you’re forcing reviewers into what is often the default – taking a neutral position. So you’re forcing reviewers to send a signal. Because based on a whole lot of experience, reviewers tend to go with the neutral option, which makes it frustrating to assemble reviews. In addition, this scale was intentionally not 1-10. There is a tendency for reviewers to see 1-10 like academic grades so they may give something a 5 which would be an academic “F” but the definition of 6 for DOE might be “satisfactory” since it is right in the middle of the scale. By pushing reviewers to a scale they are less familiar with, it helps mitigate historic mental biases associated with certain numbers and, in theory, leads to more correct scoring.

One of the other things that we decided to incorporate into scoring is a basic question – should we fund this. It’s not a commonly asked question in reviews – a simply binary yes/no should this get funded. Yes gets a 6, No gets a 1. Why did we add it in? Because it helps to provide clarity, especially when a reviewer feels mixed about a project. Even if it has deficiencies, would you fund it? And sometimes we’d get baffling results – a high score with a “no” score of 1 to not fund. But it gives the people like us assembling the reviews a better signal, and it helped when stacking ideas against each other where they were on the bubble of selection.

Does this mean that this review scoring should all look like this? Probably not. Should there be more rigor into scoring to clarify what funders want? Undoubtedly yes. Is it harder on the sponsors? Yes, it is. Coming up with solicitation-specific individual statements takes a lot of time (and heated debates – very nerdy, very well-meaning debates) and also thinking through what is it that you want to see in order to have confidence that something will work or is meritorious is hard.

This can be a way to both increase confidence in each of the selected projects, but it also increases the empathy toward an applicant. Forcing companies and teams to read the tea leaves of a sponsor is aggravating at best, demoralizing at worst. By coming up with more detailed scoring criteria, funders can increase the quality of applications and increase the integrity of the review.

Take Aways

Solicitation scoring is always somehow fraught. A lot of thought and attention goes into the categories of what funders are looking for (for the most part). But there are areas of improvement, as highlighted by the above case study.

Be Clear In What You Prioritize as a Funder: Scoring should be the strongest indicator of what you value in selected projects. At a minimum, include at least 4 subcriteria.

Tell Applicants What You’re Looking For: Similarly to above, do not just give categories. What’s the evidence you’re looking for to meet these criteria? This can be accomplished in multiple ways, whether it’s in the solicitation writeup or in a helpful table. But the more clarity you can provide, the less tea reading that applicants have to do. Don’t be ambiguous unless you want varied responses.

Never Use a Neutral Score: Having parsed through thousands of concept papers, full application, and prize reviews, for the most parts reviewers go with the wishy-washy median score. As an example, if you’re going with a -1, 0, 1 scoring, just get ready for a whole bunch of 0s. Force reviewers to take a stance on how they feel about the solicitation. Even consider using the binary scoring example from above - yes fund, no don’t - as a way to parse through the myriad ways people score solicitations.

Consider Using Statements Rather than Categories: Coming up with prize statements took a lot of time and debate. But we had better solicitation responses to review. This may be most feasible with smaller dollar awards or prizes than large, complicated solicitations. But it again goes back to the principle - don’t be ambiguous, or get ready for ambiguous responses.

These are just some considerations when settling on scoring methodology.

Subscribe now

Today, the Clean Economy Project published the first report in their Innovation Initiative Blueprint Series, a proposed redesign of how the Department of Energy runs its Small Business Innovation Research (SBIR) program. We partnered with Clean Economy to write it, drawing on our collective experience building and running commercialization programs inside DOE.

We wanted to use this post to explain how the paper came together, why we focused on SBIR, what readers will find in it, and why we think the timing matters. For those who want to jump straight to the report, you can find it here.

How this paper came to be

The four of us at Waypoint Strategy Group spent a combined 50+ years inside DOE. We led or helped build programs across the department’s innovation portfolio, from early-stage R&D and SBIR execution to prize competitions, the Technology Commercialization Fund, technical assistance programs, the Powering the Blue Economy initiative, Office of Clean Energy Demonstration program design, partnership intermediary agreements and the Foundation for Energy Security and Innovation. SBIR was one of many diverse efforts allowing us to see the myriad of ways impactful programming can work in DOE within the same set of authorities, processes, and institutional constraints. We know what the system can do when it works well because we have seen it, and we know where friction can impede well intentioned programs because we have felt it.

We were all involved in SBIR through the technology offices we led and worked in, but we were not in a position to fundamentally impact how SBIR was run. The program sat in the Office of Science (SC), which managed it under a set of norms and processes that reflected its own research-first mission. The technology offices (the offices within what was formerly EERE, Nuclear Energy, Fossil Energy, and others) participated in SBIR by writing topics and reviewing applications, but the fundamental program architecture was set by SC. Over the years, each of us bumped into the consequences of how SC ran the program from different angles: founders who were confused by the process, program managers who wanted to do things differently but couldn’t, technologies that graduated from Phase II with no clear place to go inside DOE. In this construct, SBIR was underperforming its potential. We talked about how SBIR could be improved among ourselves but never had the time or the platform to write it all down in a way that could actually be useful to someone trying to fix it.

Clean Economy’s Innovation Initiative gave us a platform to share the SBIR program design we had imagined. The Innovation Initiative is building what they call the Blueprint Series, a library of implementable policy designs developed in partnership with former federal leadership, practitioners, founders, investors, and stakeholders who understand how the system operates in practice. Each blueprint targets a specific structural failure in the federal energy innovation system and is designed to be actionable enough that agencies and policymakers could begin implementing it immediately. They asked us to take on SBIR as the first blueprint, and that partnership gave us the space to do something we think has been missing from the SBIR conversation: not just a diagnosis of what is wrong, but a detailed operating design for how the program could actually work. We are grateful to Adria Wilson, Farah Benahmed, and the broader Clean Economy team for backing this effort and for the editorial rigor they brought to it.

Several current and former federal officials reviewed drafts along the way. We will not list them all here, but their willingness to engage with the details and push back where we were wrong or imprecise made this a considerably better product.

We want to be clear about what this paper is and what it is not. It is an attempt to start a conversation about how SBIR could work at DOE, not to deliver a final answer. The program has a lot of people who care about it, inside and outside the department, and have worked for years to improve it. The goal is to give them something real to react to, build on, and improve.

Why SBIR

Of all the programs in DOE’s portfolio, SBIR occupies a unique position for early-stage energy companies. It is often the first non-dilutive federal funding a small company can access. There is no equity stake, no cost share requirement, and no need for a prior federal relationship. A Phase I award is only $200,000, but when the program works well, it carries a validation signal worth more than the dollars: DOE looked at your technology and found it worth funding. That signal can open doors with investors, prospective customers, strategic partners like offtakers, and other parts of the federal system.

Because of this, SBIR functions as DOE’s de facto front door for startups. How a company enters through that door shapes how they are perceived and the network they have access to going forward. The front door shapes almost everything that follows, which is why getting SBIR right has outsized consequences for DOE’s ability to meet the program’s dual mandate of supporting scientific excellence and driving commercialization outcomes.

The paper documents what we believe are the core structural reasons the program underperforms its statutory commercialization mandate. A few of them are worth summarizing here because they illustrate why incremental fixes have not worked and why a redesign is warranted.

The first has to do with where the program sits inside DOE. SBIR has been managed by the Office of Science since 1982, when Secretary Edwards assigned it there because SC had the competitive grant-making infrastructure. That made sense at the time, but SC is fundamentally a basic research organization whose culture, review norms, definition of success, and hiring priorities all reflect that orientation. Running a commercialization-focused program out of a research-focused office produces exactly the kind of misalignment you would expect. That is not a criticism of SC staff, who are doing good work within the system they have. It means the program is being asked to produce commercialization outcomes inside an office that was built to produce research outcomes.

The second is the degree of fragmentation inside the program. What looks from the outside like one program is actually thirteen micro-programs, one under each technology office, operating under a shared submission portal but with different topic priorities, different review cultures, and different levels of engagement with applicants. A startup applying to “DOE SBIR” is really applying to whichever technology office wrote the topic that most closely matches their work, and if no topic matches, there is usually no way in. Open topics, which would allow companies to propose solutions within broad technology areas, have been rare because technology office staff, who carry SBIR as one of many duties, cannot handle the application volume that open topics would generate under the current staffing model.

The third is the weight of the application process relative to the size of the award. Phase I is a $200,000 feasibility study, but the application includes approximately 14 components: a mandatory letter of intent, active registrations in both SAM.gov and Grants.gov, a technical narrative, a commercialization plan, a detailed budget, key personnel qualifications, and compliance documentation. For a company that has never navigated federal grants, this process can take 150 hours or more, and some of them look at it and decide a chance at $200,000 is not worth the effort. DOE never has a chance to support those companies, who may have breakthrough solutions to further DOE’s mission, because they never enter the pipeline.

The fourth is what happens (or more accurately, what does not happen) after a company finishes Phase II. A company can complete SBIR Phase II with a validated prototype and still have no defined pathway into DOE’s broader demonstration, deployment, or procurement programs, because no office inside the department owns that handoff. When transitions do happen, it’s because an individual technology office program manager went out of their way to make a connection, not because the system was designed to produce that outcome. The paper also discusses SBIR Phase III, a statutory authority that allows agencies to award follow-on contracts to SBIR companies without running a new competition. DOE has almost never used it, but other agencies have. The Department of Defense reported $1.44 billion in Phase III contract transitions through AFWERX in FY2024 alone, which gives some sense of the scale of what DOE is leaving on the table.

What the paper proposes

The report lays out ten interconnected reforms organized across four areas: (1) how the program is designed and managed, (2) how applications are formatted and reviewed, (3) what award mechanisms are used and how to manage the awards, and (4) how awardees are connected to markets and measured after the funding ends.

While wholesale adoption is ambitious, it is achievable. But even if DOE elects to not implement the entire reform, these concepts can fundamentally transform the SBIR program, even if adopted incrementally or in part. And this is a critical starting point to work from for all stakeholders thinking about SBIR reform, whether inside the department, on the Hill, or in the broader ecosystem. But we also recognize that this is a detailed starting point; we have offered it as a hypothesis to be refined, and we genuinely welcome the conversation that hopefully follows.

Why now

SBIR’s authorization lapsed last October for the first time in the program’s 43-year history, and for roughly six months eleven federal agencies were unable to issue new SBIR solicitations or make new awards. Congress has since passed a five-and-a-half-year reauthorization (S. 3971, the Small Business Innovation and Economic Security Act) with broad bipartisan support. The reauthorization includes some structural changes, including new proposal limits, enhanced foreign risk due diligence, and a new Strategic Breakthrough Award mechanism, but it does not address the deeper program design issues at DOE which this paper focuses on. The legislative debate over the past year opened a window for this kind of conversation, and we hope that window is still open even as the program restarts.

The tools that would make reform possible also exist now in ways they did not during previous conversations about SBIR modernization. General purpose AI tools are reaching a level of practical capability where they could support the development of trusted screening and compliance processes, which would be necessary to manage the application volumes that open topics would generate. The partnership intermediary infrastructure has been established and is in active use across DOE for other programs. The new DOE foundation, which has an explicit mission to support commercialization - the Foundation for Energy Security and Innovation - has launched and is staffing up. Technical and Business Assistance vouchers are an available but chronically underutilized authority. Recent Section 309 appropriations language enables fund flexibility for open topics. None of these existed in their current form the last time DOE had a serious conversation about SBIR reform.

And the proof points exist: AFWERX, ARPA-E, NASA, NSF, and DOE’s own (former) Solar Energy Technologies Office and (former) Water Power Technologies Office have all demonstrated, under the same legal framework that governs DOE SBIR, that federal innovation programs can move faster, reach more companies, and produce better commercial outcomes than what DOE’s SBIR program currently delivers. The reforms we are proposing have already been tested in other federal contexts, and the question at this point is less about whether they work and more about whether DOE will apply them to a $300 million program that has been running on essentially the same design for over four decades.

Read the paper

We have written before on Innovation Waypoints about the process of designing federal programs from scratch, the sequence of decisions that determines whether a well-intentioned program actually produces the outcomes Congress intended. (Our Common Language post walked through those steps in detail.) This paper is that thinking applied to a specific program with real consequences for the companies and technologies DOE is supposed to be supporting.

If you have been through the SBIR system, whether as a founder, a program officer, a reviewer, or an investor who watches federal funding signals, you have a perspective that matters here. We hope you will read the report and tell us where we got it right and what we missed. We want this to be the beginning of a longer conversation about how DOE supports the next generation of energy companies, not the last word on it.

The full report is available here.

Clean Economy’s introduction to the Blueprint Series.

Innovation Waypoints is brought to you by Waypoint Strategy Group.

We’re excited to release our report discussing the opportunity that Other Transactions present, where the flexibility they provide is most valuable, and recommendations on how to realize their full potential.

We wish to thank our contributors and reviewers for their thought partnership, feedback, and support.

Below is the executive summary of the report. You can find the full report here.

The U.S. Department of Energy (DOE) is the federal government’s primary agency for funding energy research, development, demonstration, and deployment. With annual discretionary budgets exceeding $46 billion and supplemental appropriations from legislation such as the Infrastructure Investment and Jobs Act and the Inflation Reduction Act, adding tens of billions more, DOE wields substantial financial resources to advance the nation’s energy innovation agenda. To deploy these funds, the department relies on several mechanisms, including traditional procurement contracts, grants, and cooperative agreements, each governed by extensive federal acquisition regulations and statutory requirements.

Among DOE’s available instruments is Other Transaction (OT) authority, a contracting mechanism defined primarily by what it is not: OTs are legal agreements that are neither contracts, grants, nor cooperative agreements, and are therefore not subject to the Federal Acquisition Regulation or many other federal procurement laws. This negative definition gives OTs substantial flexibility to tailor terms, negotiate intellectual property arrangements, and attract nontraditional contractors who might otherwise avoid government work. Other federal agencies, notably the National Aeronautics and Space Administration (NASA) and the U.S. Department of Defense (DOD), have used OT authority far more extensively than DOE, which has historically underutilized this tool despite possessing the statutory authority.^[1] (DOD is “using a secondary Department of War designation,” under Executive Order 14347, dated September 5, 2025).

OT authority offers flexibility that can meaningfully improve execution speed and reduce administrative burden while enabling a balanced risk-management approach. This report argues that realizing OT’s potential requires matching the tool to situations where it can improve outcomes for energy and industrial innovation and commercialization while building the institutional capacity and infrastructure, including frameworks, policies, procedures, and staff capacity, to use it effectively. Leveraged well, OT becomes a powerful component of DOE’s toolkit, leading to improved industry experience and outcomes. Leveraged without preparation, even valuable tools underperform.

This report offers a practitioner’s perspective on OT at DOE, grounded in direct experience with completed award cycles. Our central argument is that OT is valuable in specific contexts but is not a universal solution, and that implementation capacity matters far more than statutory authority. DOE already has OT authority. It lacks the institutional capacity and infrastructure to use that authority effectively as a unique tool to achieve real impact.

We organized the report around two questions that should guide DOE’s approach:

1. When does OT actually help? Not all projects benefit from OT flexibility. Understanding which project types, performer profiles, and program objectives are well-served by OT enables strategic deployment rather than default adoption.

2. What does DOE need to use OT effectively? OT authority alone is insufficient. Effective use requires institutional infrastructure: clear policy, established frameworks and appropriate term sheets, simplified timelines for approvals, clarified roles and responsibilities across federal project managers, legal and contracting, trained staff, and organizational capacity.

Key Findings

Different projects need different approaches. We identify five project archetypes with distinct OT needs, driven by differences in project risk profiles and team compositions: early-stage R&D, mid-stage demonstration, first-of-a-kind deployment, consortia and novel structures, and demand-side mechanisms. Each requires a different OT variant with appropriate risk tolerance and structured flexibility. Universities, by contrast, are generally better served by standard financial assistance.

Project Archetypes and Mechanism Fit

Cross-cutting policy issues persist regardless of the mechanism. Seven issues must be addressed in every award, whether OT or another mechanism. Without clear, Secretary-level (or Secretary-delegate) policy positions, every negotiation re-litigates fundamental issues, extends negotiation timelines, shifts policy risk to career staff, and potentially creates inconsistent terms across awards. This ambiguity can damage projects due to delays and financing availability, erode trust when different teams receive apparently different treatment, and increase risk on both sides.

The flexibility requires structure. OT can modify cost accounting requirements, negotiate different IP arrangements, structure milestone payments, and tailor administrative requirements. These flexibilities help certain projects and partners move faster, reduce administrative burden, and manage risk. But flexibility without structure can lead to bespoke negotiations that exhaust staff, disadvantage small companies, delay work starting, and produce inconsistent outcomes. Practitioners can mitigate this tension through deliberate design. Establishing agency-level policy positions, a focus on real program needs, objectives, and risk tolerance, program-level prototypical frameworks (~80% complete), and project-level bounded negotiation together preserve meaningful flexibility while enabling speed, predictability, balanced risk management, and institutional learning.

Considerations for Intentional Program Design

Intentional design is essential for any program, regardless of mechanism utilized. The program objectives and the expectations and needs of potential participants should be characterized before talking about best-fit approach.

1. What are you trying to accomplish, and why? “Support clean energy innovation” is not specific enough. “Fund demonstration of novel refining approach for critical minerals at pilot scale through partnership with a Series C company needing government co-investment” is specific. Articulating the underlying reason increases the likelihood that final agreements will align with true objectives.

2. What terms are needed? What performer type? What government involvement level? What payment structure? What IP arrangement? What oversight? These functional requirements should drive selection.

3. Can existing mechanisms work? Can requirements be effectively met with grants or cooperative agreements? Would a prize provide the right incentives and constraints to accomplish desired goals? If the barrier is institutional bureaucracy rather than statute, fixing cooperative agreements might be the solution.

4. If OT, which archetype? If OT flexibility is genuinely needed, which of the five variants fits? If none fits, either the program is genuinely novel or the analysis hasn’t identified actual needs.

Institutional capacity is the binding constraint. DOE has limited OT-warranted contracting/agreements officers, limited General Counsel (GC) staff with OT expertise, no standardized templates, inconsistent approaches across offices, and established expertise in other mechanisms that may bias or anchor staff to existing ways of operating. These constraints, not statutory limitations, determine what is achievable. Expanding OT use without building capacity and guardrails can lead to additional failures that could discredit OT for years, removing a useful capability from DOE’s toolkit and blunting its ability to support energy innovation.

Core Recommendations

There are immediate and near-term actions that would increase the utility and value of OT.

Note: As with the policy decisions, many of these recommendations improve overall governance and institutional capacity and, as such, apply to mechanisms beyond OT.

The Bottom Line

OT can provide real advantages for the right projects—speed, flexibility on terms, compatibility with commercial financing expectations and structures. The goal is to enable better outcomes for energy and industrial innovation and investment by using the right tool effectively for each situation. This requires deliberately building institutional capability: policy clarity, standardized frameworks, trained staff, disciplined deployment, and a balanced risk approach. Done well, OT becomes a valuable component of DOE’s toolkit, accelerating and streamlining projects and programs while improving outcomes. Done poorly, it becomes another disappointing cycle of enthusiasm and disillusionment, potentially limiting legitimate and impactful uses of OT in the future and blunting DOE’s ability to effectively support advanced energy innovation and commercialization.

Proactive decision making and deliberate action now are crucial to realizing the full value of OT.

If you’d like to learn more, the full report is here.

Innovation Waypoints is brought to you by Waypoint Strategy Group.

If you follow energy policy or federal funding, you have probably encountered growing enthusiasm for “Other Transactions” (OTs) at the Department of Energy. Executive orders, Congressional interest, and a steady stream of commentary from practitioners and policy thinkers have put OT squarely in the spotlight.

This post is a primer: what OT actually is, where it came from, why it matters at DOE, and what it can (and cannot) change about how federal energy programs operate.

DOE’s Funding Toolkit

To understand OT, you need to understand the landscape it sits within. DOE deploys direct funding through several distinct mechanisms, each with different characteristics.

