Not in the casual, “I watch Netflix sometimes” way. I mean the kind that justifies two TVs in a one-bedroom apartment -- one in the living room, one in the bedroom -- I might as well justify this as market research.

But let me get the record straight and clarify that I’m particular about my content consumption. There is nothing I love more than a show I can live inside, one that pulls me into its world and keeps me there long after the credits roll. While the internet has handed audiences para-social bonds with influencers who feel like friends, I keep returning to fictional universes and the characters who inhabit them. The essay I wrote, “What Six Months of Don Draper Taught Me,” is essentially a confession of how deep that goes.

So when micro dramas started generating real industry conversation, my first reaction was the same skepticism I heard from nearly everyone in this business: is this the pivot? Is this actually where we are headed?

Then I looked at the numbers.

And I could not look away.

First things first: let’s get the story straight.

The conversation in Hollywood right now goes something like this: micro dramas are proof that audiences have lost patience, prestige storytelling is on life support, and the future of content is essentially a TikTok that remembers to hire a real writer. The creative community hears this narrative and panics, which is understandable.

What I want to push back on is where that panic leads.

What Hollywood is doing right now is the intellectual equivalent of a wine snob in 2005 holding up a box of Franzia and declaring the death of viticulture. Nobody seriously drinks that, right?

Well well well… boxed wine has quietly outsold entire vineyard regions and became one of the fastest-growing segments in the alcohol industry.1 The box did not kill good wine. It expanded who wine was for. (fwiw I like Spanish boxed wine, Quadrum, it’s pretty good and ~$30 a box).

The assumption that highbrow equals long-form is not a law of physics. That’s a bias. And it is an expensive one to keep mistaking for logic.

When you look at the total runtime of a micro drama season, you are often looking at something equivalent to a movie, sometimes longer. While the story length has not changed, the packaging has. Instead of one two-hour block, you get seventy episodes of two minutes a piece, sometimes shorter.

And it’s evident this format is resonating with consumers, the numbers are not shy.

Global revenue from micro-drama apps surged from $178 million in Q1 2024 to nearly $700 million in Q1 2025.2 The U.S. market alone hit $1.4 billion in 2025, and Q3 2025 revenues outside China doubled year-over-year.34 ReelShort crossed $1.2 billion in gross consumer spending last year.5 DramaBox turned a profit on $323 million in revenue.6

Meanwhile Hollywood’s box office is still down 23.5% from 2019.7 Production volume dropped 40% from pre-strike levels.8 The industry has not recovered the 18,000 full-time positions lost since 2022.9

None of this is happening at the expense of long-form storytelling. The audience devouring micro dramas on a Tuesday lunch break could be the same one finishing a prestige series by Friday night.

These are not competing behaviors.

They are complementary ones, coexisting on the same phone, within the same attention span, inside the same week. While the format changes, the appetite does not.

There are two audiences here. Most studios are only seeing one.

Here’s another upfront confession: I am not the target audience for micro-dramas. I cannot claim to feel what their core viewer feels or naturally empathize with why the format works for them. However, when I put on my objective business hat, I set my bias aside and asked the questions that actually matter:

• Why is micro-drama viewership rising, and why now?

• What is this format offering that long-form is not?

• Who, exactly, is watching?

• And what does any of this mean for the broader content business?

Those questions led me somewhere the industry has been slow to go.

The pattern I keep seeing plays out the same way: A studio or network acknowledges the explosive growth of micro-dramas, runs the numbers on the market, and promptly treats it as another format to fund in an already overcrowded content portfolio.

What does not happen is the harder work: understanding how micro-drama fits into the full arc of how a viewer moves through content, across connected TVs, mobile, and everything in between. The crossover behavior, the viewer who belongs to more than one format depending on the day, stays invisible because no one is looking for it.

That is the gap. And it starts with a misread of who is actually in the audience.

Because when I look at who is watching micro-dramas, there is more than one customer segment. I see two. And the distinction between them is not a minor footnote.

Audience Segment #1: The Pure Micro-Drama Viewer

Profile 1: The Void Filler

You definitely know this person. You might even be this person. This person pulls out their phone before the subway doors even close. Not because they have messages to answer or anything urgent happened, but because twelve seconds of standing there with their own thoughts is apparently too much to ask. Five episodes of a three-minute drama later, they have successfully avoided eye contact with every single person in the car and arrived at their stop feeling oddly accomplished.

Short-form is not a preference for them. It is a coping strategy with a data plan.

Life does not happen in two-hour blocks. It happens in cracks. The commute. The lunch is somehow only fifteen minutes long. The couch, after the kids are down, when your brain wants something but cannot commit to a full episode of anything. Owl & Co’s State of Microdrama 2026 report found that 70% of viewers are watching vertical series in bed before sleep, and 59% on the couch.10 These are not people clearing their calendar for content. These are people filling the margins of their day with something that finally fits there.

They are already paying. Figure out what they want and serve it to them.

Profile 2: The Daydream Buyer

This one says more about where content is actually going.

This person is not zoning out, they are the browsing type. Scrolling through someone else’s morning routine, someone else’s apartment, someone else’s income bracket, holding it up against themselves to see what fits. The restaurant they will probably never eat at. The wardrobe that belongs to a version of themselves they are still negotiating with. They are not killing time. They are borrowing a life, or at least the curated portion of one that makes it to the feed.

Short-form is not a distraction for this viewer, it’s a window. And they will pay to keep it open.

That second profile is more understated to the industry and willing to pay more.

Audience Segment #2: The Crossover Viewer

This is the person at a steakhouse on Thursday, ordering the ribeye, discussing the season finale of Severance with genuine conviction. And then at 3am, the same person, ordering McDonald’s through the app while burning through six episodes of a micro-drama they will never mention at work.

The audience for this format is not who the industry assumed it would be, and it is not staying still.

Women built the early core of micro drama viewership, and that foundation has not gone anywhere. But look at what has happened alongside it. The male audience on My Drama grew 28-fold over seven quarters, climbing from 1.1% to 30.3% of monthly active users.11 It’s a structural shift in who is finding this format and deciding it belongs in their life.

Both of those people at 3am are real. The one the industry built its assumptions around, and the one it quietly decided did not exist. Neither cancels the other out. What is harder to defend is a content pipeline that insists on treating them as two separate audiences, sorting micro-drama and long-form into walled-off worlds, as if no single person could belong to both. Most people belong to both.

What problems do micro-dramas actually solve for the viewer and the entertainment industry?

Once you accept that both audiences exist and the format is real, the more useful question is what problem micro-dramas are actually solving?

And the answer is not one thing.

1. Monetizing Attention at the Pocket Level

Our lives are not built around blocks of time anymore. They are built around gaps: the three minutes before a meeting, the Uber ride, waiting in line, the window between getting into bed and actually falling asleep. Nobody is starting a prestige drama in those moments.

But they will absolutely watch one more episode of something that meets them right where they are.

That behavior is not new. Micro-dramas just figured out how to put a price tag on it. And that is not a small thing. That is the entire attention economy finally being monetized at the unit level where it actually lives. Unlike major streamers that offer a bottomless buffet, micro-drama pricing is a mobile-gaming-inspired model: a series offers the first 5 to 15 episodes free before hitting a paywall, requiring users to spend virtual coins that cost roughly $0.10 to $0.20 per episode.12

While major streamers like Netflix want you to watch as much content as possible, a pay-per-view model might actually increase profit margins even if it reduces aggregate hours viewed. That is a trade worth taking seriously.

2. Solving the Customer Acquisition Problem

Here is where it gets interesting for anyone watching subscriber growth numbers flatten in real time.

Major streamers are stuck in the same loop: spend more on marketing, green-light another tentpole, hope the next big show moves the needle. Micro-drama platforms are running a different play entirely. With pay per view or weekly subscriptions, consumers invest a dollar or two before anyone asks for a monthly commitment. The ask is small enough that saying yes requires almost no convincing.

But here is the part nobody is talking about: who is actually making that first transaction?

A meaningful chunk of is likely to be the crossover viewer. The person who already has a Hulu subscription, already watches prestige drama, and is also quietly spending three dollars on a micro-drama app at 11pm because it asks nothing of them and delivers immediately. They are not replacing one format with the other, it’s an addition. And they are proving, with their own wallet, that the appetite for both can coexist in the same person on the same night.

That is not a churn problem. That is an acquisition opportunity that the major platforms are currently handing to someone else.

3. The Long-Tail Retention Flywheel

This is the part the industry is most underestimating.

A person who finishes a micro-drama and wants more has already done something valuable: they proved they will pay for narrative content. They are a warmer lead for a full subscription than someone who has never engaged with your platform at all. The discovery cost is already spent and as a bi-product, the trust is already partially built.

Micro-dramas do not have to be the destination. In fact, they can be the front door. And for platforms bleeding subscribers at the top of the funnel while spending aggressively to replace them, a cheaper, lower-friction entry point that converts into long-term retention is not a side business. It is a lifeline.

Whether micro-dramas can effectively subsidize ad-free tiers for major streamers is still an open question. But from a revenue architecture perspective, the model at least deserves a seat at the table.

4. A Cheaper Method for Content R&D

Here is the angle that should appeal most to anyone with a spreadsheet open.

You can film a season of a micro drama in a week. The production cost is a fraction of a standard pilot. And you can have real engagement data back from a live audience within days. It’s not just a cheaper product, but it’s an easy method of solution validation (as the tech bros would say).