Grants provide funding with minimal federal involvement. The recipient proposes work, receives funds, and executes with limited oversight. Grants are appropriate when the government’s interest is in enabling activity without directing it.

Cooperative agreements provide funding with “substantial involvement” by the federal government. Unlike grants, cooperative agreements contemplate active federal participation: technical guidance, milestone reviews, collaborative problem-solving. The substantial involvement is a feature. It reflects legitimate government interest in partnership, and for projects where DOE expertise can improve outcomes, it is the right tool.

Procurement contracts are used when the government is acquiring something for its own use. The Federal Acquisition Regulation (FAR) governs procurement.

Prizes pay for outcomes rather than activities. The government defines a challenge, competitors attempt solutions, and winners receive awards.

Partnership Intermediary Agreements (PIAs) create an intermediary between DOE and program participants, enabling flexible structures for technology transfer, advising, and convening.

Each of these mechanisms has rules, reporting requirements, and administrative structures. Each requires different oversight staff and capabilities. And each was designed for a specific set of circumstances.

How OT Fits into the Toolkit

OT is defined in statute by what it is not: an agreement that is neither a procurement contract, nor a grant, nor a cooperative agreement. This negative definition creates both flexibility and confusion.

Because OTs fall outside the Federal Acquisition Regulation and many provisions of 2 CFR 200 (the framework governing federal financial assistance), they can be structured differently than standard awards. Specifically, OT can modify:

Cost accounting requirements. Standard financial assistance requires compliance with federal cost principles. OT can accept commercial accounting standards (like GAAP) instead, removing the need for companies to retrofit their financial systems for government work.

Intellectual property rights. Standard provisions (Bayh-Dole for financial assistance, FAR clauses for procurement) establish default IP treatment. OT can negotiate different arrangements within statutory limits.

Payment structures. Cooperative agreements typically use cost reimbursement. OT can use milestone-based payments, fixed amounts, or other structures suited to project needs.

Competition requirements. Standard mechanisms have specific competition requirements. OT provides more flexibility in how competition is structured.

Administrative requirements. Reporting, documentation, and oversight can be tailored rather than following standard templates.

None of these flexibilities are automatic. Each requires deliberate decisions about what the project needs and what terms serve program objectives. An OT agreement can range from something barely distinguishable from a cooperative agreement to a highly customized arrangement that looks nothing like a standard federal award.

What OT Does Not Resolve

One of the most important things to understand about OT is what it does not change. Significant policy questions around intellectual property, domestic content requirements, environmental review, cost sharing, and other areas recur across DOE programs and must be addressed in every major award regardless of funding mechanism. OT changes the framework within which these questions are negotiated. It does not eliminate the need to negotiate them. No mechanism does. When these overarching policy questions lack clear positions from leadership, specifically political leadership, every negotiation relitigates them from scratch, extending timelines and creating inconsistency across awards. That dynamic persists whether the award is an OT, a cooperative agreement, or anything else.

Where OT Authority Comes From

Unlike most agencies, DOE has two distinct OT authorities.

Section 7256(a), from the DOE Organization Act of 1977, authorized the Secretary to enter into agreements “of any kind” necessary to carry out the Department’s mission. This organic authority mirrors NASA’s Space Act authority, which has supported decades of flexible partnerships. As Congressman John Dingell stated during the original debates, this provision was intended to give DOE flexibility such that the Secretary “can do literally almost anything he wants to in terms of expending money and making agreements.”

Section 7256(g), enacted as part of the Energy Policy Act of 2005, provides explicit OT authority with specific intellectual property flexibility provisions. This authority sunsets in 2030 unless reauthorized, which creates uncertainty for long-term planning.

The 7256(a) authority does not sunset, providing a durable foundation for OT use even if 7256(g) expires. However, the relationship between the two authorities and their respective IP provisions creates complexity that requires careful legal navigation.

A Brief History

DOE’s OT history is shorter and more complicated than you might expect. Despite having broad authority since 1977, DOE retreated into R&D, nuclear weapons work, and environmental cleanup after attempts to terminate the agency in the 1980s. Serious deployment efforts were largely dormant until the Energy Policy Act of 2005, which granted explicit OT authority and imposed a 90-day deadline for implementing regulations, reflecting Congressional frustration with DOE’s contracting limitations.

The first significant OT experiment was the Technology Investment Agreement (TIA) framework, implemented in the early 2000s. TIA was deployed with heavy constraints and provided limited practical flexibility beyond IP negotiability. When some TIA-funded bioenergy projects failed (failures driven primarily by market conditions and technology readiness rather than the funding mechanism), the negative outcomes became associated with OT itself. This created a lasting institutional consequence: a cohort of DOE staff learned to view OT as complicated, risky, and low-value. Legal interpretations became more restrictive. Skepticism persists in some quarters today.

DOE is therefore working from a complicated starting point. Expanded OT use must overcome the usual challenges of building new capability, along with the specific challenge of undoing constrained interpretations that took root after TIA.

DOE’s OT situation also differs from other agencies in ways that matter. Importing approaches from the Department of Defense (also referred to as the Department of War) or NASA without adaptation produces poor results.

The commercial dynamics are different. In the defense space, the Department of War is often the procurer of the solutions it funds. DOE is not. DOE-funded companies need to succeed in private markets, which means fundamentally different business strategies, financing structures (project finance, tax equity), regulatory environments (state public utility commissions, FERC), and commercial relationships (utilities, offtakers, project developers).

The institutional capacity is different. DOE has historically had minimal OT infrastructure and little institutional experience. While this has improved, DOE remains far from the distributed capability that effective execution requires. NASA and the DOD, in comparison, have spent decades using OTs and developing the associated institutional capacity and processes.

What Comes Next

This primer is a starting point. The challenges and opportunities for OT vary considerably across project types, performer profiles, and mission objectives. Early-stage R&D companies face different constraints than first-of-a-kind deployment projects integrating complex private financing. Consortia and multi-party structures have different needs than standalone awards. Understanding where OT adds genuine value, and where existing mechanisms remain the better fit, requires careful exploration of these differences and intentional planning and program design. Without that deliberate work, even valuable tools underperform.

We will be publishing more on these topics in the coming months. OT is a valuable addition to DOE’s toolkit, and we hope this overview helps ground a shared conversation about how to use it effectively.

Innovation Waypoints is brought to you by Waypoint Strategy Group.

References and Further Reading:

• DOE OT Guide (2023)

• DOE OT Regulations (2 CFR 930)

• Energy Policy Act of 2005, Section 7256(g)

• GAO-08-798R: DOE Implementation and Use of OT Authority

• Keith Boyea’s Substack on OT and Federal Acquisition

• Statecraft: How to Fix a Department’s Funding Tools (Narayan Subramanian interview)

• Rick Dunn, Other Transactions Guide, Third Edition

If you’ve spent any time in energy technology commercialization, you’ve encountered the alphabet soup: TRL, ARL, CRI, MRL, SRL, IRL. Each of these metrics was built to answer a version of the same question: how ready is this thing? And each, in its own way, falls short of giving you a complete answer.

And in most cases, that’s ok.

Metrics and yardsticks are highly important in this space. They enable quantitative assessments. They allow comparisons across projects. They give program managers a shared vocabulary for articulating goals and tracking progress. But they are seldom sufficient on their own. In our experience designing and running DOE programs, the most valuable function of a readiness metric is the conversation that objective setting and the scoring process forces you to have. The number itself is secondary.

Why Metrics Matter for Program Design

From the perspective of a program designer at DOE (or any mission-aligned funder), readiness metrics serve several concrete purposes, and their value begins well before a program launches. During program scoping, structured conversations around readiness across multiple dimensions can reveal the true maturity and challenges of the industry a program intends to serve. That understanding shapes better objectives, more realistic eligibility criteria, and program structures that reflect the actual state of the field rather than assumptions about it.

Once a program is in motion, readiness metrics help target specific challenges: if a program is aimed at bridging the gap between bench-scale validation and pilot demonstration, saying “we’re looking for TRL 4-6 technologies” communicates that instantly. They help manage applicant burden: a well-articulated readiness threshold tells prospective applicants whether a program is meant for them, reducing wasted effort on both sides. And they provide a basis for tracking progress, which feeds into program evaluation and helps the agency learn from outcomes to improve future funding actions.

These are real, practical benefits. The challenge arises when the metrics are asked to do more than they can.

Without a consistent, third-party assessment and validation capability, groups often overstate or misstate their readiness levels. This is especially true with TRL, where the incentive to claim a higher number is strong and the verification infrastructure is thin. A team might genuinely believe their technology is at TRL 6 when a more rigorous assessment would place it at TRL 4. That gap is rarely dishonest. TRL definitions leave enough ambiguity that reasonable people can disagree, particularly as system complexity increases. But it means that readiness scores, taken at face value across a portfolio, can be unreliable as absolute measures. They work much better as comparatives, useful when you’re evaluating closely related technologies against each other, less useful as standalone verdicts.

The bottom line: Readiness metrics are important, particularly to articulate goals and objectives for projects and programs. But use them mostly to drive deep conversations and relationship building, rather than getting hung up on exact definitions and the precise placement of projects on each scale.

Technology Readiness Levels (TRL)

TRL is the original and most widely used readiness framework. Developed by NASA in the 1970s to assess technological maturity for complex systems (think Apollo-era hardware), it tracks a technology’s progression from basic scientific discovery (TRL 1) through laboratory validation, prototype demonstration, and ultimately to a system proven in an operational environment (TRL 9). DOE - like many other agencies and research organizations - adopted and adapted the framework for its own technology portfolio. Today TRL is embedded in nearly every funding opportunity announcement across the agency.

TRL’s strength is its simplicity and universality. For standalone technologies with clear performance benchmarks (a new photovoltaic cell chemistry, a novel membrane for carbon capture) TRL provides a useful shorthand for where the technology sits on the development continuum. It’s particularly well-suited for applied R&D, where the primary question is whether the technology works and at what scale it has been validated.

But TRL has real limitations. As system complexity increases, it becomes harder to assign a single TRL score that means anything. A long-duration energy storage system might have a core storage medium at TRL 7 while its power conversion system is at TRL 4, and its integration into a specific grid application is at TRL 2. What TRL is the project? The answer depends on what you’re scoping, and reasonable people will scope it differently. As we explored in our post on the necessity of a systems approach to risk, system-level complexity demands a different analytical lens than any single readiness metric can provide.

More fundamentally, TRL measures technical readiness, hence its name. It tells you nothing about whether the technology can be manufactured at scale, whether there’s a market for it, whether the regulatory pathway is clear, whether the supply chain exists, or whether communities will accept it. A technology can be at TRL 9 and still be decades away from commercial deployment if these other barriers remain unaddressed. For anyone who has worked in energy commercialization, this is familiar territory. TRL is often most useful and complete early in the commercialization spectrum, and becomes progressively less complete as a metric as the solution or technology matures.

TRL is also, among all the readiness metrics, the most susceptible to wishful thinking. Because it’s the most commonly used and often the most consequential for funding decisions, there’s a natural pressure to adjust to align with funding requirements. Without independent validation, self-reported TRL scores should be taken as conversation starters, not conclusions.

Adoption Readiness Levels (ARL)

The ARL framework was developed by DOE’s Office of Technology Commercialization (formerly the Office of Technology Transitions) and the Office of Clean Energy Demonstrations in partnership with industry stakeholders to address exactly what TRL leaves out. Where TRL asks “does the technology work?”, ARL asks “what else needs to be true for this technology to actually get adopted?”

ARL assesses 17 different dimensions of adoption risk across four core risk areas, then translates that assessment into a 1-to-9 readiness score. The four risk areas cover:

1. Value Proposition—Assesses the ability for a new technology to meet the functionality required by the market at a price point that customers are willing to pay, to meet the market demand (a broadened definition of “product-market fit”).

2. Market Acceptance—Captures the target market(s) demand characteristics and risks posed by existing players — including competitors, customers, and other value chain players.

3. Resource Maturity—Determines risks standing in the way of inputs that are needed to produce the technology solution.

4. License to Operate—Identifies the societal (national, state, and local), non-economic risks that can hinder the deployment of a technology.

The breadth is the point. ARL’s value is in forcing a comprehensive conversation about all the reasons a technology or solution may fail beyond core technical performance. A team might be laser-focused on hitting a performance target in the lab, but the ARL process pushes them to confront questions like: Is there actually demand for this? Can you get the raw materials or components? Who’s going to build the factory? Will communities where it will be deployed accept it? Is the regulatory pathway understood? These are the questions that determine whether a technology makes it to market or dies in the valley of death despite having a high TRL score.

At the program level, ARL is particularly useful for scoping programs and articulating their challenges and objectives. When OCED designed its demonstration programs, walking through the ARL dimensions with industry stakeholders before programs launched helped the team understand the real maturity and barriers facing potential applicants. That process shaped which adoption risks each program should target, which were outside its scope, and how program structures could be tailored to the actual state of the industries involved. At the portfolio level, it enables comparative analysis across subsectors and investment areas. Because OCED’s portfolio spans everything from carbon capture to industrial decarbonization to small modular reactors, a consistent framework for comparing non-technical risks across these very different technology areas is essential for strategic decision-making.

The trade-off is complexity. With 17 dimensions, an ARL assessment is inherently more involved than a single TRL number. It’s harder to communicate succinctly, which can limit its utility as a quick screening tool. But for anyone making or informing a significant investment decision, the additional nuance is worth the effort. The Commercial Adoption Readiness Assessment Tool (CARAT) that operationalizes the ARL framework is designed to be deliberately simple. Its power is in quickly identifying where critical barriers exist, and the scoring precision matters less than the conversation it generates.

Commercial Readiness Index (CRI)

The CRI was developed by the Australian Renewable Energy Agency (ARENA) in 2014 and takes a different but complementary approach. Where TRL tracks technical development and ARL spans the full range of adoption barriers, CRI focuses specifically on commercial and financial maturity: essentially, the financeability of the technology.

CRI uses a six-level scale, from a hypothetical commercial proposition (CRI 1) to a bankable asset class comparable to mature energy technologies (CRI 6). It evaluates eight indicators: regulatory environment, stakeholder acceptance, technical performance data availability, financial cost and revenue propositions, industry supply chain and skills maturity, market opportunities, and company capabilities.

CRI’s design reflects its origin: it was built to help a national renewable energy agency decide where to deploy limited public funding to greatest effect. It serves as a close partner to TRL. In ARENA’s framework, CRI explicitly begins where TRL starts to plateau (around TRL 7) and extends through the commercialization journey that TRL doesn’t cover. If TRL asks “does it work?” and ARL asks “what could prevent adoption?”, CRI asks “can this be financed?”

That’s a valuable question. In energy infrastructure, the ability to attract private capital at reasonable terms is often the binding constraint. A technology that works, has a market, and has community support can still fail if lenders and investors can’t underwrite the risk. CRI provides a structured way to assess and communicate that financial readiness.

However, like TRL, CRI is still fundamentally focused on the core technology and its immediate commercial proposition. It’s less useful for discussing system-level effects: the interactions between a new technology and the broader grid, market structures, policy environment, and infrastructure that determine whether deployment actually scales. And as a framework developed for the Australian renewable energy context, some of its indicator definitions need adaptation when applied to other sectors or geographies.

The Others

There’s no shortage of additional readiness frameworks. Manufacturing Readiness Levels (MRL), originally developed by the Department of Defense, assess manufacturing maturity from concept through full-rate production. Supply Chain Readiness Levels focus on the maturity and resilience of input supply chains. System Readiness Levels and Integration Readiness Levels attempt to address the system-of-systems complexity that TRL struggles with, evaluating how well individual components work together rather than assessing them in isolation.

Each of these was developed to solve a real problem, and each has a legitimate use case. But proliferating metrics can create its own problems. If every dimension of readiness has its own scale, you can end up with a dashboard so complex that it obscures rather than illuminates. The goal should be clarity, not comprehensiveness for its own sake.

How to Use Them Well

Based on our experience, a few principles emerge for how to use readiness metrics effectively. (We explore related themes in our posts on the necessity of a systems approach to risk and the wicked problem of program evaluation.)

Use multiple metrics in tandem. No single framework captures the full picture. TRL paired with ARL gives you both the technical and adoption perspectives. Adding CRI brings in the financial lens. The combination is more powerful than any one alone.

Seek consistent, third-party validation where possible. Self-reported readiness (i.e. from a potential recipient or a researcher) scores are useful but come with their own set of biases based on internal perception and, on occasion, the specific funding being sought. Where you’re comparing closely related technologies or making investment decisions, having a consistent assessor apply the framework improves reliability significantly. This is particularly important for TRL, where overstatement is common.

Match the metric to the decision. If you’re designing an applied R&D program, TRL is probably your primary lens. If you’re designing a demonstration program that needs to address commercialization barriers, ARL provides the right framing. If you’re evaluating the bankability of a near-commercial technology, CRI may be your best option. Using the wrong metric for the decision at hand creates confusion and can lead to poorly targeted programs. It takes work to set the targets within a solicitation, and it takes work from every applicant to assess their solution against those metrics. Poorly set targets and poorly chosen metrics causes undue burden, distorts the ground truth of the companies and technologies, and can complicate decision-making and funding landscapes inside and outside of the federal ecosystem.

Treat values as comparative, not absolute. A TRL score or ARL score in isolation tells you less than you think, but consistent evaluation across a set of related technologies and projects, by the same assessor, using the same methodology: that can be useful. Readiness metrics are strongest when used to compare, weakest when used to make binary go/no-go decisions.This is particularly a challenge for managers or organizational leadership. There is an inherent desire to boil large complex programs or portfolios of projects down to simple metrics. While the desire to distill the complex to the simple is only human, there is the risk that rigidly adhering to specific numbers can lead to misunderstandings or a false sense of certainty. Pushing for specificity on these values when it is not imperative to moving work forward, can lead to busy work for staff and disenchantment with using the metrics at all leading to a loss of all value.

And perhaps most critically:

Use them to facilitate and sustain conversations. This applies before programs launch, not just during selection or oversight. Sitting down with industry stakeholders and walking through readiness dimensions during program design surfaces assumptions about industry maturity, identifies barriers that the program should (or should not) try to address, and builds shared understanding that makes the eventual program more grounded in reality. During selection and project management, the same process surfaces blind spots and builds shared understanding across disciplines that might not otherwise be in the same room.

Ensuring energy sector competitiveness, affordability and sustainability depends on moving technologies from lab to market faster and more reliably than we ever have before. Readiness metrics are one of the tools we have for navigating that journey. They’re most useful when we’re honest about what they can and can’t tell us, and when we treat them as the start of a deeper conversation rather than a final answer.

Innovation Waypoints is brought to you by Waypoint Strategy Group.

Every Administration leaves its mark on the U.S. Department of Energy (DOE), including by reorganizing the agency. Over the last few months, Secretary of Energy Chris Wright and his team at DOE have been reshaping the agency through organizational shifts.

In November, the Secretary announced a realignment of the agency, as seen below.

Figure 1: DOE Organizational Chart Effective November 2025

When the realignment was announced, several offices disappeared from the organizational chart, but a new one appeared - reporting directly to the Secretary - the Office of Critical Minerals and Energy Innovation (CMEI), highlighted above. While it was generally understood that several offices would be realigned under this new office, details on the structure of CMEI were still pending. This week, DOE announced the structure of CMEI:

Figure 2: Missions of the Offices of CMEI

But what does that mean in terms of where the previous work - and appropriated funding - go within this organization? Taking a step back, to explain the new office requires a walkthrough of the other offices now realigned under CMEI.

Overview of CMEI and Organization Realigment

CMEI is a combination of several offices, including: the Office of Energy Efficiency and Renewable Energy (EERE), the Office of State and Community Energy Programs (SCEP), the Office of Manufacturing and Energy Supply Chains Office (MESC), the Office of Clean Energy Demonstrations (OCED), the Federal Energy Management Program (FEMP), and parts of the Grid Deployment Office (GDO) and the Office Fossil Energy (FE).

Below is an org chart reflective of the Biden Administration. The highlights include the various offices that now are realigned under CMEI. Note that pursuant to the org chart below, offices with dashed lines represent offices that are only partially realigned. As an example, portions of the Office of Fossil Energy are also housed now under the new Hydrocarbons and Geothermal Energy Office (HGEO).

Figure 3: DOE Organizational Chart Effective January 2025 with CMEI Highlights

Several of these offices - GDO, MESC, SCEP - were created in February 2022 by the last Administration, principally in response to the substantial investment into the Department through the Infrastructure Investment and Jobs Act (IIJA). FEMP and parts of SCEP - including the Weatherization and Intergovernmental Programs Office (WIP) - were actually under EERE under the first Trump Administration, but were spun out into their own offices to tackle larger missions.

The combination of all of these offices represents a substantial portfolio of tens of billions of appropriated dollars - through IIJA, the Inflation Reduction Act (IRA), and annual appropriations.

To tackle this portfolio, the Administration has broken the work out into the three offices as described above, and reflected in the organization chart below:

Figure 4: CMEI Organization Structure Effective January 2026

The leadership structure, with an Assistant Secretary, Principal Deputy Assistant Secretary (PDAS), and Associate Principal Deputy Assistant Secretary (APDAS) is largely retained from EERE under the last Administration, with political appointees in the Assistant Secretary and PDAS positions, and APDAS as a career official. This org chart also represents a simplification/consolidation of the Operations pillar.

But where a lot of the change and amendments are made are to the three technical pillars - or as referenced in the announcement, Offices: (1) Critical Minerals, Materials and Manufacturing; (2) Energy Technology; and (3) Innovation, Affordability, and Consumer Choice.

Below is how EERE was structured before the re-org:

Figure 5: EERE Organization Structure Prior to November 2025

The three pillars were: Buildings and Industry, Renewable Energy, and Sustainable Transportation & Fuels. All of these offices have now been realigned to each of the new Offices under CMEI.

First, the Energy Technology Office:

Figure 6: Office Realignment for Energy Technology Office

On the left are the new offices under Energy Technology, mapped to their former office titles. This Energy Technology Office now represents most of the former Sustainable Transportation and Renewable Energy pillars. This office spans across clean energy and transportation research and development. Notably, the Geothermal Technologies Office that was in EERE has shifted to the new Hydrocarbons and Geothermal Energy Office.

Bioenergy and Hydrogen have been aligned with Alternative Fuels & Feedstocks, and Integrated Energy Systems now houses the solar, wind, and grid efforts. Transportation Technologies includes the non-battery portion of the former Vehicle Technologies Office. And reflective of the Administration’s prioritization of hydropower, the Water Power Technologies Office has been renamed to Hydro Power and the Hydropower Incentives that were in GDO have shifted to this office as well.

The Innovation, Affordability, and Consumer Choice Office is more straightforward, with principally offices that were separate from EERE now back under the same larger office structure, as seen below:

Figure 7: Office Realignment for Innovation, Affordability, and Consumer Choice

This office will be responsible for administering substantial programs, including appliance standards, energy efficiency rebates, weatherization funding, assisting federal agencies with energy planning, and research and development for industrial and building technologies.

And last - and the most substantial in terms of size and mandate - is the Critical Minerals, Materials and Manufacturing Office.

Figure 8: Office Realignment for Critical Minerals, Materials and Manufacturing

While this looks relatively straightforward, some of these activities being realigned under this Office were very large and complex, with billions of dollars of funds remaining under the IIJA and Inflation Reduction Act. It appears that all of MESC is now a part of this organization. The activities under MESC included all of the battery manufacturing and recycling work funded by IIJA, as well as the execution of the 48C tax credit program.

This new organization is also the new home for OCED. But unlike its previous iteration, not all of the programs that were once housed under OCED will be under management of CMEI. Large programs like the Advanced Reactor Demonstration Program and much of the carbon capture programs have been shifted to other parts of the agency - in the Office of Nuclear Energy and the Office of Hydrocarbons and Geothermal Energy. The full list of projects and programs now under management in CMEI is still forthcoming.

Takeaways

So what does this all mean, in general? It means that while the Administration is realigning much of the Department, a lot of the core work and equities that have been built over decades are still within the purview of the agency. And the reality is that much of what these offices will focus on still remains to be seen.

The upcoming Fiscal Year 2027 budget request from the President, which is expected to be released sometime after the State of the Union, will be one indicator. That may give some insight into the priorities, structure, and key upcoming activities throughout CMEI.

Another will be any forthcoming strategic plans - both at the larger CMEI level, as well as through all of the new offices. This could include meshing together of existing plans, or entirely new strategic plans and directions to reflect the priorities of the Administration. Strategic plans, multi-year program plans, and other planning efforts will be no doubt taking shape during the course of the Administration.

But in general, as a first step, understanding where various activities will reside in the new CMEI is important for all stakeholders who are seeking to work with the agency.

(Note to readers, this is being crossposted with the DOE Alumni Network and views expressed in this post are entirely my own)

Technical progress alone doesn’t guarantee market success. An improved solar inverter or racking solution will change little if no one will buy it. Yet for decades, federal applied R&D programs measured success almost exclusively through technical milestones such as operational or installation efficiency gains and other technical performance improvements. Market relevance was commonly treated as an issue to consider in the future once a technology is proven.

The Department of Energy’s SunShot Incubator program proved there’s a better way. I started my DOE career working on the Incubator program and saw first hand how integrating commercial milestones into projects improved outcomes for taxpayers, companies, and the broader innovation ecosystem. The approach has since spread across DOE, validating what may seem counterintuitive: market feedback belongs in the lab.

The SunShot Incubator Model

Launched in 2007, the SunShot Incubator program was designed to help startup companies bridge the gap between initial concept and commercially viable prototype. The program was one of a suite of programs in the office supporting technology development and played a specific role related to technology commercialization, complementing other research efforts at labs, universities and other entities. Incubator provided early-stage financial assistance to help businesses cross technological and commercial barriers to getting a product to market while encouraging private sector investment. Over more than a decade, hundreds of companies received awards spanning photovoltaics, concentrating solar power, grid integration, system installation, and soft costs. The results speak for themselves. According to DOE’s Solar Energy Technologies Office (SETO), the program generated a 25x investment multiplier, with nearly $14 billion in follow-on funding tracked from 296 participating companies as of 2024.

But these outcomes didn’t happen by accident. They resulted from a deliberate shift in how the program defined and measured success.

From Technical-Only to Technical-Plus-Commercial

In its early iterations, then called the Photovoltaic Incubator program, the Incubator program structured projects around purely technical milestones. The program started at the National Renewable Energy Laboratory (now the National Lab of the Rockies) focusing mostly on improving solar photovoltaic cell efficiency. Given that program was developed by a lab, it is no surprise that technological progress took center stage. An award might require a company to improve solar cell efficiency by a certain percentage or demonstrate the throughput of a new manufacturing technique. These goals were measurable and scientifically rigorous, but insufficient.

The problem wasn’t the technical milestones themselves, it was what they left out. A company could hit every efficiency target while developing a product no customer wanted to buy because the appropriate commercial feedback loop did not exist. Technical success and commercial success were treated as more sequential rather than parallel.

Incubator program management was transferred to DOE in 2011. Led by my current colleague and former SETO teammate Victor Kane, the program was restructured to include commercial goals as part of each project milestone. The logic was straightforward: if companies weren’t in touch with potential customers and designing products to fit market needs from the outset, even the most impressive technical achievements would likely fail to translate into commercial impact.

This wasn’t about adding bureaucratic requirements. It was about building recurring market feedback into the development process itself.

Subscribe now

Push Back on Commercial Milestones

Integrating commercial milestones into projects wasn’t always smooth—resistance came from multiple directions.

Internal DOE Considerations

Initially there was discussion about whether commercial milestones were an appropriate role for government funding. We addressed this by framing the milestones as a mechanism for greater impact of federal funds: early market feedback could help ensure federal R&D investments ultimately reached deployment rather than languishing in the “valley of death.”

We also had to work with DOE legal and contracting staff to clarify that these milestone funds weren’t for marketing or promotional activities (which federal financial assistance regulations prohibit). Instead, the purpose was gathering critical feedback from potential customers and end users to guide R&D direction.

Awardee Hesitations

Awardees raised their own objections, which generally fell into several categories:

• “The technology will sell itself.” Some believed market traction would follow automatically once they hit their technical performance targets.

• “Competitors will steal our ideas.” Others worried that engaging with potential customers too early would expose their innovations to copying.

• “That’s not my job.” Some researchers saw themselves as technologists, not businesspeople, and viewed commercial activities as outside their scope.

• “This distracts from real work.” Related to the above, some felt time spent on commercial engagement detracted from technical progress.

• “End users can’t tell us anything useful.” A few dismissed the value of customer feedback entirely.

• “I am an expert and know more than anyone else.” A few thought they knew more than DOE or customers and did not see the value in thinking about commercial factors.

• “These milestones are out of our control.” Unlike technical milestones where success depends on the team’s own execution, commercial milestones often require action from external parties—potential customers, partners, or investors. Some awardees found this uncertainty uncomfortable. (More on how we addressed this below.)

While we did not agree with those points of view, we took the time to hear out and consider feedback like this. Dismissing them would have undermined buy-in and made the milestones feel like bureaucratic hoops rather than valuable learning opportunities. The key was taking the time to understand what would be valuable commercial milestones for that awardee during pre-selection interviews and award negotiations. The goal was agreeing on milestones that provided genuine and valuable insight while remaining achievable and within reasonable control of the project team.

Designing Effective Commercial Milestones

Not all commercial milestones are created equal. Poorly designed ones create busywork, such as spending valuable time calling 100 people about a technology. Well-designed ones accelerate commercialization, such as understanding product specifications from potential customers. Several principles emerged from our experience.

Negotiate, Don’t Impose

Milestones should be developed collaboratively with the awardee, not handed down from program managers. When companies help set their own commercial goals, they own those goals and are bought into their success. This creates mutual accountability and makes any future changes or project cancellations more of a shared decision than a unilateral action. A company that agreed to specific commercial targets understands why missing them matters. In most cases, the documentation and information prepared to meet a commercialization milestone should be directly usable and useful in conversations with potential partners, funders, and customers. Work to show commercialization progress should not be structured for the DOE alone.

Prioritize Third-Party Validation

The most valuable commercial milestones involve feedback from parties beyond DOE and the awardee. This might include:

• Letters of interest or support from potential customers confirming that the problem being solved is real and the proposed solution is attractive

• Independent review of performance and cost metrics validating that what the company is developing a product to solve is useful to potential buyers

• Third-party validation of cost models, such as analysis by a national laboratory, ensuring economic projections are grounded in reality rather than the optimism that can come from the technologists developing products

Third-party validation accomplishes two things. It provides companies with honest market feedback they might not otherwise seek. And it gives DOE confidence that taxpayer-funded projects are addressing genuine market needs.

• Less Useful Milestone: Call ten potential customers

  • This is fully in the awardees control and may not lead to useful feedback on the commercial viability of the technology

• More Useful Milestone: Get 3 non-binding letters of support from potential customers or users confirming the specifications you are designing to are useful

  • This requires a third party to think critically about the technology, provides actionable feedback and shows market interest.

Ramp Up Progressively

Commercial milestones should intensify as products mature. Early-stage projects might focus on customer discovery and letters of interest. As development progresses, milestones can require:

• Identifying demonstration sites (particularly important for products that require scale for meaningful testing)

• Third-party testing to validate that performance aligns with commercial customer requirements

• Manufacturing or production demonstrations that focus on key steps performed using representative methods and rates

• Ultimately, securing order commitments or deployment agreements

This progression mirrors the natural commercialization path while ensuring companies don’t postpone market engagement until it’s too late to incorporate feedback.

• Less Useful Milestone Progression: Nothing commercial for the first half of the project, then “Get 5 purchase orders for your product”

  • This waits too long to ensure what is being made is actually of use

• More Useful Milestone Progression: “10 non-binding letters of support”, “third party review of cost model”, THEN “Get 5 purchase orders for your product”

  • This progression builds on prior deliverables to ensure customers are aware of the product and that cost assumptions are rational before asking the awardee to get customer commitments.

Maintain Flexibility

Commercial milestones should be phrased broadly enough to accommodate technical pivots. A milestone requiring “validation from three potential utility customers via letters of support that the product addresses a significant operational need” allows for changes in the specific product being developed. One requiring “letters of support for the Model X-7 inverter that will be 1’ x 2’ x 1’ in size” does not. Unless the size is absolutely critical to its use case, milestones should focus only on critical performance factors.

Technology development is inherently uncertain. Markets shift. Customer needs evolve. Commercial milestones should test whether a company is building something the market wants without locking them into a specific technical path.

When Commercial Milestones Save Projects—By Ending Them

The clearest vindication of commercial milestones comes when they reveal market misalignment that technical milestones would miss entirely.

We funded a company developing a large, centralized inverter for utility-scale photovoltaic systems. The technology was promising, and the team executed well. Technical progress was solid by every conventional measure.

But the market shifted during the project’s lifetime. When DOE selected the project, centralized inverters were standard for large installations. By the time the technology matured, the industry had moved toward smaller inverters deployed in distributed configurations. This architectural shift offered advantages in fault tolerance, maintenance, and system flexibility that centralized approaches couldn’t match.

Prior to starting the award we negotiated a commercial milestone requiring the awardee to secure a letter of support from an entity willing to host a prototype test in a commercial environment. Despite persistent outreach, they couldn’t find a site. Every potential customer conversation revealed the same problem: the market conditions that existed when DOE selected the project had fundamentally shifted. No developer or utility partner would commit to hosting a demonstration. Not because of technical shortcomings, but because the business case had eroded.

This wasn’t a technical failure. The team had delivered on their engineering goals. But commercial milestones revealed what technical metrics couldn’t: the product no longer had a market.

DOE ultimately cancelled the project with the awardee’s support. Funds were preserved. The company redirected staff toward products with actual market potential. And taxpayers avoided continuing to fund development of technology no one would deploy.

This is exactly what commercial milestones are designed to accomplish. They surface market misalignment early enough to act on it, whether that means pivoting the technology, refocusing the customer base, or, when necessary, ending projects that can’t succeed commercially regardless of their technical merits.

This is only one example of times when commercial milestones saved government funds and awardee time. In addition, we got a lot of feedback from awardees who were skeptical of commercial milestones at first, but after the award got going saw the value and told us that engaging with potential customers, or others that could provide feedback on their products commercial viability, helped them create better solutions and speed their path to market confident they knew what customers needed.

Implications for Program Design

The SunShot Incubator experience suggests several principles for technology development efforts:

Start commercial engagement early. The instinct to wait until technology is “ready” for market feedback delays valuable information. Customer conversations at the concept stage can shape technical development in ways that dramatically improve eventual commercial prospects.

Make commercial milestones collaborative and transferrable. Imposed requirements generate compliance; negotiated goals generate ownership. Work with awardees to define commercial targets they believe they can achieve and are motivated to pursue. Structure the deliverables such that they are immediately usable and useful for partner, financier, and customer discussions.

Tie milestones to third-party validation. Self-reported progress is valuable but insufficient. External validation—from potential customers, independent labs, or industry experts—provides credibility and surfaces blind spots.

Preserve flexibility for pivots. Commercial milestones should test market fit without constraining technical approaches. Frame them around customer outcomes rather than specific product features.

Be willing to cancel technically successful projects. This may be the hardest lesson. When commercial milestones reveal fundamental market misalignment, continuing to fund technical development wastes resources and delays inevitable reckoning. Early cancellation, handled transparently, benefits everyone.

Conclusion

Technical excellence is necessary but not sufficient for commercialization. The SunShot Incubator demonstrated that integrating commercial milestones into early-stage R&D programs improves outcomes without compromising scientific rigor. Companies stay connected to market needs. Program managers get earlier signals about project viability. Taxpayers fund technology development with better odds of real-world impact.

The $14 billion in follow-on funding attracted by Incubator participants didn’t happen because companies developed technically impressive products in isolation. It happened because the program structure ensured those products were developed with commercial reality in view from the start.

For any organization funding applied R&D, whether federal agencies, state programs, foundations, or corporate R&D divisions, the lesson is clear: commercial milestones belong at the earliest stages of technology development, not as an afterthought once technical work is complete. The SunShot Incubator proved it works. The only question is why more programs haven’t followed suit.

Innovation Waypoints is brought to you by Waypoint Strategy Group.

In Part 1, I defined technical assistance (TA) as programs that connect organizations with expertise to evaluate, plan, and deploy technologies and walked through DOE models ranging from cohort networks like SEIN and SolSmart to multi-year partnerships like E2C and ETIPP. Those examples showed what TA looks like. This post addresses the factors impacting why some programs succeed where others fall short.

Not all technical assistance is created equal. A community that receives a few hours of expert time for a complex, multi-stakeholder solar deployment often walks away with partial understanding and incomplete information. Meanwhile, programs like SolSmart that have robust, integrated support for community needs have demonstrated measurable impact. SolSmart alone resulted in a 17% increase in solar installations in designated communities compared to peers. The difference lies in program design.

This post examines the principles that differentiate high-impact TA from well-intentioned but ineffective efforts. I’ll cover common misconceptions about what TA can and cannot do, the design choices that matter most, and practical guidance for two audiences: organizations building TA programs and those seeking to benefit from them. Whether you’re a federal program manager designing the next ETIPP or a rural cooperative trying to navigate your first clean energy deployment, the goal is to ensure that knowledge transfer translates to lasting capacity and real-world outcomes.

What Technical Assistance Is Not

While the first post spent a while defining what TA is for this discussion (I suggest going back to that post if you need a refresher or have not read it) I want to spend some time discussing what it isn’t. Common misconceptions can lead to poorly designed programs or misaligned expectations.

TA is not a substitute for direct technology funding. Technical assistance helps organizations make better decisions about technology, but it does not replace the capital needed to develop, acquire, install, or operate that technology. A community may receive excellent TA on energy deployment options, but still need grants, tax credits, and/or financing to actually build projects.

TA is not consulting. While consultants are hired to solve specific problems, effective TA aims for knowledge and skill transfer leading to recipient independence. While some programs may appear to act in this manner, the goal is to transfer knowledge and build capacity so that organizations can handle similar challenges on their own in the future.

TA is not one-size-fits-all. Generic webinars and downloadable guides have their place, but the most effective TA is tailored to the specific context, resources, and goals of each recipient. A rural electric cooperative in Alaska faces different challenges than a major metropolitan utility, and TA should reflect those differences.

TA is not a one-time transaction. While some questions can be answered in a single conversation, building organizational capacity typically requires sustained engagement. Programs that provide a few hours of expert time for complex, multi-stakeholder deployments often leave recipients with partial understanding and incomplete information. TA should allow for multiple engagements with experts overtime to ensure awardees are growing their knowledge and can further refine their understanding of challenges and solutions as they work through them.

TA is not just information dissemination. Publishing a report or hosting a webinar is not the same as providing technical assistance. Effective TA requires two-way engagement, iteration based on feedback, and adaptation to what recipients actually need to know. However, the outputs of a TA program can have information dissemination elements to help as many similar entities with similar challenges as possible, such was the case with SEIN.

Subscribe now

Principles for Effective Technical Assistance Program Design

For organizations designing new TA programs or improving existing ones, several principles emerge from DOE’s experience.

Meet Awardees Where They Are

Effective TA programs offer multiple entry points for organizations at different stages of readiness. The Energy to Communities (E2C) program exemplifies this with its three-track approach: in-depth partnerships for communities ready for sustained engagement; peer-learning cohorts for those seeking structured learning opportunities; and expert match for communities with specific near-term questions.

Early-stage communities may need help with goal-setting, feasibility assessment, and understanding what questions to ask. More advanced communities may need validation of specific technical approaches, support for financing negotiations, or assistance navigating regulatory processes. A single-track program design can run into challenges serving populations equally well.