Right now the way studios validate a concept is through development: pitch meetings, script notes, pilots that cost millions before a single viewer has weighed in. The bet is enormous before you have any signal at all. While Micro dramas give immediate feedback, sometimes time and the zeitgeist WoW can impact metrics, and that’s another debate about what is the definition of success…

Anyhow, if you produce enough of them and actually read the data, patterns emerge fast. The Owl & Co State of Microdrama 2026 report found that romance accounts for 76% of female viewing and 54% of male viewing on leading micro drama platforms.13

That’s not a coincidence.

That’s a signal, and it is exactly the kind of revealed preference data that is nearly impossible to extract from a traditional development process.

A particular emotional note: yearning, unresolved romance, revenge. Something is generating outsized engagement compared to everything else. You’re not guessing, you’re reading revealed preferences from a paying audience.

It is pilot season with a fraction of the risk.

Not every micro drama concept will translate to long-form. Some formats do not stretch, and some stories are exactly as long as they need to be. But for studios trying to test a new concept or a new genre approach before committing to a full series order, the micro drama format is as close to an MVP as this industry has ever had access to.

There is a creator dimension to this too. The format is accessible in a way that traditional Hollywood pipelines are not. An emerging writer or director does not need a development deal or a legacy relationship to make something in this format and put it in front of an audience. That means studios investing in micro drama ecosystems are also, whether they plan it this way or not, building a pipeline of emerging talent whose traction is already visible before anyone takes a blind bet on them. The first look deal of the future might not start with a pitch. It might start with a micro drama that quietly racked up two million views.

5. The Legacy IP Sitting Untouched

There is a another opportunity that does not get enough attention, and it starts with what studios already own.

Right now there is an entire generation of viewers who are culturally fluent in The Sopranos without having watched it. They know Tony’s duck. They have seen the memes. They have watched the “best moments” recaps. The show is alive for them in a real way, even if they have never sat through a full episode.

The traditional view is that this is a loss. To be candid, it is an unactivated asset.

That person already has an emotional relationship with the characters. They understand the cultural weight of the property. The gap between them and a full subscription is smaller than it looks, because the hardest part of the acquisition problem, getting someone to care, is already done.

The opportunity is to meet them where they are. Reintroduce the show through a format they will actually engage with: structured recaps, character breakdowns, season summaries that are entertaining on their own terms and function as a pathway into the full library. Not as a concession to short attention spans. As a deliberate on-ramp that compounds over time and serves an audience that legacy content already has a claim on, but is not doing anything to activate.

Studios that own deep libraries are sitting on reintroduction opportunities they have not started to build. The format is right in front of them!

The Real Question

The conversation Hollywood keeps having is whether short-form content threatens long-form storytelling. That is the wrong frame.

The right question is whether the industry is willing to stop treating its audience as a monolith with one mode of consuming content, and start building for how people actually show up to media across a week.

Sometimes it is a Saturday night with dinner, company, and the full season ahead.

Sometimes it is a Tuesday at midnight with a phone and twenty minutes.

Both of those moments can belong to the same person, and sometimes they are your audience. Right now, one of those moments is being captured by platforms the industry is calling a threat instead of a lesson.

The numbers are not subtle and the behavior is already there. The only question left is who builds for it first.

As you know, I usually write the deep analytical pieces about the turbulent masochistic industry of media and entertainment. This is not one of those.

Everything here is public information, nothing proprietary. But since I’ve been asked “what’s the latest?” so many times this month that I figured the most honest way to cover it is satire, because what’s happening in the media industry right now is beyond serious analysis. It’s worse than puberty. At least puberty has an end date.

So grab your beverage of choice and find a lounge chair. The Warner Bros. Discovery saga is a full season of Love Island that no one asked for but everyone is watching. 👀

This Season’s Line-up

Warner Bros. Discovery (WBD) is our girl. She walked into the villa carrying HBO, the DC Universe, Harry Potter, Game of Thrones, and CNN in her luggage. Absolute catch. But she’s also carrying $33 billion in debt, which, in Love Island terms, is the emotional baggage from a previous relationship that she just can’t stop talking about. Her previous coupling with Discovery was... let’s call it a “journey.” And her agent, David Zaslav? He’s the one negotiating every deal on her behalf, except nobody’s totally sure if he’s looking out for her or for himself. Paramount even offered to keep him on as co-CEO in one of their bids, which tells you everything about who he’s really negotiating for.

David Ellison (Paramount Skydance) is the guy who just walked into the villa after already completing his own glow-up. Son of Oracle billionaire Larry Ellison, fresh off merging Paramount and Skydance in mid-2025. He’s got his dad’s money backing him up and he’s got confidence. He’s the contestant who says “I know what I want” in his intro package, and you half believe him.

Ted Sarandos (Netflix) is the one who’s been on the show for five seasons. He’s comfortable. He’s established. He doesn’t need to be here but showed up anyway because he heard someone interesting was in the villa. Cool, calculated, and if the price isn’t right, he’s happy to go home alone.

Comcast is the bombshell who arrives on Day 3 and leaves on Day 4. We’ll get to him.

Episode 1: “Can I Pull You for a Chat?”