An example of where this was an issue, but eventually resolved, is Solar Ready Vets. The original version of the program focused on providing training at military bases that specialize in transitions out of the military. While this made sense at first blush, it was soon discovered that the transitioning members were being connected to jobs around that base, not where they wanted to live long term. Future iterations of the program focused on facilitating training and job placement where the trainee wanted to be leading to more impactful outcomes for trainees and employers.

As part of the TA program development process, make time to understand the community you want to help and those that they will work with. Having input on the design process, or even co-designing a program, will lead to more effective outcomes and more satisfaction among participants.

Build in Peer Learning

The cohort model used by SEIN, ETIPP, and other programs creates valuable peer learning opportunities. When communities facing similar challenges work together, they learn not just from experts but from each other. A city that struggled with siting or utility interconnection processes can share hard-won lessons with peers, multiplying the impact of the expert facilitation.

These communities of practice often outlast the formal program. Relationships formed during cohort sessions become ongoing resources as participants face new challenges. The network effects can be as valuable as the direct technical support and stretch the impact of every dollar spent.

Plan for Sustainability

The ultimate measure of TA effectiveness is whether recipients can apply lessons independently. Programs should explicitly plan for sustainability by:

Creating replicable tools, frameworks, and guides that remain useful after direct engagement ends

Documenting lessons learned and sharing them broadly so that future communities benefit

Building internal capacity through training, not just solving immediate problems

Connecting recipients with ongoing resources and networks beyond the program

Make Expertise Accessible, Not Just Available

Technical expertise exists in a myriad of locations, such as, but not limited to, national laboratories, universities, industry associations, specialized consultancies, and within experienced practitioners. The challenge for TA programs is not finding expertise but making it accessible to organizations that lack the networks, resources, or knowledge to engage it directly.

DOE programs illustrate this well. The national laboratory system houses extraordinary depth across technologies and disciplines, but navigating it requires understanding which lab has relevant expertise, how to initiate engagement, and what resources are needed. Programs like E2C and ETIPP solve this by serving as intermediaries that match community needs with appropriate researchers and managing engagements to keep them community-focused.

This intermediary function applies regardless of the expertise source. Whether your program leverages university researchers, industry consultants, or peer practitioners, effective design includes:

• Pre-established relationships with expertise providers, reducing lead time for recipients

• Intake processes that translate recipient challenges into actionable expert engagements

• Active management of expert-recipient interactions to ensure relevance and prevent scope drift

• Clear expectations for both parties on deliverables, timelines, and communication

Expertise that recipients cannot effectively access is expertise that does not exist for practical purposes. Program design should treat accessibility as seriously as availability.

Consider Hybrid Approaches

Technical assistance can be made more impactful when delivered in combination with other funding mechanisms. The Community Power Accelerator Prize combined TA with prize competitions. ETIPP provides both in-kind technical support and direct funding for implementation. The Solar Ready Vets Network connected workforce development TA with employer engagement and credentialing programs.

These hybrid approaches recognize that knowledge alone may not be sufficient. Organizations or individuals often need resources to act on what they learn. By combining TA with funding from DOE or a line of sight to funding from others, programs can ensure that capacity building translates to concrete and enduring outcomes.

Getting Started

For Organizations Designing Technical Assistance Programs

If you are creating a new TA program, start with these questions:

Who is your audience? Are you serving communities, developers, utilities, workforce organizations? What is their baseline knowledge and capacity?

Who should be involved? Are you trying to increase capacity for multiple entities at once? Are you trying to bring different entities together to build connective tissue between them? Who is the appropriate entity to lead the effort, if it has multiple stakeholders?

What outcome are you seeking? Do you have a clear understanding of the gap to be addressed and why it would be impactful to your goals? Is the goal immediate problem-solving, long-term capacity building, or both? How will you measure success?

What level of engagement is appropriate? Do recipients need sustained partnerships or targeted expert consultations? How can the program help provide enough space/time/money to give stakeholders the ability to engage?

How will you facilitate peer learning? Can you group recipients into cohorts? How will you capture and share lessons across the program?

What happens after the program ends? How will recipients sustain their new capabilities? What resources will remain available?

For Organizations Seeking Technical Assistance

If your organization could benefit from technical assistance, consider:

Define your challenge clearly. What decision are you trying to make? What knowledge gaps are preventing progress?

Assess your readiness. Do you have staff capacity to engage with technical experts and implement recommendations?

Identify the right program. DOE has offered numerous TA programs across technology areas and community types. Programs like E2C, ETIPP, and the Solar Energy Innovation Network each served different needs. Also consider pathways beyond federal government support such as programs from cities, states or non-profits.

Engage early. Many programs have application windows and limited capacity. Building relationships with program managers before formal applications open can improve your chances and help identify the best fit.

Conclusion

Technical assistance is essential for acceptance, adoption and deployment of emerging and advanced energy technologies and solutions. While most DOE funding supports the development of new technologies and solutions, TA ensures that communities, organizations, and individuals have the knowledge to use them effectively. It is the bridge between what is technically possible and what actually gets built.

The most effective TA programs share common features: they meet recipients where they are, build peer learning into the design, plan for long-term sustainability, leverage deep expertise, and often combine knowledge transfer with other forms of support. Programs like E2C, ETIPP, SEIN, and the Community Power Accelerator Prize demonstrate these principles in action.

For organizations seeking to deploy clean energy technologies, technical assistance offers a path to navigate complexity without building all expertise in-house. For organizations designing support programs, TA provides a model for impact that extends far beyond individual grants or projects.

The knowledge base for adopting and using technology is as important as the technology itself. Combined with funding and communities of practice, technical assistance creates powerful programs that prepare stakeholders for clean energy’s introduction to the market and beyond.

Subscribe now

Innovation Waypoints is brought to you by Waypoint Strategy Group.

Introduction

The U.S. Department of Energy (DOE) and similar funding organizations exist primarily to support technology development. But developing a technology is not the same as deploying it. For new technologies to achieve market impact, stakeholders, such as communities, utilities, states, and workforces, need the knowledge and capacity to put them to use. Technical assistance programming bridges this gap.

Defining technical assistance (TA) precisely matters because the term is often used loosely to describe any effort that builds technical knowledge. Under this broad interpretation, TA could mean helping a community understand solar economics or helping a startup refine its manufacturing process. This discussion focuses on a specific subset: TA as programs that give organizations access to specialized expertise and resources for evaluating, planning, and deploying new technologies. Technology development support, bringing products from concept to commercial readiness, is a separate undertaking and will not be addressed here.

Designed correctly, TA is a force multiplier that provides training and expertise to help organizations, such as communities, cities, states or private sector entities, understand a new technology, conduct vetting and energy system planning, build capacity, and learn how to exploit it for their needs. TA accelerates the deployment and understanding of new technologies by bringing expertise and/or capacity to non-expert actors. In addition, it can be made even more effective by creating communities of practice to share lessons among similar entities. Combined, powerful programs can be created to more effectively prepare those involved in using or integrating new technologies.

During my 13+ years working in the DOE’s Solar Energy Technologies Office, we used TA regularly to help accelerate the deployment of solar energy technologies by bringing together stakeholders to increase understanding of the technology and facilitate deployment. During this period the industry saw a massive acceleration of solar energy deployment in the US. While steep reductions in hardware costs were a critical driver, it was imperative to develop and run programs to increase the capacity of a wide range of actors to ensure they understood solar energy and could work with one another to increase deployment in as friction free a process as possible. The examples I provide below are from the Solar Energy Technologies Office or cross office efforts related to renewables. While the examples are solar/renewables centric, the lessons learned about effective TA programming are widely applicable to government and non-government TA program providers alike.

This post is the first of two exploring technical assistance at DOE. Together, they will define what TA means in this context, provide examples of different program models, identify what distinguishes effective TA from ineffective efforts, and offer guidance for organizations designing or seeking such programs. This first post focuses on definitions and examples. The second will turn to program design principles and practical advice for applicants.

What Is Technical Assistance?

By my definition from above, technical assistance encompasses the knowledge transfer needed to move products from the manufacturing line or off the screen into communities and markets. Rather than funding technology development, TA programs fund expertise that helps organizations make informed decisions about technologies already available to them.

These TA programs share several core design principles. They support entities that lack technical depth on a topic, connecting them with subject matter experts who can help solve near-term challenges while also building transferable knowledge for future use. Funding flows to the TA provider, reducing barriers for resource-constrained organizations, and sometimes to the recipient themselves to allow them to acquire needed capacity to engage on the implementation of new products. And increasingly, these programs cultivate communities of practice where recipients share lessons learned and best practices, ideally extending beyond the funding period itself.

This last element is particularly important. The most effective TA programs recognize that peer-to-peer learning multiplies the impact of expert engagement. When a city in Ohio learns from a city in Colorado that faced similar challenges, the knowledge transfer extends far beyond what any single consultant could provide.

A note on terminology: technical assistance can sound similar to voucher programs (discussed previously). In essence their difference is the manner in which the assistance is delivered. Voucher programs generally provide access to capacity at a technical organization, predominantly national labs, for a well-defined service. The awardee is given access to that entity and it is up to them to leverage it. TA programs as discussed here, are programs that tend to leverage broader expertise than labs alone, have multiple entities involved (instead of just the voucher awardee and the technical provider), and are meant to create enduring capabilities over potentially one off use of a technical facility. TA may also include funding to program recipients as well to facilitate participation in the TA.

Subscribe now

Key Elements to Technical Assistance Programs

The intention behind TA programs is to create a combination of access to expertise and the resources necessary to help communities navigate certain challenges.

There are a few key critical elements to DOE TA programs.

1. National Laboratory Expertise and Facilities: Navigating the national labs can be challenges for outside parties, but the labs house deep expertise on technical and non-technical subject areas related to the deployment of technologies.

2. Private Sector Partner Expertise and Facilities: In addition to the resources at the labs, some TA programs will pull in other subject matter experts if they cannot be found at the labs. This could include standards groups, financiers, installers, or other expertise necessary to help with energy planning and deployment.

3. Funding Resources: Some TA programs also provide direct funding to groups participating in programs. This can help organizations/communities hire more surge direct support to go through the program, or help augment operational resources.

4. Facilitators: To facilitate peer learning, TA will bring cohorts together to share lessons learned. Having professional facilitation can increase the probability of sharing impactful, usable information during in-person or virtual gatherings.

5. Community Organizations: For TA programs that impact more remote or rural communities, having trusted intermediaries as a part of the TA program can have enormous value. Having organizations embedded and familiar to communities can ease implementation and create better communication between the TA technical providers and recipients.

The next post in this series will cover more about how to combine these various elements and what decisions program designers should make when creating TA programs.

Models for Technical Assistance at DOE

DOE has developed a range of TA delivery models, each suited to different contexts and community needs. Understanding these models can help both TA seekers and designers identify the right approach.

Innovation Networks and Cohorts

Solar Energy Innovation Network (SEIN): This program assembled multistakeholder teams representing utilities, local governments, community organizations, and other partners to research solutions to real-world solar adoption challenges. Over 18 months, teams received direct funding, analytical support from national laboratories and other industry experts, and facilitation through multiday working sessions by trained staff. Crucially, teams were grouped into cohorts based on shared challenges, enabling the cross-pollination of ideas.

SEIN’s companion initiative, Innovation at SCALE (Solar Community Assistance for Local Equity), extended the insights from completed SEIN projects to new communities. This model recognized that innovations developed in one context often apply elsewhere with adaptation. Rather than reinventing the wheel, communities could build on proven approaches.

The cohort model creates a multiplier effect: teams learn not just from experts but from each other. For example, when Tampa Bay developed a framework and toolkit for enhancing resilience through solar plus storage, those insights became available to communities, cities and states nationwide through targeted technical assistance.

SolSmart: Launched in 2016, SolSmart is a national recognition and technical assistance program that helps local governments reduce barriers to solar energy development. The program provides no-cost technical assistance to help communities streamline permitting, improve planning and zoning, enhance community engagement, and develop local solar markets. Communities achieving designation receive national recognition at Bronze, Silver, Gold, or Platinum levels. To date, over 570 local governments across 44 states have achieved SolSmart designation, representing one in three Americans. Independent research found that SolSmart-designated communities saw an average of 17% more solar installations per month than non-designated communities. The program has since expanded under DOE’s Energy Ready initiative to include Charging Smart for electric vehicle infrastructure and Distributed Wind Smart for sub-utility wind assets.

Interconnection Innovation e-Xchange (i2X): Originally led by DOE’s Solar Energy Technologies Office and Wind Energy Technologies Office, i2X convenes diverse stakeholders to enable simpler, faster, and fairer interconnection of energy generation resources while enhancing grid reliability and security. Interconnection queues have long been an issue for utilities, grid operators and generation developers. DOE had tried funding technical solutions, such as new modeling products for grid entities, or providing TA and tools to help system developers understand the best locations on a grid to deploy. However, these efforts came up short and it was clear a more coordinated, multi-stakeholder program was needed. i2X facilitates stakeholder engagement, data collection and analysis, strategic roadmap development, and technical assistance. i2X has developed both a Transmission Interconnection Roadmap and a Distributed Energy Resource Interconnection Roadmap to guide stakeholders in addressing current interconnection challenges. The program brings together utilities, grid operators, state and local governments, energy industry groups, related experts, and others to share data, develop roadmaps, and implement solutions to improve interconnection procedures.

Multi-Year Deep Partnerships

Energy to Communities (E2C): This ongoing program offers three tiers of support with the goal of ensuring communities have de-risked plans for meeting their clean energy goals through access to national laboratory expertise and cutting-edge analysis. In-depth partnerships provide up to three years of sustained engagement, including direct subcontract funding and extensive technical support from DOE’s national laboratory complex. Peer-learning cohorts bring together communities with shared challenges for facilitated group learning. Expert Match connects communities with 40-60 hours of targeted assistance to address specific near-term questions.

Energy Technology Innovation Partnership Project (ETIPP): Focused on coastal, remote, and island communities facing unique energy challenges, ETIPP offers two tracks: strategic energy planning (4-8 months) and deep-dive technical assistance (12-24 months). Selected communities receive up to $50,000 in direct funding alongside substantial in-kind support from national laboratory researchers at NREL, Lawrence Berkeley National Laboratory, Pacific Northwest National Laboratory, and Sandia National Laboratories.

The key design feature of these programs is their community-driven approach. Communities have the financial capacity to engage deeply and define their priorities, and technical experts adapt their support accordingly. This stands in contrast to prescriptive programs that tell communities what they should want.

Workforce Development Technical Assistance

Some TA programs focus on building human capital rather than project development:

Solar Ready Vets: Recognizing that military veterans bring valuable skills to the clean energy workforce and that employers were looking for qualified, reliable candidates, this program connected transitioning service members with training, certifications, and employment pathways in the solar industry. The network provided resources for both veterans seeking careers and employers looking to recruit military talent, including SkillBridge programs and registered apprenticeship pathways.

EMPOWERED (Education Materials for Professional Organizations Working on Efficiency and Renewable Energy Developments): About a decade ago DOE saw that the deployment of solar, building and vehicle technologies was or could be hindered by the emergency response or permitting communities if they did not understand the technologies adequately. An example of this is constraints that were imposed on battery developers in New York City due to safety concerns among officials. The program was developed and run by multiple offices (solar energy, buildings technologies, and vehicles technologies) to address this issue by working with first responder and building official organizations nationwide to build a greater understanding of these technologies. Technical Assistance was provided to or by project partners (Interstate Renewable Energy Council (IREC), International Code Council, and National Association of State Fire Marshals) to develop and vet educational resources for their stakeholders. This content is now available through CleanEnergyClearinghouse.org and CleanEnergyTraining.org. It reached over 75,000 building and safety professionals across all 50 states with training on distributed energy resources including solar, energy storage, electric vehicles, and grid-interactive buildings.

These workforce-focused programs demonstrate that TA extends beyond project-specific support to building the broader ecosystem required for technology deployment.

Hybrid Models: Technical Assistance Plus Prizes

Some of DOE’s most innovative programs combine technical assistance with prize competitions, using cash awards to build capacity while TA ensures that capacity translates to lasting capability.

Community Power Accelerator Prize

The Community Power Accelerator Prize illustrated how TA and prizes can work together. This $10 million competition was designed to fast-track new and emerging solar developers, but it is structured to build organizational capacity, not fund projects.

The prize unfolded across three phases:

• Phase 1 (Ready!): Teams submitted initial project portfolios and received $50,000 upon selection for staff capacity to engage in the prize stages.

• Phase 2 (Set!): Teams completed instructor-led Learning Lab courses, work with technical assistance coaches to refine project profiles, and develop investor pitch decks. Winners receive $200,000 to continue to be able to dedicate staff capacity to learning about project development and financing

• Phase 3 (Grow!): Teams secured investment commitments for their projects through the Community Power Accelerator platform, competing for final prizes.

Image From: Round 3 November informational webinar

What made this model powerful is the integration. Prize money provided the resources teams needed to hire staff and expand operations. Technical assistance built the skills to use those resources effectively. The online platform proposed projects needing funding creating an ecosystem connecting developers with investors and philanthropic organizations. By the end, winning teams had not just completed projects; they had become capable organizations positioned to develop community solar at scale.

The Community Power Accelerator demonstrated that TA is not just about answering questions. It is about building institutions that can continue the work long after DOE’s direct involvement ends.

Conclusion

This post established a working definition of technical assistance, programs that transfer knowledge to help organizations evaluate, plan for, and implement technologies already available to them, and illustrated the diversity of forms this can take. From cohort-based learning networks like SEIN to multi-year deep partnerships like E2C and ETIPP, from workforce development initiatives to hybrid prize-plus-TA models, DOE has developed a toolkit of approaches suited to different stakeholder needs and deployment challenges.

This diversity is intentional. Deployment barriers vary by technology, geography, and stakeholder type, and effective TA must be tailored accordingly. What these examples share is a common purpose: reducing barriers by building capacity among the actors who must ultimately adopt, permit, install, operate, or integrate new technologies. SolSmart streamlines local permitting processes. ETIPP helps remote communities develop energy plans. The Community Power Accelerator builds emerging solar developers into sustainable organizations. Each addresses a different barrier, engages different stakeholders, and operates on different timescales.

The next post will examine what separates effective TA programs from ineffective ones, exploring the design principles that distinguish high-impact programs and offering practical guidance for both program designers and organizations seeking technical assistance.

Subscribe now

Innovation Waypoints is brought to you by Waypoint Strategy Group.

“In complex systems, cause and effect are not closely related in time and space.” — Peter Senge

Program evaluation is a foundational part of good governance. It’s how organizations assess impact, learn from experience, and improve future efforts. At the Department of Energy, evaluation is taken seriously: programs are designed with multi-year goals, results are reported to Congress through the Government Performance and Results Act (GPRA) processes, and there is genuine cultural commitment to learning and doing better over time.

And yet, in practice, evaluation in energy programs often struggles to deliver on its promise.

This is due, at least in part, to the fact that DOE energy programs operate in what behavioral economist Robin Hogarth calls a “wicked learning environment”—one in which learning from experience is unreliable, feedback is delayed or misleading, and outcomes are difficult to attribute to skill rather than luck. David Epstein later popularized the concept in Range: Why Generalists Triumph in a Specialized World, contrasting fields where expertise develops reliably with those where it doesn’t.

Understanding that wickedness—and its implications—is essential if we want evaluation to meaningfully inform energy policy and investment, and if we want to avoid wasting effort on evaluation when it will not.

Kind vs. Wicked

Hogarth distinguishes between “kind” and “wicked” learning environments.

In kind environments, rules are stable, feedback is timely and accurate, and cause-and-effect relationships are relatively clear. Chess is a classic example—you know almost immediately whether your move worked. So is tennis. Learning is straightforward: observe what worked, adjust, and improve.

Wicked learning environments are the opposite. They are characterized by:

• Rules that are unclear or changing. There may be no rules at all, or they may change without notice.

• Data that is messy. Information is often incomplete, inconsistent, or misleading.

• Feedback that is delayed or inaccurate. It’s difficult to connect an action to its consequence.

• Evolving contexts. What worked in the past may not work in the future due to volatile, game-changing events.

• Difficulty distinguishing skill from luck. Outcomes can be misleading, and it’s hard to tell if success is due to ability or chance.