September 2025

Paramount makes the first move. David Ellison, barely done unpacking the wedding gifts from the Skydance merger, slides into WBD’s DMs in September. The opening offer? $19 per share. In Love Island terms, this is the equivalent of “You’re fit, do you want to share a bed?” without even learning her last name.

WBD’s board takes one look at this and does the villa equivalent of saying “I see you more as a friend.” On September 22, Zaslav sends a letter back to Paramount: the board has “unanimously concluded” the proposal is inadequate. Thanks, but no.

Paramount, undeterred, comes back five days later. Bumps it to $22 per share, 67% cash, a $2 billion breakup fee if regulators kill it, and, here’s the sweetener, a proposal for Zaslav to stay on as co-CEO and co-chairman. That last part is the reality TV equivalent of telling a girl’s best friend you think she’s great too.

Still a no.

Episode 2: “I’m Open to Getting to Know Other People”

October-November 2025

WBD does something that changes the whole dynamic. On October 21, she announces she’s “considering alternative options.” In Love Island language, this is the girl standing up at the fire pit and saying “My head could be turned.”

And just like that, the boys start lining up.

By November, it’s a full bidding war. Paramount, Netflix, and Comcast all submit first-round bids by the November 20 deadline. Three suitors. One rose ceremony.

Let’s talk about Comcast for a second.

Comcast is the contestant who arrives as a bombshell, says all the right things in the Beach Hut interview, but only wants part of the package. He didn’t bid for all of WBD.

He only wanted the studio and HBO Max. That’s like telling someone you love their personality but you’d really just like to date their legs. Bloomberg reported that Comcast knew the odds were low going in. He came, he saw, he left. Classic Day 4 elimination.

Episode 3: “I’ve Made My Decision”

December 2025

December 5 hits, and WBD makes her choice. She picks Ted Sarandos.

The deal is valued at $82.7 billion. Netflix agrees to pay $27.75 per WBD share, a mix of cash and stock. He’s taking home the studio, HBO, the streaming platform, and all the IP that comes with it. And the Discovery linear networks, the cable channels, all that baggage she’s been carrying around? Those get separated out and spun off as “Discovery Global.” Ted’s not just proposing, he’s saying “I’ll take you, but we’re leaving your baggage at the door.” The perfect prince. All her problems are about to magically disappear.

Netflix secured $59 billion in financing from a bank consortium. He showed up with a ring and a house.

But here’s where it gets messy.

Three days later, on December 8, Ellison launches a hostile takeover bid. $30 per share, all cash, straight to WBD’s shareholders. Bypassing the agent entirely. This is the guy who shows up at the villa uninvited, stands outside with a boombox, and says “I’m not leaving until you hear me out.”

The total value of this hostile bid? $108.4 billion, including WBD’s debt. And unlike Sarandos, Ellison isn’t just cherry-picking the good parts. He’s offering to take all of WBD. The cable channels, the debt, the baggage, everything. He wants the whole woman, not just the highlights.

WBD’s board unanimously tells shareholders to reject it. WBD chairman tells CNBC that choosing Netflix over Paramount “was not a hard choice.” That’s the kind of line that gets replayed in the “previously on” segment for weeks.

Episode 4: “My Head’s Not Been Turned... Yet”

January 2026

Sarandos, sensing the pressure, amends the deal on January 20. Converts the whole thing to an all-cash transaction. Same $27.75 per share, but no stock component anymore. This is the moment on Love Island where Sarandos sees Ellison pull WBD for a chat and suddenly starts cooking her breakfast, cleaning the villa, and writing her a speech for the next recoupling. He was already committed, but now he’s making sure she feels it. The move simplifies the deal, gives WBD shareholders more certainty, and accelerates the shareholder vote timeline to April. Sarandos not panicking, but he’s definitely trying to lock this down before she has time to think.

Meanwhile, Ellison’s hostile tender offer is just... still there. Sitting on the kitchen counter like a love letter nobody asked for. It got amended on December 22, extended to January 21, and WBD’s board tells shareholders to reject it again. That’s now six formal rejections. Six proposals, six no’s. At this point Ellison isn’t even getting rejected anymore. He’s just part of the furniture. The villa has accepted his presence the way you accept a weather pattern. He’s not leaving, and somewhere deep down, everyone knows it.

Episode 5: “It’s a Lot”

February 2026

This is where the whole season flips.

On February 24, Ellison comes back with a revised offer: $31 per share, all cash. The full picture is $110.9 billion all-in. That includes a daily ticking fee of $0.25 per share per quarter, a $7 billion termination fee if regulators block it, and Paramount covering Netflix’s $2.8 billion breakup fee. That’s Ellison saying he’ll personally pay Sarandos back for the engagement ring. He’s not just outbidding the guy, he’s settling the tab on the way out. And because WBD’s board has a fiduciary obligation to its shareholders, they can’t just ignore a higher offer because they like the other guy better. They have to take it to the table. So the board determines that Ellison’s proposal “could reasonably be expected to lead to a Superior Proposal,” which triggers a matching period under the existing Netflix agreement.

In villa terms: Ellison just walked back in after six rejections, baggage and all, and said “I want all of you, and I’m offering a better life than he is.” Not just more money. He wants the cable channels, the debt, the whole mess. And this time, the numbers are too loud for the villa to pretend they don’t hear it. She just told the fire pit she might actually like the other guy more.

Under the terms of the Netflix deal, Sarandos has four business days to match. Four days. That’s one very long recoupling ceremony.

On February 26, Sarandos walks. “At the price required to match Paramount Skydance’s latest offer, the deal is no longer financially attractive, so we are declining to match.” Looks at the camera, shrugs, gets in the cab without crying.

WBD pays Netflix a $2.8 billion termination fee and officially couples up with Ellison.

Now, a note from the audience: There’s a contingent watching this season who thinks the real happy ending was always WBD ending up with nobody. It’s 2026. A woman doesn’t need a man. An independent WBD, restructured and unburdened, might have been the most exciting outcome of all. And if she had to couple up, the sentiment leaned Sarandos. He was only taking the best parts, sure, but at least the vision felt like a true new start for WBD. Ellison’s offer is bigger, but bigger isn’t always better when you’re absorbing an entire company’s worth of complexity on top of your own.

Then again, $31 per share is $31 per share, and the shareholders aren’t here for a love story.

Episode 6: “The Final”... or So We Thought)

April 2026

On April 23, WBD shareholders vote overwhelmingly to approve the Paramount deal. $31 per share. A 147% premium over WBD’s unaffected stock price of $12.54. And in one delicious detail, Zaslav, the agent, tried to claim a massive payout for brokering the whole thing. His pay package went up for a vote at the same meeting, and shareholders shot it down. “We’re happy with the wedding, but the matchmaker doesn’t get a commission on this one.”

The DOJ cleared the deal in June. Everything looked like it was heading toward a Q3 2026 close. The couple was about to walk out of the villa hand in hand.

The Reunion Special: “We Need to Talk”

July 2026

Except this season isn’t over. Because the parents got involved.

See, in any good Love Island finale, the couple has to meet each other’s families. In this case, the family is the U.S. government, and the parents don’t agree. Dad (the DOJ) gave his blessing in June. He looked at Ellison, looked at the numbers, and said “Yeah, I’m fine with this.” But Mom? Mom is twelve state attorneys general, led by California AG Rob Bonta, and Mom has been sitting quietly at the table this whole time, watching everyone celebrate, waiting for the right moment.

On July 13, she stood up. Twelve AGs filed an antitrust lawsuit to block the merger, alleging the combined company would violate the Clayton Act by reducing competition in wide-release theatrical distribution, top-grossing theatrical distribution, and basic cable licensing. Mom doesn’t think this man is good enough for her daughter, and she brought receipts.

A federal judge granted a temporary restraining order on July 20. And on July 24, Ellison agreed to postpone the merger until five days after an antitrust trial or June 1, 2027, whichever comes first.

The venue is booked. The invitations went out. Dad’s already in his suit. But Mom just stood up at the ceremony and said “I object,” and the officiant looked around and said “...yeah, actually, we should probably hear her out.”

Where We Are Now

The couple is in limbo, and the meter is running. After September 30, Ellison owes WBD shareholders a ticking fee that could exceed $1 billion if the wedding keeps getting pushed. But the money isn’t even the scariest part. As Puck’s Eriq Gardner put it, “the company David Ellison buys six or 12 months from now may not be worth what it is today.” The villa doesn’t pause just because you’re stuck in a recoupling. The other contestants keep moving. Cord-cutting, AI, shifting consumer habits -- WBD is standing still in an industry that isn’t.

Ellison’s play is to push for a full trial, potentially as early as November, and he’s hired Beth Wilkinson, the lawyer who helped Microsoft beat the FTC’s Activision challenge, to make his case. He’s not fighting the parents over dinner anymore. He’s taking them to court.

And then there’s Zaslav. The agent. On June 4, 2027, he acquires the contractual right to walk away from the entire deal and collect a $7 billion regulatory termination fee. If the wedding doesn’t happen by that date, the agent gets paid more than anyone. And you wonder whose interests he’s been representing this whole time.

Sarandos, meanwhile, is at home on the couch, scrolling through the villa, quietly grateful he didn’t overpay.

Stay tuned for the trial. Next season will be even spicier.

All information in this piece is sourced from public filings, press releases, and news reporting.

It’s 2026, and this year’s content calendar reads like something most people could have predicted half a decade ago without a single proprietary dataset: Toy Story 5, a new Star Wars, another Spider-Man, another Avengers, The Devil Wears Prada 2, a Harry Potter reboot. The most sophisticated audience intelligence infrastructure ever built looked at everything it knows and concluded that the best move was to do it all again.