The stock market is a classic wicked environment. So is politics. And, as it turns out, so is federal energy program management.

Why Energy Programs Are Wicked

To set the stage, this discussion focuses on applied research where the expectation is market impact, not scientific research where the explicit goal is to increase knowledge in a domain. Consider what it takes to develop and commercialize hard technology. A typical DOE award lasts 2-4 years for earlier-stage work, 5-10 years for later-stage demonstrations. Most technologies need three or four projects across different stages to really reach the market. From initial research to commercial deployment, we’re often talking about 15-20 years.

And the world does not stand still. Over those same time horizons:

• Political administrations turn over

• Policy environments shift

• Competing technologies advance

• Markets evolve in unpredictable ways

• International competitors move quickly

A cost target set in 2010 may be meaningless by 2025, either because the market moved faster than expected or because the competitive landscape changed entirely. The assumptions that shaped program design may no longer hold by the time outcomes are visible.

This creates a fundamental problem: the feedback loops are too slow. By the time you know whether a program actually worked—whether it moved the needle on commercialization, market adoption, or system transformation—the people who designed it are long gone. The context that shaped the original decisions has changed. And the lessons, even if someone bothers to extract them, may not apply to current conditions.

This is the essence of wickedness: the learning signal is delayed, distorted, missed, irrelevant, or all of the above.

The Culture of Evaluation at DOE

DOE frequently talks internally about the importance of evaluation and impact. The Department has multi-year goals reported to Congress through GPRA. There is genuine commitment to learning and improving.

But this commitment runs into two challenges. One, it consistently takes a back seat to more immediate pressures: launching programs, obligating funds, overseeing projects, responding to near-term oversight demands. Staff resources are always stretched thin. The urgent crowds out the important. And evaluation work—the slow, careful analysis of what actually happened and why—never quite makes it to the top of the priority list. Two, GPRA goals tend to be high level, such as reducing the installed cost ($/Watt) of solar energy. While this kind of goal can help track an industry’s evolution, it is too high-level to assess the impact of DOE funded technologies.

This creates predictable problems.

Staff and leadership want wins. So they focus on metrics that are easily tracked and show leading indicators of engagement rather than metrics that would actually demonstrate system impact. The easy metrics include things like number of applications submitted, competitiveness ratios, cost share amounts, and technology performance against interim targets. These are useful operational indicators, particularly for early R&D, but they’re not the same as commercial impact.

Impact data isn’t defined up front—and collection ends when projects do. The data required to assess long-term impact is often not formally included in project reporting requirements. When it is included, the burden on performers is high enough that compliance is spotty. You can’t evaluate what you didn’t measure.

But even when the right data is defined, there’s a structural problem: for most projects, reporting requirements end when the project does. Typical DOE contract structures haven’t included extended data collection, either by DOE staff or by a third party. The award closes out, the performer moves on, and the thread is lost. By the time the outcomes that actually matter start to materialize—commercial traction, follow-on investment, market adoption, workforce effects—there’s no mechanism in place to capture them.

This is an opportunity for Other Transactions* (OTs) and other flexible contracting mechanisms. Unlike traditional grants and cooperative agreements, OTs can be structured to include post-project data collection requirements, third-party evaluation relationships, or other provisions that extend visibility beyond the award period. The metrics that would actually matter—relative competitiveness against incumbent solutions, follow-on investment and commercial deals, expansion of organizational capacity, workforce development and spillover effects—require this kind of extended visibility. They unfold over years, not months. Without contractual structures that maintain the relationship past project closeout, they mostly don’t get tracked.

*OTs as a funding tool will be discussed in a future post

Staffing changeovers break continuity. Over the duration of a multi-year program, staff turn over. New people inherit projects without full context on the core program goal and original project intent. It becomes easy to rigidly follow the written plan even when the underlying logic would allow—or demand—that objectives shift.

Results from previously funded projects frequently do not get housed in adequate data repositories - they are passed on through either peer reviews or through lessons learned acquired by staff who have been around long enough to see the various cycles, which increases sensitivity to staffing changes.

Project outputs get checked; program outcomes don’t. There’s a clear process for verifying whether a project delivered its technical milestones. There’s much less infrastructure for tracking whether the portfolio of projects actually achieved the program’s intended market or system impact. DOE has historically funded some third-party retrospectives, but data is fragmented and incomplete, limiting the conclusions that can be drawn. Or outcome-based evaluations through traditional impact evaluation call for a counterfactual, which can be nearly impossible - whether that’s because of data privacy rights of applicants that did not receive funding, or just finding a corollary to a project in the private sector that did not receive federal funds.

Feedback loops are too slow to inform current decisions. Even when evaluation does happen, the signals arrive too late to meaningfully influence program design within a single administration. You’re often flying at least partially blind, making decisions based on incomplete data from programs that operated in different contexts.

This challenge became particularly visible during the implementation of major, time-sensitive efforts like those funded under the Bipartisan Infrastructure Law (BIL). Extensive review and approval processes, overlapping equities, and limited internal capacity slowed execution. Meanwhile, external conditions continued to change—sometimes faster than programs could respond. For these programs, like many prior DOE efforts, evaluation frameworks often assumed a kind learning environment where none existed.

The Attribution Problem

Beyond the timing issues, there’s a deeper problem: attribution. When programs fail to achieve their desired market impact, the failures get blown out of proportion and attributed to poor decision-making or technology failure—not to market shifts, policy changes, or system-level dynamics that were outside anyone’s control.

This creates perverse incentives. Program managers may seek to avoid anything that might fail visibly, even if taking that risk would have been the right call given the information available at the time. Similarly, they may avoid drawing attention to ambivalent or negative results, even if they could meaningfully inform future decisions.

Project successes are often similarly misattributed or incompletely considered. Offices may be encouraged to claim credit for successes that may have had little to do with their decisions, or were likely due to multiple activities from multiple offices and entities. The role of luck and favorable shifts to policy, market conditions, or other landscape actions are discounted. The feedback, such as it is, reinforces the wrong lessons.

Both situations disincentivize program evaluation - the positive outcomes are perceived to be self-apparent and easy to tell a story around, while more detailed assessments on opportunities to improve invite criticism and negative conclusions. This turns end-of-award evaluation into a clearance exercise for execution and compliance—checking boxes—rather than an analytical one. It’s a stark contrast to the rigor applied during continuation reviews, where real scrutiny happens. Impact becomes a “who cares” issue. Not because no one values it, but because no one is incentivized to own it at the end. This issue is compounded by the limited bandwidth of DOE staff and their need to move onto the next program or project to ensure the work the office is measured against gets done.

The Trust Connection

This connects to themes we’ve explored in prior posts. In our discussion of trust in federal energy programs, we noted that without strong relationships between program staff and performers, it’s difficult to weather the uncertainty inherent in long-duration technology development. And in our post on systems approaches to risk, we explored how DOE manages compliance and execution risk well but often incompletely addresses impact risk—the risk that programs fail to change the energy system in meaningful ways on a timeframe that matters.

The wicked learning environment compounds both problems. “Trust but verify” is the standard mantra. But verify what? Original targets may no longer be relevant. Technical performance requirements may have shifted with the market. The goalposts move, and without trusted relationships that allow for honest conversation about evolving value, everyone defaults to checking boxes against outdated plans.

The hard conversations—”this project isn’t working but we learned something important,” or “the market moved and we should pivot”—require trust to be productive. In its absence, performers tell program managers what they want to hear, program managers report upward what leadership wants to hear, and everyone loses the chance to actually learn.

What To Do About It

The practical question isn’t whether evaluation matters in the absolute—it does. The question is whether additional evaluation effort will actually change decisions in the timeframe that matters. Sometimes the honest answer is no. In those cases, the better investment is focusing on program design up front and then staying the course.

If you care about impacts in the next five years, you probably aren’t going to get new data that allows you to meaningfully change approach in time.

The feedback loops are simply too slow. In that case, the right investment is in collaborative program design at the outset—bringing in diverse perspectives, pressure-testing assumptions, and making the best decisions you can with available information. This is why we’ve emphasized program design as a craft learned through apprenticeship—the upfront work matters precisely because mid-course correction based on impact data is often not possible.

In these cases, program evaluation still has a place, but it’s more to build a body of knowledge around technology and market progressions than to inform near term decisions. The level of effort and burden should be set accordingly.

If you care about impacts on the decadal scale, different approaches are needed:

Establish an independent evaluation function. Program evaluation will always take a back seat to time-sensitive operational work. Existing staff will never find time to prioritize it. An independent group—within DOE or external to it—with dedicated responsibility for tracking long-term impacts is the only way to sustain attention on this work. The retrospectives this group produces can gradually adjust DOE and other government programs over time, even if they can’t inform any single administration’s decisions.

Establish different targets for different stages. Cost and performance targets are useful for early-stage work, where you’re trying to establish potential for competitiveness and demonstrate forward progress. They’re less useful for later-stage commercialization, where the cost-value proposition in real markets is what matters. Measure alignment with industry needs in the short to mid term—deals, follow-on financing, project announcements. Measure relative value, not just cost. Focus on system value, not just asset value.

Build data collection into the relationship. The independent evaluation group needs a formalized relationship with project performers that gives them the ability to get the data they need over extended timeframes. Make this a condition of government support. Use the flexibility available in contracting mechanisms (like Other Transactions) to build in these requirements without excessive burden. And ensure the evaluation group has a formalized relationship with DOE offices to inform how program objectives are defined and what data gets collected.

Accept that some learning will be retrospective. The best we can do for many programs is enable good retrospective analysis that benefits future programs, even if it can’t benefit the current one. That’s still valuable—but only if someone is actually doing the retrospectives and the institution has mechanisms to absorb the lessons.

Communicate explicitly. Be clear from the start about how evaluation will be conducted, how it will be used, and who is accountable for carrying it through. Ambiguity about ownership is how evaluation becomes performative—generating activity without insight.

The Honest Version

Here’s the uncomfortable truth: for most federal energy programs, we will never know with confidence whether they “worked” as well as they could have. The time horizons are too long, the confounding variables too numerous, the counterfactuals too unknowable. We can track outputs. We can measure some outcomes. But definitive, complete attribution of system-level impact to specific program decisions? That’s probably beyond reach.

This doesn’t mean evaluation is pointless. It means we should be realistic about what evaluation can and cannot deliver. It means we should invest in program design—in getting the logic right up front, in building trusted relationships that allow for adaptation, in constructing portfolios that can succeed across a range of scenarios. And it means we should be humble about claiming credit or assigning blame after the fact.

The learning environment is wicked. We can make it somewhat less wicked through better processes and sustained attention. But we can’t make it kind. The best we can do is acknowledge the wickedness and design our institutions and learning systems accordingly.

Innovation Waypoints is brought to you by Waypoint Strategy Group.

The Department of Energy - among other agencies in the federal government - houses and owns some of the most unique and complex technical facilities. Equipment you cannot access anywhere else in the world. From particle accelerators, to wave tanks and solar module testing equipment. They’re housed at the 17 different National Laboratories, spanning the country.

And it’s not just the labs that have incredible equipment and talent. Universities across the United States, and private industry labs, host a wealth of equipment and expertise that has been built and is maintained to advance technologies from across the energy and broader deep tech ecosystem.

But often this equipment and expertise is inaccessible to entrepreneurs and researchers outside of the institution that houses that equipment. The lack of access to this equipment stalls progress. Validation facilities that could be used to derisk companies’ technologies are out of reach. Requests for third party validation to companies cannot be fulfilled without access to this equipment. Expertise that could help address key technical questions remains out of reach.

Increasing the availability of this “behind the walls” equipment and expertise can help accelerate progress, facilitate investment into technologies and entrepreneurs, and leverage the investment made primarily by the federal government in these facilities and equipment at national labs. Pairing innovators, communities, corporations, you name it with the facilities they need to validate, test, or progress innovation is such a critical component to commercialization and adoption.

And so how do you do that? Well, to address this issue, we created vouchers. What’s a voucher? It’s basically a mechanism we created that allows the agency to pay for services at a lab on behalf of a company (or community, or other recipient). You receive a voucher for services (a.k.a. in-kind) that you need. By having an agency or a funder pay, that eliminates the need for companies to pay out of pocket for these services. It can also (theoretically) reduce the friction inherent in the system for paying for services at places like the labs for companies - which can be a particular challenge for startups. By serving as an intermediary, the government (or a designated intermediate) can match the companies that need access to services and expertise to the facilities that can provide it. This also creates a reciprocal value of the laboratories as they build relationships with innovators and are exposed to cutting edge technologies.

But managing that facilitation process requires key programmatic design. How do you connect companies with these services and facilities? What’s the best mechanism? What are the key roadblocks that would prevent startups from accessing these facilities?

These were all things we worked through to unlock access to facilities nationwide. This post explores how we implemented vouchers and recommendations for the future.

A Bit of Voucher History

Back in 2012, I took over management of a program called the Innovation Ecosystem initiative. Launched in 2010, the grant had seeded five approaches to supporting commercialization. There were several university commercialization approaches, an incubator, and an award with the Fraunhofer USA in Boston.

Under a 3-year, $1 million award, Fraunhofer established Fraunhofer TechBridge. The idea was to pair startups with Fraunhofer’s industry-grade facilities and expertise to design and execute testing, validation, and demonstration projects. I still remember visiting their sparkling new facility in downtown Boston, viewing the equipment that was validating the performance of inverters, new electrical panels, advanced building technologies, and more.

As we were planning to launch what would come next, the Technology to Market team that I was on held a peer review of the previous awardees. And TechBridge was highly reviewed, near the top supporting startups in the three year time period. What we heard was that access to these facilities provided the independent validation was an accelerant for companies. (I should note it clearly has shown value to Fraunhofer as well as TechBridge still continue to this day).

At the same time as the TechBridge program was concluding, we were also in the process of standing up the Lab Impact Initiative, which sought to increase the impact of the labs on commercialization. One of the ideas that was being proposed was similar to TechBridge, but on a bigger scale - unlocking the national laboratories by standing up a voucher program.

What ultimately was rolled out was the Small Business Voucher initiative. This was a full, cross-office program that included the labs and every office within the (fka) Office of Energy Efficiency and Renewable Energy. The program catalyzed relationships with hundreds of startups and the labs. More than 1200 companies expressed their interest through the Request for Assistance. However, the program was sunset after three rounds due to a change in Administration.

Through these early pilots vouchers had proven to be a useful tool. They were integrated as part of the winding story for prizes, as we had identified vouchers as key tools to support startups. So we used vouchers in the Solar Prize through the American Made Network. These vouchers allowed prize participants that got past the Ready! Stage to work with a national lab of their choice to help further their technology in preparation of the next phase of the competition This was also the first time that we expanded to not just include the labs, but private facilities nationwide. Then in the Water Power Office (WPTO), we used vouchers as critical tools in prizes like the Fish Protection Prize - which focused on downstream fish passage around hydropower facilities. These vouchers enabled participants to work with the National Labs to advance their solution to the final stage and vouchers were offered as part of the prize pool itself.

And in WPTO, building upon the Solar Prize and the concepts developed in the Fish Protection prize, we launched the TEAMER - or Testing & Expertise for Marine Energy - program. This next iteration of vouchers unlocked further potential via improvements in program design. TEAMER ran Requests for Technical Services (RFTS) with streamlined and simplified application processes easing the burden of access resources on applicants. In parallel, WPTO had a rolling application process to make more test facilities available to TEAMER participants. An additional change was having it run by a private sector organization - POET - who facilitates the process, lowering the lift on WPTO staff. Finally, TEAMER would post regular reports on uses of the TEAMER facilities to more quickly disseminate any lessons learned. Below is an illustration of how the RFTSs work.

We also dramatically improved how fast these vouchers could be distributed, with the process illustrated below:

While the models above appear simple, getting the architecture and structure ready for these programs took a lot of design and work. Ironically moving fast in government is…..hard.

And in 2023, vouchers got another huge boost. We launched the first Partnership Intermediary Agreement (PIA) through EnergyWerx and one of the specific lines of work we asked was to run a voucher program. In partnership with the Office of Clean Energy Demonstrations and others, fueled by funding from the Infrastructure Investment and Jobs Act through the Technology Commercialization Fund, these expanded vouchers could support the vetting and acceleration of demonstration-ready companies. This moved beyond the traditional early-stage company support, and expanded to commercialization assistance, permitting and siting, and expanded validation efforts. Through this new Technical Voucher Program, we created even more innovation in vouchers execution.

DOE Technical Voucher Program Structure

So how does this process work?

First, we start with facilities. Voucher providers describe their capabilities. This included: (1) Testing and Validation Support; (2) Commercialization Support; and (3) Permitting and Siting Support. The DOE Technical Voucher program expanded beyond the traditional testing and validation to include commercialization and permitting support.

Once facilities and their capabilities are evaluated and catalogued, these capabilities are then advertised to potential recipients. Through a simple application form, recipients apply to use specific capabilities. And then the intermediary (private third-party like POET in TEAMER), in consultation with DOE, awards the vouchers.

Here are some examples of the types of matches.

Seems simple, but this is a complicated process.

And popular.

In 2024, DOE announced that over 170 companies, tribes, and local governments would receive support to advance commercialization and facilitate demonstration and deployment of clean energy technologies in the form of DOE-funded vouchers. Almost 60 organizations, including six National Labs, were selected to provide this support valued at more than $20 million.

And in the last year, as part of the Senate Appropriations process, the vouchers program was specially called out as a key thing for Office of Technology Commercialization…”The Committee directs the Department to continue to administer the Technology Commercialization Fund Voucher Program without interruption,” portending a good sign for the future of the program.

What Did We Learn Along the Way

Importance to Startups and Companies

We’ve heard resoundingly that not just the monetary value, but the ease of which vouchers are now facilitated is incredibly valuable to companies. Many have formed long-term relationships with the labs and facilities where they did testing. Even original companies through the Small Business Voucher program are working years later with the labs.

Facilitating these relationships are critical to the maturation of the companies and for labs to learn to work with a broader set of stakeholders.

And the technical validation received by companies has helped these groups raise follow-on funding, led to acquisitions, and helped companies achieve critical milestones.

Vouchers are a high-impact, often low-dollar way for DOE to achieve its mission of supporting the maturation and commercialization of energy technologies. And they should be sustained at DOE and other agencies.

The Agreement Part is Tough

Finding the right contractual agreement between the company, facility, and sometimes the intermediary can be tricky. The most common mechanism that is used between labs and industry are Cooperative Research and Development Agreement (CRADA). The issue we first faced was that CRADAs can be lengthy to execute and require a number of (unneeded) signatures. So in partnership with the lawyers at DOE, we created a Technical Support CRADA - or TS-CRADA. That helped to streamline. For some labs, we also sought to execute a Master Scope of Work (MSW), which could likewise help accelerate the timeline. That still is not ubiquitously executed across the labs, but is with some. But getting in a MSW could further streamline signatures.

With the RFTS, we also worked through streamlined agreements with facilities. Setting up the right expectation of what the agreement terms would be helped to speed the signatures needed to get to an agreement to start the work.

But this is one area there still is a lot of work needed. As organizations contemplate vouchers, understanding the best ways to continue to drive down the timeline from project selected to work starting is critical.

Matchmaking is….Hard

While we’ve iterated with many different ways to matching up facilities with companies, this is where you make or break a voucher program. Understanding the bandwidth of expertise and facilities on the provider side, and truly understanding what a company needs (versus what they may have originally stated) is a delicate dance to ensure that recipients are matched with the right service providers.

This could include relying on external merit panels, having councils that are delegated to ensure proper representation of the facilities, or some other way to have good input on matching companies with facilities.

Communication is Key

Communication includes a few variables.

• Have a broad dissemination plan for both the facilities and the vouchers.

• Ensure that you are communicating what is being offered in a voucher - vouchers are in kind, so applicants need to understand that this will not be a check cut to them, but a service provided for them with a described monetary value.

• Strong communication with the facilities is important. Planning and bandwidth understanding is imperative. Also make sure to communicate to the facilities and the principal investigators what the outcomes you’re trying to achieve with the program.