I find that genuinely fascinating. Not because these films will fail to make money (most will, and the models will confirm it), but because it reveals how deeply the industry has over-indexed on one signal: popularity as a proxy for likelihood to watch. That signal is not wrong. People will, in fact, watch these films. The problem is that “will they watch it” is the wrong question to optimize around when the actual decision on the table is what to build next.

I think about this more than I’d like to admit because I drive strategy for building and enabling insights from first-party data products for a streaming platform. I have watched this industry accumulate more behavioral data while using it the way a retail day trader reads a stock ticker: always staring at the last candle, always one beat behind the move that already happened. The infrastructure is imperfect, the dashboards are full of numbers that very smart people spent real money building, and the strategic output of all that machinery consistently amounts to an elaborate, expensive description of what already occurred, paired with the hope that it occurs again.

I’ve written about the technical dimension before in pieces such as Nobody Can Tell You What’s Actually Making Money or Streaming Has Trading-Floor Infrastructure and a 1990s Cable Playbook . Those essays catalog the tactical wreckage: fragmented first-party ecosystems, data stacks that survived three mergers and still cannot communicate with each other.

But the deeper failure is not a plumbing problem. It is an interpretive one.

Nobody Wants to Say This Part Out Loud

Legacy media companies are content companies, they are not tech companies and the distinction is crucial.

People come to streamers like HBOMax or Disney because there’s a desire to consume a specific piece of content on the platform, not because you’re hyped about the change of a feature or a button (maybe that too). I say this as someone who works in the product department at one of these companies, fully aware of the irony…

Content comes first, and the tech exists to surface it, to help people find and stay with the things that matter to them. When that order quietly inverts, when tooling starts shaping creative priorities instead of illuminating them, the customer experience fractures in ways that are difficult to diagnose from inside the building, because every individual decision still looks defensible in isolation.

From the customer’s perspective, it looks like a product built by people who could not agree on what it was supposed to do. Product teams optimizing for one set of metrics, data teams measuring against a different set, marketing pulling in a third direction, content teams operating on instinct because the analytics they were handed do not describe the thing they are trying to understand. Meanwhile, someone is sitting on their couch trying to find a show.

What follows are three misconceptions that keep surfacing in public discourse about how data operates in this industry, not because the people advancing them are unintelligent, but because the distance between how data gets discussed publicly and how it actually functions inside a streaming operation is wide enough to produce consistently wrong conclusions.

Misconception One: Trends Tell You What to Make

I’m talking to you development execs and producers!

This conversation happens constantly: someone pulls up a chart showing a trend, maybe romantic comedies are surging, maybe there is a spike in anime viewership among 18-to-34-year-olds, and the room nods along like something important just happened. Someone says, “the data is telling us something.” Then the room proceeds to treat a pattern as a prescription.

“Horror is having a moment.” “Limited series are outperforming returning seasons.” “Audiences are craving nostalgia.” Each of these statements may be statistically true at a given point in time, and none of them, on their own, tell you what to green-light next.

A VC Would Never Invest Like This. So Why Does Hollywood?

When a venture capitalist evaluates a startup, they are not reading the market as it exists today and extrapolating forward. They are modeling the world as it is becoming, asking what the world is going to be ready for in five to ten years, and whether this team can build it before anyone else realizes it needs to exist. The best VCs are not looking for the next Facebook by funding another Facebook.

They are placing bets on conditions, on climate, on the underlying forces that will make a particular product feel inevitable by the time it arrives. When a financial institution hires quantitative analysts, those analysts are not drawing trend lines forward into next quarter. They are modeling the environment: political climate, regulatory shifts, consumer sentiment indicators that collectively shape the conditions under which a particular asset might perform.

Here’s a meme that’s not far off but proves the point:

The climate is the input and the trend is the output. The entire discipline is built around understanding the inputs well enough to place a bet before the output becomes obvious to everyone reading the same chart.

Prediction Models based on Popularity are NOT the Answer

Legacy Media, by contrast, tends to operate as though it is trying to find the next Facebook by building another Facebook, which is why 2026 looks the way it does. On a spreadsheet, those titles will perform. Nobody needs a data scientist to tell them that a movie starring Matt Damon, Ben Affleck, and Taylor Swift is going to open big. That is not insight, that is arithmetic. And confusing the two is how you end up with a release calendar that makes money quarter over quarter while slowly bleeding out the cultural conditions that make long-term franchise value possible in the first place.

Here is my AI generated movie poster (just goes to show humans are truly still needed for this, smh)

Cultural impact is what actually compounds in this business. The story itself is what turns a show into a franchise, a character into a reference point that outlives its own decade, a subscriber into someone who stays for years instead of months. The spreadsheet will always recommend the safe bet because the spreadsheet has no way to model something that has not happened yet.

Misconception Two: Statistics Settle the Argument

This one shows up whenever someone presents data in a room as though the conversation is over once the numbers hit the slide. “This genre is trending up 40%.” “Completion rates on eight-episode seasons outperform ten-episode seasons.” These are facts, and they are also, on their own, almost entirely useless for the decision that actually matters: WHAT we should green-light and WHY.

The problem with statistical observations divorced from causal understanding is the same problem you’d have if you noticed that more people are investing in AI stocks without understanding the actual health of these companies.

SpaceX Stock and Short-form Content have the Same Problem: Hype is Not Value

SpaceX is widely celebrated and commanding a valuation that reflects enormous belief in its future, and yet the stock tells a more complicated story. It has been plummeting, and it may be sitting right around where it was when it first IPO’d. Highly valued does not mean financially healthy. The company has not been consistently profitable overall, and the enthusiasm baked into the price has always been a bet on future conditions, not current ones.

AI investment operates on similar logic, and so does the current obsession with short-form content. Shorts are having a moment and the engagement numbers look promising. So the industry responds the way it always does: everyone rushes to make shorts.

However, high engagement and durable value are not the same thing.

A viral short delivers a specific, narrow kind of value. And yet, it’s neglected to address the longevity of short form content.

• Does it compound in ad-revenue?

• Can short-form content alone build a fan-base akin to several seasons of a show?

Which is why the observation that something is getting traction might prompt a good question, but it should never be confused with an answer.

Creative executives, at their best, operate on instinct and cultural intuition, asking what emotional territory should we be exploring, what cultural conversation has not been had yet. Data and algorithm teams are asking a completely different set of questions: given what this user watched last, what are they most likely to click next, what thumbnail variant reduces scroll time?

Both sets of questions are legitimate and are grounded in something real. However, they are operating at different altitudes, and the assumption that answering one automatically informs the other is where the industry keeps getting stuck.

The creative instinct that matters most has always been the ability to sense where the culture is heading rather than where it has been. Data, by its structural nature, is retrospective. It tells you what people did. However, it cannot tell you what they are ready to feel for the first time. That is not a criticism of data. It is a description of a structural limitation.

As someone who stares at numbers and evangelizes graphs all day, the people who use data most effectively understand that limitation intuitively. They use it to validate instincts, pressure-test assumptions, identify underserved audiences, and optimize the surfaces through which content reaches people. They do not use it to replace the human judgment that decides what gets made, because that substitution is how you end up with a content slate that tests well in every model and moves nobody.

Misconception Three: Streamers Know What Their Data Is Worth

This is the misconception closest to my daily work, and it is the one that genuinely frustrates me the most.

From the outside, streaming platforms look like data juggernauts: hundreds of millions of viewing hours tracked in real time, behavioral signals on every click, every pause, every moment someone hovers over a thumbnail and decides to keep scrolling. The assumption, both inside and outside the industry, is that this data is already being leveraged to its full potential.

I can tell you from inside the building that this is not the case, and it is not even particularly close.

The core problem is that most media companies are still trying to answer the fundamental question of what constitutes real signal versus noise.

Inside a streaming platform, content gets categorized along axes that make sense for portfolio management: how many crime shows did we green-light this year, how does our genre distribution compare to last quarter. This is useful for managing a content budget. It is also almost entirely disconnected from how an actual human being relates to what they watch.

Even still, most platforms are measuring one variable, viewing time, while remaining functionally blind to the three or four variables that actually explain why someone stays or leaves.

It is the equivalent of a fund manager who tracks trading volume obsessively while remaining indifferent to whether anyone is actually making money on the trades.

That is not a strategy problem you can solve with a better dashboard. It is an infrastructure problem baked into the organizational DNA of every post-merger media conglomerate that inherited incompatible tech stacks, conflicting definitions of success, and a measurement culture built for linear television that has been duct-taped into the streaming era without anyone stopping to ask whether the underlying assumptions still hold.

The Real Argument

The media industry keeps reaching for data to confirm what it already believes rather than using it to discover what it does not yet understand.

The quant at a financial institution does not look at a stock chart and say “this went up, so it will keep going up.” They model the environment, simulate the conditions, and price the difference between the version of the world that makes an outcome likely and the version that makes it unlikely.

The best creative executives have always done something analogous without calling it by that name. They read the cultural temperature, sense the restlessness and the unnamed feeling a critical mass of people are carrying around, and build something that gives that feeling a name, a face, and a story to live inside.

The 2026 summer slate proves the point. Five major sequels, every one greenlit by backward-looking data.

• The Devil Wears Prada 2, a sequel to a twenty-year-old film with no franchise architecture, returned nearly seven times its budget because someone read where the cultural conversation around work and ambition had moved.

• The Mandalorian and Grogu had everything the algorithm loves (the Star Wars brand, a pre-built audience, a character that moved billions in merchandise) and became the lowest-grossing live-action Star Wars film in history.