• Work to ensure there is good communication between the facility and the selected performer. Lack of understanding of the scope and value of the voucher can lead to a disintegration of the relationship.

• Communicate the data out of the vouchers! Under applicable rules for data release, get the data out there to show the value of these vouchers.

Predictability and Repeated Rounds

These voucher programs have ended out being quite popular. But running only one or a few rounds can be discouraging all around. Plan to have multiple rounds and through data convince leadership that it’s important to keep running these programs. There’s no lack of companies looking for access to facilities; these voucher programs should be a part of nearly any program design,

These are just a few of the lessons learned and findings.

Where Could Vouchers Grow

And what are some of the ways that vouchers could be improved?

Creating New Agreement Mechanisms to Speed Execution

We’ve really only started to push the boundaries of making these agreements simpler from an execution mechanism. There are ways to explore new ways to streamline.

Take the National Labs as an example. As mentioned earlier, CRADAs are slow - even the TS-CRADAs. But there’s a new variable - lab foundations and the Foundation for Energy Security for Innovation.

It’s never been explored about the efficiencies that could be gained by having a foundation serve as the intermediate for vouchers. There’s a potential that this could cut out some of the unnecessary signatures that occur between the site offices and the entrepreneurs.

Lab foundations also have the ability to have more discretion over what type of funding agreement is used in the context of vouchers; this malleability could help speed up execution if a foundation can bypass some of the unnecessary paperwork (much of the paper is based on joint discovery agreements, which isn’t pertinent to vouchers.)

But this isn’t the only potential mechanism. Organizations like DOE or others could play more with prizes in the form of vouchers, better business to business contract agreements, or more philanthropic funding that could be arranged in simpler agreements to support access to these facilities.

Annual Dedicated Funding for Vouchers

While the Senate language is a good start for support for DOE vouchers, having sustained appropriations language could benefit DOE and entrepreneurs. Predictability is a common theme when it comes to challenges for funding, and the same goes for vouchers. If companies and facilities have adequate time to plan for voucher relationships, then faster and more high quality projects could be achieved.

TEAMER is a good example of this in that it’s been running for 7+ years and has high predictability and consistent dedicated funding language in both the House and Senate appropriations.

Integrating Vouchers into Solicitations

Sometimes vouchers have been treated as their own program with a time cycle separate to other office solicitations. But there’s an opportunity to integrate vouchers better into solicitation processes.

Rather than vouchers being provided separate and distinct to awards like cooperative agreements and grants, offices could set aside funding specifically for vouchers during an award period. A use or lose set aside pot of funding identified by an office to specifically provide surge support/technical validation/testing during the period of performance.

WPTO integrated vouchers into the award package available to companies if they successfully won an award and that ended out being critical for many companies, including doing technical derisking and validation prior to in-water deployment for marine energy systems. WPTO set aside a certain amount of funding that would go to the National Labs as part of the overall budget for a funding program. This isn’t used ubiquitously and requires trade off and planning for where to dedicate resources. But having an integration of vouchers not just separate to awards could help improve project success.

Getting Philanthropies in the Game

Vouchers thus far - at least in the energy context - have been almost entirely funded by the federal government. But these are relatively low-cost, high-impact investments. In addition to augmenting the funds towards vouchers in technology areas in alignment with DOE, there are lots of areas that are currently either underfunded or unfunded by DOE or other energy agencies. Philanthropy could help bridge the gap between great entrepreneurs or communities that need validation and the technical and physical resources available.

As an example, there is very little funding available for water efficiency technologies or water purification technologies, but lots of resources and facilities exist. A philanthropy that cared about water access, purification, or efficiency could set up a simple voucher program available to companies and communities who need access to information or technical validation - just like the government does - and help achieve their programmatic objectives by serving as a critical bridge.

These are just a few ideas. If you have others to improve voucher programs or are interested in learning more, let us know!

In Part 1 of this series, we told the story of how prize authority at DOE went from dormant law to functioning tool. Part 2 walked through the full-stack design of the Solar Prize and how its success led to over 100 prize competitions across the agency. Part 3 examined when prizes are the wrong tool. Part 4 addressed what’s missing in the broader ecosystem and why the field needs a community of practice.

This post is for people ready to act. It’s the checklist and decision framework for designing and launching a prize competition. If you want the story or the theory, go back to the earlier posts. This is the reference guide.

Just to note: unlike our previous grants deep dives, this is more for program designers specifically, and not necessarily for applicants/competitors. Though understanding the motivations and structural decisions behind prizes can be helpful for people considering applying to prizes in the future.

Step 1: Is a Prize the Right Tool?

Before designing anything, work through the first layer of the full-stack program design framework. What specific problem are you trying to solve? Who needs to apply and what will motivate them? What is your budget? What does success look like? What are the measurable outcomes that will tell you whether it worked?

Those answers should drive the mechanism choice, not the other way around. Part 3 of this series covered the conceptual boundaries in detail, but the short version: prizes work well when the path to success is uncertain, when results can be objectively assessed, when competition will drive innovation, and when you want to attract solvers who wouldn’t normally engage with your organization’s funding. Prizes are the wrong tool when you need close oversight, when the path is well-known and you just need execution, when you want to build capacity in equal amounts across many participants rather than select winners, or when success is subjective.

If you’ve worked through that analysis and concluded a prize is the right approach, the next question is whether your organization has done this before.

Step 2: Is This Your Organization’s First Prize?

If your organization has run prizes before and has existing approval and announcement processes, skip this section. If this is your first one, work through it before you start designing.

The first barrier you’ll hit is simple inertia. “That’s not how we do things here.” Your organization has established ways of deploying capital, whether grants, contracts, or something else. Those processes exist because people spent years building them, and staff have built careers learning to navigate them. Proposing a different mechanism means asking people to learn something new, which feels like unnecessary effort and risk when the existing approach is perceived as working well enough. The response to this isn’t to argue that prizes are better than grants (they aren’t, for many purposes). It’s to identify a specific problem that your current tools aren’t solving well and position the prize as a targeted experiment to address that gap.

The second barrier is a misconception about what prizes actually are. When most people hear “prize competition,” they picture a large, winner-takes-all technical challenge. Something like the Ansari XPRIZE or the L-Prize: a single audacious goal, a big pot of money, one winner, program over. That model exists and has its place, but it represents a narrow slice of what prizes can do. As we covered earlier in this series, prizes can be structured in many ways: multi-stage competitions with increasing award amounts, multiple winners at each stage, support services layered on top of cash, technical vouchers, network connections. Expanding leadership’s mental model of what a prize can look like is often necessary before you can have a productive conversation about whether to try one.

Once you’re past those threshold objections, the more specific concerns emerge. Leadership will worry about the “no strings attached” nature of prize funds. We covered this dynamic in Part 1: the fear that winners would take early-stage prize money and walk away. The concern comes from people whose mental model is grants, where oversight happens through invoices, quarterly reviews, and project management. For someone who has spent a career in that world, the absence of those controls feels like risk rather than feature.

The way through is to start small. Propose early-stage amounts low enough that if someone did walk away (which in our experience across 100+ DOE prizes, nobody has), it’s manageable and does not involve large, irreversible sunk costs. Then make sure in the next stage there is a prize that will incentivize competitors to continue work rather than walking away with their part of the purse. The Solar Prize started at $50,000. Win that, and you’re one of 20 teams competing for 10 prizes worth $100,000 each. Hard to abandon a 50% shot at doubling your money.

You also need to answer the authority questions before you’ve invested months in a design you can’t execute. For federal agencies: Does your office have prize authority? Is it delegated? At what dollar thresholds do approvals change? At DOE, prize authority defaulted to the Secretary for years, which meant anything over $1 million required Secretarial sign-off. That barrier alone kept most offices from trying. Foundations and private sector organizations face different questions (legal review, board approval) but still need to map them early. Ironing out decision authority early will be especially critical when you want a fast moving prize, e.g. being able to select winners on the same day as a pitch competition.

Beyond these prize-specific hurdles, the full-stack framework applies. The ecosystem scan and coalition building work from Layer 2 is just as important here, and just as tempting to skip under time pressure. Right-size your first prize to prove the model before expanding. A successful small prize opens doors. A failed ambitious one closes them.

Step 3: The Design Dimensions

Every prize requires decisions across a set of design dimensions. What follows is a common list you will need to work through. The decisions you make here are what create the incentive framework that drives the behavior you want or sets up the prize for failure from the outset. The majority of the impact of prizes is generated during this design phase not the selection phase.

1. Competition & Winner Structure: How does the competition work and who wins?

• First-to-achieve, single winner: Race to a threshold. First to hit the target wins, competition ends. The L-Prize worked this way.

• Best-by-deadline, single winner or ranked placement: Competition closes on a fixed date, top performer(s) win. Can award just first place or tiered prizes (1st/2nd/3rd).

• Best-by-deadline, threshold: Everyone who clears a defined bar wins the same amount. The Ready stage of the Solar Prize worked this way: 20 teams each received $50,000.

• Rolling/ongoing: Continuous intake with periodic judging windows. Less common.

2. Stage Structure: How many phases?

• Single-stage: One competition, one set of winners.

• Multi-stage: Sequential phases where advancement depends on winning prior stages. The Solar Prize used three: Ready, Set, Go.

3. Evaluation Method: How are results judged?

• Objective technical validation: Measurable, verifiable. Did the prototype hit 25% efficiency? Send it to a third-party lab, get data.

• Subjective expert assessment: Judges evaluate quality, potential, team strength, market fit.

• Hybrid: Technical threshold to qualify, then subjective assessment for final ranking.

These can combine. The Solar Prize used a threshold at Ready (clear the bar, you’re in) but the Go stage was closer to ranked placement (judges selecting top performers from finalists).

Note: If your evaluation method depends on validated performance data, make sure you actually have access to the facilities and expertise to gather it. A prize that requires third-party testing only works if competitors can get that testing done within your timeline and budget.

4. Award Type: What do winners receive?

• Cash only

• Cash, plus any combination of:

  • vouchers (technical services, lab access)

  • network/support services

  • Other things of value

• Non-cash only (recognition, vouchers, access)

5. Funding Amount: How much, and how is it distributed?

• Total prize pool: Overall budget allocated to prizes.

• Per-winner amount at each stage: The $50k → $100k → $500k escalation creates different incentives than a flat structure.

• Ratio between stages: How much does the jump incentivize continued participation? Win $50k with a 50% shot at $100k and walking away gets hard.

• Fixed vs. variable: Are amounts predetermined or can judges allocate a pool based on submission quality?

• Administrative costs/overhead: How much funding is needed to execute the program which can include staff time, payment to third parties for support, stipends to reviewers, etc.

One practical note: prizes work well for small awards where grant overhead is prohibitive. The administrative burden of a $50,000 grant is nearly the same as a $5 million grant, which is why program offices avoid small grants. Prizes don’t have that problem. The same structure that awards a $500,000 grand prize can award twenty $25,000 early-stage prizes without scaling administrative cost. At the other end, very large prizes (above $1-5 million) raise questions about whether no-strings-attached is appropriate or if a cooperative agreement with oversight makes more sense. This is a decision government funders may have to consider, but it may not be as much of an issue for foundations or private entities.

6. Eligibility: Who can compete?

• Open: Anyone.

• Restricted: US only, small business only, students only, etc.

• Progressive restriction: Open for phase 1, must meet additional criteria for phase 2 (incorporated, domestic manufacturing) to receive award and advance.

• Team composition requirements: Must include a university partner, community organization, etc.

7. IP Treatment: Who owns what?

• Competitors retain all IP: The default for federal prizes based on AMERICA COMPETES Prize authority.

• Government gets license: Winners grant government rights to use the solution.

• Open source/data sharing requirement: Winners must make results publicly available.

• Anonymized data or output sharing: Winners will have some data on performance or other aspects of their work shared with competitors to evaluate relative strengths at a task or of a product, but it will be done in a way that protects individual entities (e.g. Solar Forecasting Prize).

8. Application Burden: How much work to apply?

• Minimal: The Solar Prize Ready phase required a 5-page written plan, a 90-second video, and a summary slide. For $50,000, hard to argue it isn’t worth the effort.

• Moderate: More detailed plans, budgets, letters of support.

• Heavy: Approaching traditional FOA requirements, which can defeat a major benefit of doing a prize.

Every additional page loses potential applicants. The instinct to ask for “just a bit more information, just in case” works against you. If your evaluation method requires third-party data (product performance testing, independent validation), think through who bears that cost, who gathers the data, and whether the timeline is realistic. Asking competitors to get lab testing done in a 3-month window when lab queues could run that long is setting them up to fail.

9. Support During Competition: What help do competitors get?

• None: Pure competition, sink or swim.

• Passive network: Directory of resources competitors can access on their own.

• Active support: mentorship, office hours, recruiting assistance.

• Technical vouchers and access to facilities: Funding for lab access or specialized services usable during competition.

If you’re paying organizations to help competitors (Recognition Rewards, Power Connector contracts), you need to work through the details: what triggers payment, when it gets paid out, how you assess whether the support was actually useful, and how those costs fit within your total budget alongside prize purses and administration. And you need to understand the overall costs associated with facilities access, particularly for validation purposes.

10. Timeline: How long does everything take?

• Competition period length per stage: The Solar Prize initially used 3-month windows, then shifted to 4 months after learning that was too tight for real progress and voucher processing.

• Time between stages: Including voucher redemption windows if applicable.

• If recurring, frequency of cohorts: Annual, semi-annual, rolling.

Design timelines for competitors, not staff convenience. A competition window over the holidays that gives your team a break forces competitors to prepare when their teams and partners are unavailable. This doesn’t mean your staff should work at the worst possible times. It means thinking through when each step happens and aligning deadlines with periods when competitors, reviewers, and partners can all actually engage.

11. Venue/Demonstration Requirements: Where does judging happen?

• Fully remote: Submissions only.

• In-person pitch component: The Solar Prize Go stage included live pitches with same-day winner announcements.

• Physical demonstration: Testing at a specified facility.

• Hybrid: Written submissions plus in-person final round.

12. Program Cadence: One-off or recurring?

• One-off: Single competition, program ends.

• Recurring: Annual cohorts, ongoing program.

Recurring programs are significantly easier to sustain once launched. They force you to measure impact (you need to prove Round 1 worked to justify Round 2) and let you iterate on design based on experience. One-off programs struggle with both. If there’s any way to structure your prize as recurring, do it. Once the community expects another round, word spreads on its own and the outreach burden drops.

This is not an exhaustive list, and every prize will have its own constraints and requirements. But these are the dimensions most prize designers need to work through, and the choices you make here matter more than most people realize. For the federal government, a traditional FOA process filters for teams that know how to navigate government funding. A well-designed prize can reach an entirely different population: entrepreneurs who would never consider a federal grant, researchers who don’t have the overhead to manage cost share, small teams moving fast on ideas that don’t fit neatly into a three-year project plan. The Solar Prize hit 75% first-time applicants in Round 1 and stabilized around 50% in later rounds, more than double the typical rate for DOE programs. That wasn’t luck. It was a response to incentive design.

For the rest of the launch sequence, governance mapping, timeline architecture, partner agreements, and impact tracking, the full-stack program design framework applies. What follows are pitfalls specific to prizes that the general framework doesn’t cover.

Step 4: Prize-Specific Pitfalls

The governance trap. Decision authority and communication authority are separate approval chains. The Solar Prize faced a six-month delay between selecting winners and being allowed to announce them. We had authority to make selections. We assumed that meant we could communicate those decisions. It didn’t. The people reviewing the announcement package were senior officials who had never heard of the prize, didn’t understand prize authority, and surfaced all the old fears about no-strings-attached funding. Map both chains before you launch. Identify who approves the competition, who approves selections, and who approves communications and get them all up to speed on your prize as soon as practical then remind them throughout the process. They are often different people with different levels of awareness.

One-off when recurring is possible. We covered this in Step 3 under Program Cadence, but the decision is consequential enough to revisit. If you can position a new prize as the first round of a recurring program rather than a one-off competition, it opens doors that are hard to open otherwise. You can start thinking about funding sources for Round 2 a year in advance. You can frame impact measurement around what you’ll need to justify continuation. You can recruit applicants who missed Round 1 by pointing them to Round 2. You can iterate on design based on what you learned. Your prize can build a reputation and brand in the target community. One-off programs don’t get any of this. By the time outcomes materialize, staff may have moved on to other projects and nobody is around to document what worked. The recurring structure creates accountability and momentum that one-off programs struggle to generate.

Partner agreement timing. If your design includes vouchers, network support, or dedicated applicant support, those agreements need to be in place before competitors need them. Prize timelines are compressed compared to grants. A 3-month or 4-month competition window doesn’t leave room for legal agreements to work their way through procurement. The Solar Prize initially used 3-month windows and discovered that voucher agreements with National Labs couldn’t be processed fast enough for competitors to actually use them during the competition. Build in more lead time than you think you need.

Realistic technical requirements. While prizes can establish ambitious goals, setting targets that are practically unachievable discourages serious competitors and wastes everyone’s time. Before finalizing technical requirements, confirm that between available funding, expertise, and facilities, participants have a reasonable path to success.

Scope creep into capacity building. When prizes start working, they can become the default tool for everything. New staff who are trained on prizes and see their success may develop tunnel vision, reaching for a prize structure even when other mechanisms would be a better fit. The risk compounds over time. Eligibility expands. Stages get added where everyone advances. Structures emerge where most participants win. At that point the mechanism is breaking down. Prizes are competitions. They work by being selective. If your goal is to support a broad set of participants rather than select winners, a prize is probably the wrong tool. Other mechanisms exist for that purpose. The same full-stack thinking that led you to choose a prize for the right problem should prevent you from using it for the wrong one.

Step 5: Resources

One theme runs through everything we’ve learned about program design, and it’s especially true for prizes: collaborating with other practitioners always generates a better result. Prizes seem easy and intuitive to design. Define a goal, offer money, pick a winner. But there is nuance at every level, and unknown unknowns that can lower or eliminate impact. The governance trap we hit with the Solar Prize wasn’t in any handbook. We learned it by making the mistake. Someone who had run a prize before could have warned us. Seeking out other practitioners and leveraging existing resources isn’t optional. It’s a critical step.

For federal agencies, Challenge.gov is the official repository for prize competitions. All federal prizes should be posted there, and the General Services Administration runs a federal Prize Community of Practice through that platform.

NASA’s Center of Excellence for Collaborative Innovation (CoECI) provides hands-on support for challenge-based initiatives. It was established in 2011 to help federal agencies experiment with these methods before standing up their own capabilities. Most federal staff outside NASA don’t know CoECI is available to them. There is an informal distribution list of federal prize practitioners from many agencies that frequently ask each other for advice. Seek out an experienced federal prize staff member and they can likely help add you to this list.

For practitioners across sectors, the Global Prize Network launched in 2025 as a LinkedIn group connecting prize designers from government, philanthropy, and private sector organizations. It emerged from a gathering during Climate Week where organizations that fund and run prizes realized they were all solving similar problems independently.

If all else fails, shoot us an email via the website listed at the end of this post and we will do our best to help.

Conclusion

I wish there was more to share in the previous section. The resources that exist are useful but scattered, and the community of practice we described in Part 4 is still more aspiration than reality. We’re specifically interested in helping make more available.

This series started with a 13-year story about institutional resistance and ended with a checklist. The path from “that’s not how we do things here” to 100+ prize competitions at DOE required persistence, timing, and leaders willing to spend political capital on something new. We wrote it down because too much of this knowledge stays locked in the heads of practitioners who figured it out through trial and error, and disappears when they move on.

Prizes are a powerful tool when used well. They can reach applicants your organization has never seen, support hard tech that doesn’t fit traditional funding models, and generate outcomes that justify continued investment. But they’re not magic. The impact comes from design, not the mechanism itself.

If this guide helps you launch a prize, we’d like to hear how it goes. If you’ve run prizes and learned lessons we missed, we’d like to hear that too. The ecosystem gets stronger when practitioners share what they know.

Innovation Waypoints is brought to you by Waypoint Strategy Group.

In Part 1, we told the story of how prize authority at DOE went from dormant law to functioning tool. In Part 2, we walked through the full-stack design of the Solar Prize and how its success led to over 100 prize competitions across the agency. In Part 3, we examined when prizes are the wrong tool for the job. This post addresses what’s missing in the broader prize ecosystem and why the field needs a community of practice.

The Apprenticeship Problem

Throughout this series, we’ve returned to a recurring theme: the craft of designing and deploying capital effectively is learned through apprenticeship. It takes years to understand not just the mechanisms, but the governance, the ecosystem, the politics, the timing, the private sector partners, the potential applicant archetypes, and all the unwritten rules that determine whether a good idea becomes a real program. That apprenticeship model is slow, fragile, and dependent on the right people being in the right place at the right time.

This is true for all funding mechanisms, but it’s especially acute for prizes. Grants and cooperative agreements have decades of institutional infrastructure behind them. There are trainings, handbooks, acquisition workforce certifications, and career tracks built around financial assistance. Prizes have almost none of that. The knowledge that exists lives in the heads of practitioners who figured things out through trial and error, and it largely stays there.

We built the Solar Prize without a roadmap. Every office at DOE that followed had to learn similar lessons. The same is true outside the government. Foundations launching prize programs, corporate innovation teams experimenting with challenges, and private prize organizations all face the same basic questions and mostly answer them independently.

A Recognized Gap

The Energy Act of 2020 recognized this problem. Section 9004 directed the Secretary of Energy to issue Department-wide guidance on the design, development, and implementation of prize competitions and to collect and disseminate best practices on the design and administration of prize competitions. It was an acknowledgment that prizes had proliferated across DOE but the institutional knowledge remained siloed.

To date, for the most part, this unfunded mandate (another topic we will cover at a later date) has not been carried out. Prize programs continue to be designed and built through the unofficial apprenticeship model, with varying degrees of success depending on whether staff happen to connect with someone who’s done it before. Informal prize guides have circulated, and new rules documents are built off best practices from previously run and administered prizes. But there is no centralized body within DOE that a new staffer can go to that can help them design, launch, and shape a prize.

NASA offers one model for how this could work within an agency. In 2011, at the request of the White House Office of Science and Technology Policy, NASA established the Center of Excellence for Collaborative Innovation (CoECI) to mature the agency’s crowdsourcing strategy and help other federal agencies use challenges to solve mission-critical problems. CoECI guides teams through the full lifecycle of challenge-based initiatives, from problem definition through incentive design to post-submission evaluation. Through the NASA Tournament Lab and a research partnership with Harvard University, they’ve executed numerous challenges across multiple agencies while building an evidence base on what actually works. The model lets agencies experiment with these methods before standing up their own capabilities.

This is valuable infrastructure. However, CoECI has limited bandwidth and many practitioners aren’t aware it exists. Many staff in DOE, for example, don’t know CoECI is available as a resource for staff outside of NASA. And while CoECI can teach universal principles, each agency and each organization has its own unique challenges with building out prize capacity. The broader prize ecosystem, including private sector organizations running their own prizes, lacks similar connective tissue.

The Scale of the Problem

In 2025, a group of organizations that either directly shape or fund prizes in the private and public sector gathered at an event coordinated by XPRIZE and the Keeling Curve Prize during Climate Week in New York. The event was unique in that while many of the organizations had either informal or formal operating relationships, it appeared to be one of the first prize collaboration meetings with all of these groups in one room. The staff at XPRIZE and Keeling Curve had been discussing the need to bring together prize practitioners across the philanthropic landscape, but since there is no central, private-sector prize organization or network convener, this was done through volunteer time, gathering like-minded groups and individuals while everyone was at Climate Week.

The conversation surfaced real challenges and opportunities in how the prize ecosystem operates. Some of the common gaps included the lack of a standard approach for measuring prize efficacy, no shared frameworks for basic design questions, limited independent research on what makes prizes effective. Prize practitioners face the issue of reinventing the wheel on questions like what constitutes a prizable problem, how to structure stage gates, what prize amounts motivate without distorting behavior, and how to measure the impact (or lack thereof) of prize successes.

These shared challenges aren’t necessarily a bad thing. They show there’s an opportunity to create better alignment and connect groups more intentionally, whether through regular convenings, shared evaluation frameworks, or leveraging new technology to accelerate learning cycles. What struck everyone in that room was how much prize organizations want to work collaboratively. We all want to advance the utilization and impact of prizes. As a result of the meeting, the Global Prize Network launched as a LinkedIn group, with plans to expand from there.

But even the identification of these challenges is only a snapshot of the prize ecosystem. There were about 20 organizations in that room that had each independently figured out the value of prizes and wanted a place to collaborate. Meanwhile, there are thousands of foundations, government agencies, and corporate funders deploying capital toward innovation who have never seriously considered whether a prize might be the right tool. The coordination challenges among active practitioners matter, but the first-prize problem, meaning organizations that have not tried a prize, represents the larger loss in potential impact. Even if you look at the launch of a prize as an indicator, as far as I can tell, the most viable pathway for a new organization to try prizes is to hire someone who has previously run a prize and have them lead the first one, or for leadership to spontaneously decide prizes are worth considering and then hunt through their network to find someone to talk to. Neither is systematic. Both depend on luck and knowing the right people.

What Would Help

The pieces for a community of practice exist, but they’re scattered. The General Services Administration maintains Challenge.gov as a federal repository and runs a Prize Community of Practice. The National Renewable Energy Lab has been building a database of DOE prizes. There’s also a federal Prize Competition Guidebook. CoECI provides hands-on support for agencies willing to seek it out. Private sector organizations like XPRIZE have deep expertise and reach. What’s missing is something that connects all of these into a functional whole.

A real community of practice would need to address several gaps simultaneously. On the connection side, there’s no centralized way to map the continuum of prizes available to entrepreneurs or to connect prize alumni with each other. On the evaluation side, individual organizations have developed impact assessments, but they’re isolated. The last decade has produced a wealth of data on prizes, but nobody is synthesizing it comprehensively. A common impact framework that lets practitioners evaluate and compare different prize designs would help the entire field learn faster.

For organizations new to prizes, the need is even more basic: awareness that the option exists. Most foundations and corporate funders have never seriously considered prizes because the mechanism simply isn’t on their radar or they have a misconception of what they actually are and the value they can provide. Solving that requires someone waking up every day trying to spread the word: continued outreach, discussion of impact, presenting at major convenings, building the case for prizes as a tool. That person or organization doesn’t exist. The prize ecosystem is largely a closed system where practitioners talk to each other but rarely make the case to outsiders.

For the few organizations that do become curious about prizes, they need someone to talk to. A program officer who’s heard about prizes but has never run one needs a place to get briefed on how they work, whether the mechanism fits their problem, and what the first steps look like. CoECI can serve that role for federal agencies. For foundations and private sector funders, there’s no equivalent front door.

Why This Matters

Dating back centuries, prizes have helped spur new ideas, forge new pathways, transform industries, and inspire thousands of individuals to develop new technologies. In the 21st century, organizations like XPRIZE helped launch the commercial space industry. Federal agencies like DARPA and DOE have used prizes to advance autonomous vehicles, achieve breakthroughs in LED lighting and desalination, and launch new ventures. The mechanism works.

But to maximize the impact of both public and private sector prizes, new scaffolding to support prize implementation is needed. There’s an opportunity to envision prizes not as isolated competitions but a more connected ecosystem that shepherds ideas and people along a continuum. To create a more connective environment for prize applicants and alumni while fostering greater collaboration between prize designers and funders.

None of that happens without connective tissue. Analysis, advocacy, connections, shared best practices, and common evaluation frameworks need a home. The knowledge exists. We just need to systematize it.

What Comes Next

This series has covered the history, the design, and the boundaries of prize competitions. We’ve now addressed what’s missing in the broader ecosystem. In Part 5, we’ll provide the practical reference guide we promised at the beginning: a summary of prize mechanics, design considerations, and implementation steps for practitioners who want to move from theory to action.

Innovation Waypoints is brought to you by Waypoint Strategy Group.

In Part 1, we told the story of how prize authority at DOE went from dormant law to functioning tool. In Part 2, we walked through the full-stack design of the Solar Prize and how its success led to over 100 prize competitions across the agency. Both posts made the case for prizes as a powerful mechanism for reaching new innovators and supporting hard tech development.

This post is about when not to use them. The examples here come from DOE, but the lessons apply to anyone running prizes: foundations, corporate innovation teams, and private prize organizations face the same design choices.

The “Everything Should Be a Prize” Problem

When prizes started growing in popularity at DOE, an interesting thing happened. They were such a clear improvement over traditional processes in reaching new applicants that some started to wonder: why not just do everything with prizes? This was especially acute with new staff whose first program was a prize. When those initiated into funding design through the simplicity and logic of prizes were later confronted with the normal grant or contract process, they sometimes rejected it. And for good reason from their point of view. It feels unreasonable to subject applicants to so much more paperwork and process, negotiate every activity in advance, scrutinize invoices, and approve every dollar spent on the effort over years.

The issue is that prizes are not a suitable tool for all programs. Traditional funding mechanisms exist for a reason and incredible things have been done with them.

Mechanisms as Tools

The way we like to think about funding mechanisms is as tools. In construction, a practitioner with an appropriate set of tools can build anything. With a saw, hammer, measuring tape, level, drill, and square, a house of any shape can be built, or a boat, or a boxcar. But if all you have is a hammer, your opportunity set is diminished because you end up using the wrong tool for the wrong job. Proper project planning requires knowing the strengths and weaknesses of each tool and when to use it.

So what are prizes good for? Prizes are competitions. They are a mostly passive tool to incentivize a community to activate around an issue and reward only those that generate the best results. Those that compete and don’t do well are out of luck. It’s intentionally selective and reductive. The most straightforward use is a well-defined, quantifiable technical challenge: the capability to make a widget that is 25% efficient doesn’t exist. If you make one and we can verify it, you get X dollars. Done.

You can add phases to support teams working on that challenge and spur more activity. This is still a good use of the tool. You can push one step further, as we did in the Solar Prize, and state that you want to create the best and most impactful two new solar startups once a year and give them $500,000 if they reach a legally binding customer commitment. This still works. Taking 100 to 200 competitors down to two with support services that help everyone along the way is still fundamentally a competition with clear winners.

Where It Goes Wrong

Where things go sideways is when funding program designers, with the best of intentions, seek to use prizes to make more and more awards for capacity building activities. Example: a disadvantaged community struggles to apply to traditional funding. The thinking goes, let’s do a prize to award them funds to do an activity, and then we’ll do multiple phases and everyone or most can win at each stage. These “everyone wins” prizes stretch the mechanism in ways that raise questions. At DOE, Congress noticed the agency was doing more activities like this and raised concerns. With sufficient concerns, the authority for the government to use prizes could be changed or eliminated.

The response to this criticism is that these communities can’t apply to traditional grants because it’s too difficult. Or that there just is not sufficient funding to support capacity building specifically - we expect new groups and applicants into our portfolios but have limited funds in building capacity to apply into funding. Prizes became a default tool to meet that gap, And that wasn’t without merit - until recently, there wasn’t a good answer to that argument. Now DOE has a tool that is a better fit for capacity building: the Partnership Intermediary Agreement. We hope to discuss the PIA mechanism in another post, but suffice it to say it offers a low barrier to entry, increases capacity building, and provides streamlined process without the competitive nature of prizes.

The Size Question

Prize amounts have to be motivating without being distorting. Too little and nobody bothers. Too much and perverse incentives emerge, or teams focused on trying to game the system for the money rather than the mission apply. The key is calibrating to the work required and the target audience.

One practical advantage of prizes is they work well for small awards. Traditional grant mechanisms struggle to make awards under $250,000 because the administrative overhead isn’t worth the return. The time spent on proposal review, negotiation, project management, invoicing, and closeout is roughly the same whether the award is $50,000 or $5 million. Program offices avoid small grants because they consume staff time that could support larger, higher-impact projects. Prizes have much lower overhead once the structure is set. The same competition that awards one $500,000 grand prize can also award twenty $25,000 early-stage prizes. The administrative burden doesn’t scale with the number of awards or the dollar amounts. This makes prizes particularly effective for early-stage support, where small amounts of flexible capital can make a material difference but would never justify a traditional grant process.

At the other end of the spectrum, very large prizes start to raise questions about whether a prize is the right tool. At a certain size, it becomes difficult to justify issuing a large, no-strings-attached award for a result when a cooperative agreement or grant would provide oversight and risk management over several years for that level of funding. The exact threshold isn’t clear, but somewhere between $1 million and $5 million, the calculus starts to shift. The administrative savings that make prizes attractive for small awards matter less when the stakes are high enough that active project management and regular check-ins would be prudent anyway.

That said, context matters. If a $5 million cash prize can catalyze $50 million in private investment, it could make sense. Which brings us back to a point we’ve made several times in this series: start with what you’re trying to accomplish, not with the mechanism you’re comfortable using. This is the Thesis & Success Map layer from our full-stack framework. What specific problem are you solving? What does success look like? What are the measurable outcomes that will tell you whether it worked? Those answers should drive the mechanism choice, not the other way around.

If you’ve never run a prize, finding the right program concept for your first one is especially important. If you’ve run many prizes, consider whether you’re developing tunnel vision. The tendency to default to what worked last time is natural but can lead to using the wrong tool for the job. This is a dynamic space. Paying attention to what others are learning and collaborating with practitioners facing similar challenges is the best way to avoid getting stuck.

With that framing in mind, here’s how we think about the boundaries:

When to Use Prizes

The sweet spot for prizes is well-defined technical or commercialization challenges. Specifically, situations where:

• The path to success is uncertain and you want multiple approaches tried

• You can objectively assess and judge results

• Competition will drive innovation and urgency

• You want to attract new solvers who wouldn’t normally engage with your organization’s funding

• You’re facing an interdisciplinary issue that will require groups or scientific fields to tackle issues together

• Winners can be clearly differentiated from non-winners

• The work can be done on the competitor’s timeline and with their own resources (until they win)

When Not to Use Prizes

Conversely, prizes are the wrong tool in several situations:

• Outcomes will improve from close funder oversight and active project management

• The path to success is well-known and you just need someone to execute it

• You need sustained, multi-year effort with regular check-ins and course corrections

• You want to support capacity building or provide resources to help all participants

• The goal is to fund many teams doing similar work rather than selecting the best

• Success is subjective or difficult to judge objectively

• A “winner takes all” dynamic would harm the ecosystem you’re trying to build

Conclusion

Understanding these boundaries matters beyond just program design. The lessons in this post come from an environment where prizes became popular enough that staff started using them for everything. That’s a good problem to have. It means the tool is well understood and accessible.

Outside government, the more common problem is the opposite. Prizes remain underutilized by foundations and private sector organizations that default to grants or contracts without considering whether a competition might work better. The same framework applies in both directions. If you’re reaching for prizes reflexively, ask whether another mechanism fits the problem better. If you’ve never seriously considered a prize, ask whether your current approach is actually optimal or just familiar.

Knowing the full toolkit, and being honest about which tool fits which job, is what separates effective program design from habit.

In Part 4, we’ll discuss what the broader prize ecosystem is missing and why the field needs a community of practice.

Innovation Waypoints is brought to you by Waypoint Strategy Group.

Note: We thought this was going to be a 2 part series, however, there is a lot to cover on the topic of prizes so…there will be more parts! This section covers the specifics of how the solar prize was designed. What is missing in the prize ecosystem, when prizes are not the right choice, and the prize guide will all be coming.

The “8-Month Slog” That Built the Fundamental Building Blocks

In Part 1, we told the winding, multi-year story of how prize authority at DOE went from a dormant law to a real, usable tool. We ended with the inflection point: a new leader, a pressing political need, and a 2-year-old white paper being suddenly pulled off the shelf and greenlit in a week.

That approval was a victory, but it was just the beginning. This is where the real work started.

The story of how we built the Solar Prize between 2015-2018 is an early attempt at what we now call full-stack funding program design. But to be clear, we did not have this framework when we built the Solar Prize. What we had was approval to do a multi-stage prize and partner with mission-aligned organizations. The mechanics of how to do either were unknown. What resulted was an 8-month slog where we fiercely debated opinions, because there were no facts, firm legal structures, or precedents to follow. Everything could be argued, and was.

We’re going to explain what we built using the 8 Full-Stack layers. Not because that’s how it happened, but because it’s the clearest way to explain the decisions we struggled through. It’s the framework we wish we had.

The Full-Stack Breakdown of the American-Made Model

Layer 1: Thesis & Success Map

This whole effort started with the “Leaky Applicant Pipeline” issue. Our main program for supporting solar entrepreneurs was the Sunshot Incubator. It was a solid and successful program, but each year the number of new, first-time applicants was going down. The traditional Funding Opportunity Announcement (FOA) (AKA Notice of Funding Opportunity (NOFO)) model was reaching fewer new companies and funding the same shrinking pool of usual suspects who knew how to navigate grants.gov, cost-share requirements, and 100-page FOA instructions.

We had two other problems. We wanted to use government funds to support the hard things that were not attracting traditional investment. This meant the program was focused on supporting domestic manufacturing, which meant the program was seeking hardware technology development and manufacturing. This is not what one would traditionally think to use a prize for, the way you might for crowdsourcing a solution to explore data or finding a novel solution to a predefined narrow problem.

The other main challenge is that entrepreneurship is very difficult in a normal situation, and solar entrepreneurship was extremely hard at that time due to a solar boom-and-bust cycle at another low resulting in disinterest from private investment. We had heard many interesting anecdotal stories that important solar companies were just a few days from collapsing for any number of different reasons, and random collisions and serendipity resulted in them having the right connection to the right person that was needed to overcome many of the challenges. Without those connections, the company fails. This appears to be a near universal story of entrepreneurship. We wondered, how often is that critical connection not made and an otherwise viable business goes down? Is there a way we could systematize this organic, essential-to-commercialization process?

These challenges shaped our guiding thesis. The traditional FOA process wasn’t just a filter; it was a wall. High barriers to entry were blocking hardware-focused entrepreneurs. Even if they got funding, lack of systematic support meant many were failing for non-technical reasons. If we built a new pathway with a low barrier to entry, we could attract these missing innovators. If we simultaneously built a support network around the prize, we could systematize the serendipity and increase their chances of success.

This thesis gave us a clear definition of success with four requirements. First, attract new applicants. Second, support hard tech and domestic manufacturing. Third, build a functional pipeline where prize winners would eventually flow into larger DOE programs and set them up for private sector support. Fourth, prove the value of a support network as a measurable part of success. These were our strategic anchors that let us defend the unconventional design choices we were about to make.

Layer 2: Ecosystem Scan & Coalition Building

Our ecosystem scan was very bottom-up. This is the part of the process most government programs are good at: trying to understand their applicant pool. We spoke with hundreds of startups, venture capitalists, incubators, and our own awardees. We didn’t ask if they wanted prize money. We asked why they were struggling, what they actually needed to succeed, and what amount of funding would make a material difference.

The answers to the funding questions were predictable. They needed more money, faster. But the answers to “why are you failing” were revealing. We heard the same story repeatedly: near-death experiences saved not by a grant, but by a random collision with the right person at the right time. This scan validated our thesis. The serendipity problem wasn’t just our theory, it was the lived reality of entrepreneurship.

That being said, when we asked entrepreneurs what we should build into the program, nobody asked us to create a support network. They advocated for all program funds to go directly to them as cash prizes. They believed they would put every dollar to better use than anyone else. This is where we had to make a critical deduction that went against what our customers were telling us. Capital alone is not enough. The entrepreneurs were telling us this in their own stories, even as they argued for the opposite. They’d survived because of connections, not just capital.

This insight drove our entire design approach. We couldn’t just offer cash prizes. We had to build the connective tissue around the prizes that would systematize the serendipity. But this created our central design challenge: how do we build a support ecosystem robust enough to solve the challenges entrepreneurs face, without pulling so many resources away from prize purses that entrepreneurs no longer see the program as valuable?

We should acknowledge a failure on the coalition building with other funders part of this layer. At the time we were so overwhelmed with the scope of the task and being pressured to deliver something by leadership that we did not have the mental capacity or time to consider whether anyone else was doing similar work that we could partner with. This is usually the case for anyone trying to develop and launch a program in any organization, and it represents the lowest hanging fruit for amplifying impact. Were we to repeat the program design process, we would include a thorough ecosystem scan of any mission-aligned funders in other federal and state governments and also private funders. We would also have tried to engage with and learn from other expert prize practitioners, but at the time there wasn’t an obvious way to do that. This remains difficult and is something that requires a better solution. True success isn’t running a single useful program. It’s helping align all those who are seeking to support an area with funding and helping everyone do more and do better.

Layer 3: Mechanism & Project Design

This layer was the heart of the 8-month slog. We had our thesis. Now we had to design the actual prize structure that would work. The problem was we insisted on a multi-stage prize and there was the leadership fear we covered in Part 1: “They’ll just take the money and walk away.” With grants and cooperative agreements, the government maintains control through active project management. With prizes, you have none of that. Once awarded, prize funds become non-federal with no strings attached.

Our ecosystem scan told us that entrepreneurs wouldn’t walk away with the funds they won in the first stage of a multi-stage prize. When entrepreneurs have a clear target and a credible path to more funding, they don’t abandon the work. They work harder. But this argument wasn’t enough for leadership. We had to work through many iterations of prize structures to find a path forward that would address their concerns.

The answer was a multiple escalating prize structure, but with a low enough dollar amount in the first stage that if a competitor did just walk away (which leadership was certain would happen), it was a manageable risk. We made each stage exclusively available to winners of the previous stage. If you didn’t win the Ready stage, you couldn’t compete in the Set stage. If you didn’t win the Set stage, you couldn’t compete in the Go stage. This changed the math. You win Ready and get $50,000. Now you’re one of 20 teams competing for 10 Set prizes worth $100,000 each. A 50% chance of winning is hard to resist. You’re not competing against the world anymore. You’re competing against 19 other teams who also won. Similarly, once you win Set, you have a 20% chance of winning $500,000. No one walks away from those odds.

This is how we solved the no strings attached problem. We created motivational strings. The incentive to win the next, larger prize pulled everyone forward. The structure gave leadership multiple decision points. By the time we were awarding two $500,000 Go stage grand prizes from the original 20 Ready stage companies that started with a problem to solve, teams had already demonstrated a proof-of-concept, built a prototype, and secured commercial partners. We weren’t betting on an idea. We were awarding demonstrated success.

Layer 4: Support-Service Layer

The prize structure solved the walking away with cash from the first stage problem. But we still had to solve the serendipity problem and the hard tech gap. This is where the approval to partner with mission-aligned organizations came in. We had to figure out what that meant in practice.

From our research, we knew we could not just tell entrepreneurs to go network. That is a large part of what they already do. The issue is it’s hard to find the right people to talk to. Conversely, groups that want to support the best entrepreneurs also struggle to find them and identify which ones are worth talking to. We wanted to produce something that solved both sides of this issue. We believed that mission-aligned organizations - like business incubators, university entrepreneurship programs, makerspaces, and others - would already be interested in helping and some would do so organically. What was missing was a clear way for competitors to see which organizations were interested.

To solve this we created the American-Made Network as a repository of entities that could help the prize competitors. Members of the Network were called ”Connectors” to recognize their role as connective tissues in the innovation space. To ensure these Connectors were high quality, we created a sign up process and credibility check to confirm the organizations were what the Network was intended for. After that, their information would be visible to prize competitors. But we weren’t sure anyone would actually sign up and help. This is where we used the prize tool on the network itself. We set up Recognition Rewards. Competitors were encouraged to review the Network for organizations that could help them and if a Connector was identified by a winning competitor as materially helping them succeed, that Connector also got a cash prize (on the order of low single digit thousands of dollars). This created a direct financial incentive for organizations to sign up for the Network, be responsive to competitor requests, and do their best work.

As a hedge, we also set aside funds to directly contract with a few Power Connectors to accomplish specific tasks like recruiting and applicant support. By having both passive Connector incentives and active Power Connector contracts, we built a support system that didn’t traditionally exist for government programs.

While the Network connected competitors to entities that could help with business development and maturation, there was also a need to connect competitors to world-class research and testing capabilities. The final piece in the prize puzzle to serve this need was a technical services voucher. Prize authority allows the government to award cash or other things of value. When Solar Prize competitors won the Set phase, they received their cash award and a $50,000 voucher that could be used at any National Lab to leverage their technical experts and facilities. Vouchers were use-or-lose with a set time frame. This was the answer to the hard tech gap. The voucher forced competitors to engage with resources they wouldn’t have naturally pursued as National Labs were seen as difficult to contract and work with. DOE support at this stage bridged that gap.

The Ready, Set, Go model provided the cash and motivation. The Network provided the connections. The vouchers provided access to resources competitors couldn’t afford or access. This was why the American-Made model worked when other prize programs had stalled. We weren’t just running a competition; we were building an ecosystem.

Layer 5: Governance & Decision Ops

This is the layer where we truly failed. It wasn’t due to lack of experience. We knew how approvals worked. Our failure was in not anticipating what happens when a well-worn approval process built for grants suddenly encounters something it’s never seen before.

In a traditional FOA, the approval architecture is straightforward. Solicitations are typically approved by the Office Director with advice or approval from the Assistant Secretary or another Senate-confirmed position. Once a solicitation closes, selections are made by a Selection Official based on recommendations from a Federal Consensus Board and a Merit Review panel. That process is subject to change, but can typically be navigated.

Prizes operate differently. The COMPETES Act included vague language requiring the Secretary’s approval for any prize over $1 million and Congressional notification for prizes over $50 million. We’d spent years working to get that $1 million threshold delegated down from the Secretary to the Assistant Secretary level, which would let us move faster with multiple prizes (getting Secretarial approval on anything is a behemoth of a process). This delegation effort had been in motion since before the Solar Prize design work began.

The selection process was also different. Prizes use a single Judge who receives input from expert reviewers, rather than the layered Federal Consensus Board structure. There was confusion about who could be designated as a Judge. Due to the ongoing confusion around prizes, legal advisors had settled on designating the Secretary of Energy as the default Judge unless that authority was explicitly delegated. We’d worked hard to get a single-use delegation from the Secretary down to our Solar Office Director for this specific competition.

We ran the competition. Expert reviewers scored the submissions. The Director made the selections based on those scores. We thought we were done. We were not.

The problem emerged when we tried to announce the winners. We’d assumed that having authority to make selections meant we had authority to communicate those selections. That assumption was wrong. At the same moment, there appeared to be a re-reading of the statutory language in the American COMPETES Act. Since the Secretary was required to approve any prizes over $1 million, a legal question arose about what triggered that approval. Was the $1 million threshold for approving prize competitions before they launched, or was it for awarding prize funds totaling over $1 million? The distinction mattered because our multi-stage prize had a total pool above $1 million even though the first stage awards were only $50,000 each.

Before we could sort through this question, the announcement package hit a completely different approval queue with a much higher threshold. Everything froze. So much time had passed between the initial approval to launch the prize and the first round of selections (over a year) that leadership personnel had moved around. The people reviewing the announcement package were senior officials who’d never heard of the Solar Prize and didn’t understand prize authority. All the old fears came back. “They’ll just take the money and walk away.” “There are no strings attached.” “This is too risky.”

Even though the selections had been made, we could not notify the winners. The delay stretched to six months. Twenty teams who’d beaten out hundreds of competitors had done the work on their own dime and then waited. And heard nothing. For a startup, six months is an eternity.

We eventually got the approvals. It took dozens of meetings, countless emails, and leadership staff who spent political capital to push it through. Once we got through it, we’d set the precedent. The next prize program had a path to follow. But it was painful the first time.

The lesson turned out to be simple but critical: making a decision and communicating a decision are two entirely separate processes with different approvers. This seems obvious in retrospect, but it wasn’t obvious when we were in the middle of it.

The Governance & Decision Ops layer feels like a check-the-box activity when planning. But in reality, it’s the source of the most frequent and painful delays. To actually succeed at this layer, you have to map out everything the approval process involves. Everyone who needs to sign off, their availability and awareness, the external political environment, the budget process, the holidays and summer vacations, and so on. This is not fun work. But it’s the difference between a program that hits timelines and a program that leaves winners waiting in the dark for six months.

Layer 6: Building Awareness

Our failure to map out the full governance process had serious ramifications but that is not the end of the story. There were still other outstanding challenges around awareness that we needed to consider to make this prize work. We’d designed the program. We finally had approval to launch. Now we faced a different problem: how do you tell people about it? In government, building awareness is one of the most constrained parts of running any program. Government must be fair, which means everyone must have equal access to opportunities. In practice, this means targeted outreach is essentially forbidden.

Funding opportunities must be posted on grants.gov or challenge.gov. Program offices can email people who’ve signed up for their newsletter. If deemed important enough, an Office can issue a quasi press release (you can’t call it a press release) or the Public Affairs Office for the Department can elect to distribute via a formal press release - but it’s typically a high bar reserved for pushing out content to the media. And then back in the office, staff might post on LinkedIn. That’s about it. What you can’t do is what most would instinctively want to do: identify high-potential candidates and personally invite them to apply. If you email someone directly about a funding opportunity and they apply and aren’t selected, they could file a complaint. Or if they did win, and someone found out that you emailed that group directly, they could also file a complaint about unfair advantages. It sounds paranoid, but these are the scenarios federal employees are warned about in ethics training.

This creates a closed ecosystem. A specialized class of people emerges who are experts at monitoring funding announcements and navigating the application process. This is one reason why the number of first-time applicants to most DOE programs hovers around 25%.

Our strategy for the Solar Prize was to use the full-stack design to bypass this entire problem. We relied on our Support-Service Layer. Our Power Connectors and Network Connectors weren’t federal employees. They didn’t have these restrictions. They could send direct emails to entrepreneurs in their network. They could host workshops and invite specific companies. They could do literature reviews and personally reach out to researchers. All the things we couldn’t do, they could. And because we’d built Recognition Rewards and Power Connector contracts into the program, they were incentivized to do it.

Awareness is the first part of the pipeline, but the less obvious part is converting interested parties to applicants. It is unfortunately impossible to know how many high-quality potential applicants take a look at an opportunity and decide it’s not worth the trouble. Maybe they think it’s too difficult to apply, or maybe they’re concerned about their IP, or something else. From conversations with applicants, we think this number may be high. Given that we sought to achieve the minimum possible applicant material needed to make an informed selection decision, we focused on reducing barriers. This is really hard to do in government because the instinct is to always ask for a bit more information, just in case. The end result for the prize was a 2500 word maximum application (5 pages), a 90-second video, and a summary slide. That’s it. When paired with a $50,000 prize, it was hard for anyone to argue this isn’t worth it or this is too risky to my company. (For context, FOA/NOFOs ask for 20+ page full applications, detailed budgets, and a variety of supporting documents)

This is how we solved the Leaky Pipeline. In Round 1 of the Solar Prize, approximately 75% of applicants were first-time applicants to any DOE program. In subsequent rounds, the number stabilized around 50% new applicants per round. Still more than double the typical rate. We didn’t find those applicants. We built an incentivized system that found them for us. The Network wasn’t just about supporting winners. It was our entire awareness and recruitment engine.

Layer 7: Budget & Timeline Architecture

We had the structure, the support network, the funding amounts. Now we had to figure out when everything would happen and how the money would flow. When we were designing the Solar Prize timeline, we were juggling competing constraints. Staff availability, external reviewer availability, competitor needs, the political cycle, program infrastructure build time, and legal and contracting timelines.

The overwhelming temptation in government is to design programs for the convenience of yourself and your team. Launching any program is hard enough. The prevailing attitude can easily become, “Take it or leave it, we’re doing the best we can.” A recent example of this:

This timeline is ideal for staff. It goes over a holiday week so staff won’t have to handle anything while off. It is 31 days (the minimum allowable time), giving DOE staff more time for processing. But there’s a counter party to this arrangement. Applicants are now forced to prepare submissions during holidays when their own team and potential partners are out of office and unavailable. This approach is unlikely to help a program trying to reach new audiences. To be clear, we are guilty of this ourselves. Sometimes it feels like the only option, but it definitely is not the best option.

The ideal approach is to start with a different question: in a perfect world with no constraints, what is the right schedule for the entrepreneur? We initially determined that a 3-month competition period for each stage made sense. Fast enough to maintain momentum but long enough to get real work done. However, reality intervened. In practice, a 3-month window was too tight for notable progress as well as too fast for the voucher process. The legal agreements and setup time meant vouchers couldn’t start in time to actually help competitors.

We adjusted the next time we ran the program. We shifted to a 4-month timeline per competition stage. Three months had been too tight. Competitors needed more time to do the actual work required for each stage, and the voucher legal agreements needed time to process so Labs could help in a timely manner. An extra month gave everyone the breathing room they needed.

The broader principle here is worth restating: the overwhelming temptation in government is to design programs for staff convenience. An alternative that can help drive new applicants is to start with what’s best for the people you’re trying to serve. Then figure out how to make it work within your constraints. Not the other way around.

Layer 8: Impact Verification & Feedback Loops

We’d built the model. We’d survived the governance nightmare. We’d launched Round 1 and made our first awards. Now we had to prove it actually worked. This layer is where many government programs struggle. Not because they don’t create impact, but because the staff don’t always have the time, resources, or incentives to track and demonstrate impact convincingly. And without proof of impact and the ability to clearly communicate it, programs continue or end based mainly on political support (for whatever reason).

While certainly not always the case, a common challenge with impact measurement in government is that the people who design programs are often not the people who assess whether those programs succeeded. The timeline works against you. It takes 12 to 18 months to design, launch, review, select, and put awards in place. Those awards tend to last 2 to 4 years. Projects often extend an extra year (due to delays) so there is a 3-6 year gap between program justification and project closeout. DOE has roughly 10% staff turnover per year (on top of staff promotions and transfers), which means projects may have changed hands once or twice by the time the awards end. By the time you’re supposed to assess impact, you’ve moved on to your next project. Your leadership has moved on. If someone does an assessment, it may be because the organization needs to justify the program’s existence, but it’s usually a rushed exercise done by whoever is available rather than whoever created and defended the program’s original thesis.

There are many examples of good impact assessment strategies that resolve the issues described above. In the specific case of the solar prize, we avoided this trap through one critical design choice: we made the Solar Prize recurring. Not a one-off competition. A recurring program that would run every year. When you want to run Round 2, you have to show leadership that Round 1 worked. A recurring program forces you to care about impact assessment because that’s how you justify running the program again. This doesn’t work for every program, but when it’s possible, the incentives align in a way that one-off programs often can’t match.

Remember our Success Map from Layer 1? We’d identified four metrics that defined success: attract new applicants, support hard tech, build a functional pipeline, and prove the network’s value. They were testable hypotheses.

Round 1: 75% of applicants were first-time applicants to any DOE program. Rounds 2-8 stabilized at 50-60% new applicants per round. We weren’t just funding the usual suspects with a new mechanism. We were reaching an entirely new population of entrepreneurs. We tracked voucher usage. Usage rates were above 90%. We surveyed winners and asked what contributed most to their success. Winners cited the Network connections and Lab vouchers as frequently as they cited the prize money itself. This proved our support-service layer wasn’t just a nice-to-have. It was central to why the program worked. Round 1 was limited to hardware-only solutions, which guaranteed we’d hit the hard tech metric. But the other metrics - new applicant rates, network value, and pipeline function - were what actually validated the prize architecture. Those proof points became the foundation that allowed future competitions with this structure to get approved.

One of the strongest proof points came from the competitors themselves. Government work can, at times, be a little dry and boring. The Solar Prize was different. We incorporated an in-person pitch competition as the final component of the program. After submitting a written package that was reviewed in advance, the top 10 teams had 5 minutes to pitch to the prize judge and a panel of experts about why they should win $500,000. After a 1-hour deliberation, the winners were announced the very same day. For the Department of Energy, this was simply unheard of.

It was made possible by leveraging that horrible governance delay we faced in Round 1. Because we’d had to get practically every person in DOE comfortable with the first phase winners, and because only first phase winners were eligible to compete in later stages, we successfully argued that we already had approval to award funds to these groups. The later stages were just follow-on awards to what was already approved. This nuance allowed same-day review, selection, and announcement, which greatly assisted the winners.

As mentioned in Part 1, prize winners have stated the funds they receive are four times more useful than grant funds because of their immediacy and flexibility, and that the program was extremely motivating, pushing them to do far more than they could or would have done otherwise. This feedback goes directly from program participants to the DOE leadership who decide whether to fund future rounds. It creates an impact and support feedback loop that is hard to beat. The genuine joy and accomplishment shown in the below picture captures something you don’t typically associate with government funding programs.

Through six rounds from 2018 to 2023, the Solar Prize supported 140 teams with over $20 million in prizes and $5.9 million in technical support vouchers. Those teams collectively raised $300 million in follow-on funding and created 491 jobs. For a program that awarded $26 million total, that’s more than a 10x return in leveraged private capital.

This evidence enabled scaling. After Round 1, when other DOE offices asked if they should try a prize, we had a case study and metrics. The Solar Prize became the template. After Round 3, when we pushed to get prize authority formally delegated to lower levels, we had proof that prizes weren’t risky. We had run multiple prize rounds across several program offices. To date we are not aware of a prize winner doing anything inappropriate with the funds.

By 2020, prize authority was delegated down to office directors for prizes up to $10 million. This removed a massive bureaucratic barrier. The American-Made Challenge program expanded from a single $3 million Solar Prize in 2018 to over 100 prize competitions across multiple technology areas, awarding more than $188 million in funding by 2025.

That scaling happened because we could prove the model worked. And we could prove it because we’d designed impact measurement into the program from day one. The lesson: recurring programs force you to care about impact in a way that is difficult for one-off programs to replicate. One-off programs struggle to leverage learning. Recurring programs do.

Conclusion: A Prize is a Program, Not Just a Purse

In Part 1, we told the 13-year story of how prize authority at DOE went from dormant law to functioning tool. It didn’t result from a clean policy implementation. It was persistence from staff who held onto an idea after being told no, a rare alignment of political timing, and champions in leadership willing to spend political capital to try something new.

In Part 2, we walked through what happened after that approval. The 8-month slog to design the Solar Prize wasn’t just about picking prize amounts and writing rules. It was about building an entire ecosystem. The Ready, Set, Go structure addressed the motivational problem. The American-Made Network helped with the serendipity problem. The vouchers provided entrepreneurs access to cutting edge capabilities that weren’t available otherwise. The governance mapping (eventually) got us through the announcement problem. The recurring structure forced us to care about impact measurement in a way one-off programs don’t.

The success of the American-Made model wasn’t the mechanism itself. It was the design behind it. Capital alone is rarely enough. You need structure, support services, partners, timeline, and a way to prove it worked. Those prizes supported hundreds of teams who went on to raise hundreds of millions in follow-on funding and create thousands of jobs.

But that knowledge largely stayed within DOE. Each new organization running prizes, whether in government or the private sector, still has to figure most of this out from scratch. The craft of designing and deploying capital effectively is learned through apprenticeship. It takes years to understand not just the mechanisms, but the governance, the politics, the timing, and all the unwritten rules that determine whether a good idea becomes a real program. That apprenticeship model is slow, fragile, and dependent on the right people being in the right place at the right time.

We’re trying to document these lessons. The prize ecosystem needs a community of practice. The broader innovation funding ecosystem needs better coordination and shared learning. Innovation Waypoints is our attempt to share what we’ve learned and create space for practitioners to learn from each other.

If you’re working on innovation funding programs, whether in government, philanthropy, or the private sector, we’d like to hear from you. What are the challenges you’re facing? What lessons have you learned? What would help you do your job better?

Stay tuned for Part 3: When Not to Use a Prize

Innovation Waypoints is brought to you by Waypoint Strategy Group.