• The Avengers: Doomsday, at roughly $800 million all-in, needs $1.5 billion to break even because Endgame made $2.8 billion and someone looked at a chart and made the precise error a quant never would.

The data said the audience was there. It was measuring the audience that had been there.

That is not a mystical gift. It is a discipline, and data, used properly, should make that discipline more precise rather than replace it with a spreadsheet that optimizes for predictability at the expense of everything that makes this business worth being in.

The companies that bridge what the algorithm knows with what the creative team senses will build the next decade of culture. The ones that keep confusing the rearview mirror for the windshield will keep wondering why nothing they make seems to matter the way it used to.

Don Draper is the quintessential anti-hero:

• He builds an entire career on selling people authenticity he has never once possessed.

• He’s a man who stole a dead soldier’s name in the Korean War and turned the lie into a corner office on Madison Avenue.

• He’s a serial adulterer who betrays whichever woman he’s committed to.

• By season four he’s a functioning alcoholic dictating rambling late-night letters to no one in particular and passing them off as “creative breakthroughs.”

Nevertheless, the combination of charm, vulnerability, and an almost supernatural talent for selling people their own happiness makes him irresistible.

And yet after seven seasons, and thousands of YouTube videos that have fostered an ecosystem of character breakdowns still getting clicks years after the finale, what’s all the hype?

Mad Men endures not because its characters are complex, but because they're incomplete in the right places, and their contradictions feel less like writing choices and more like inevitabilities. I'd finish an episode the night before and find myself deep in rabbit holes of YouTube during my lunch breaks, embarking on videos such as Jon Hamm in an SNL skit or a video essay explaining the essence of the story engine that drives the entire show (these are some of my favorites).

I build data products for a major streaming platform, so I spend my days watching audience behavior at scale. And from that vantage point, I’ve become convinced that productive irresolution is the single most under-leveraged principle in modern TV development. It is the quality that separates a show people simply finish from a show people can’t stop watching. The yearning desire to immerse one’s self in the world of a show is what creates the supplemental content ecosystems, the decades of analysis, and the cultural longevity that every streamer is now chasing (and that most are failing to produce, for reasons the data is making painfully clear).

This essay is a character study of Don Draper as a way of reverse-engineering what that quality actually looks like when it's built right. We'll go through the relationships, the self-sabotage, the professional arcs of the people around him, all to understand the specific architecture that makes a character outlive the show they were written for.

And then we'll turn to the numbers, because Netflix and the rest of the streaming industry are hemorrhaging viewers after season one and they cannot figure out why. Shows are launching to massive first-season audiences and then losing 30, 50, sometimes 70 percent of them by season two.

The industry is treating this as a retention problem, when in fact, it is a construction problem. These shows are being built to serve a premise for a weekend rather than to create characters worth returning to in the years to come.

Once you understand what productive irresolution actually looks like, the reason becomes obvious. The fix does not live in the data science or algos team. It begins on the creative side with development teams and writers rooms.

Don Draper is one of the clearest proofs of concept I’ve encountered.

The Paradox of the Lovable Disaster

So here’s the thing about Don that should not work but absolutely does:

In the Carousel pitch for Kodak, he sells a room full of executives on the sacred beauty of family while actively dismantling his own.

He delivers the Hershey pitch as a polished origin story, then unravels mid-sentence and confesses he grew up in a brothel, orphaned, and stealing chocolate bars from the pockets of johns.

Timing Disguised as Compatibility

And then there’s the pattern with women, which the show maps across three relationships that each reveal a different facet of the same flaw.

Betty, Don’s first wife, is the perfect image, the suburban wife who validates the entire Don Draper construction. She’s beautiful, composed, and crucially, she never asks what’s underneath. Their marriage works as long as it stays on the surface. The moment Betty begins to see through the performance, the marriage doesn’t just strain. It becomes structurally impossible.

Which brings us to Faye Miller, a woman who is arguably Don’s equal.

She sees him clearly and refuses to be dazzled by the performance. She knows that Don stole a soldier’s identity and his real name before “Don” was “Dick Whitman” and doesn’t flinch. And then while stringing Faye along and slowly ghosting her while taking his kids on trip to California, he comes home and announces he’s marrying his secretary, Megan.

On the surface this is the quintessential cliché. But if you’ve been paying close attention across the full run of the show, it lands as almost inevitable.

So he chooses Megan, who feels like Betty’s opposite that’s warm, spontaneous, unguarded. But what actually draws him is that Megan adores the version of Don he’s performing. She doesn’t require honesty, she requires fascination. For Don, that’s not love, it’s relief.

Until Megan starts building her own identity, pursuing acting, becoming a person who no longer orbits him. Don doesn’t lose her because she changes. He loses her because she stops being his audience.

Three women, three versions of closeness, and the same exit. That’s not a character flaw the writers forgot to resolve. That’s productive irresolution doing exactly what it’s supposed to do.

The gap between those two readings is exactly where the show lives and why it’s so brilliant.

Tony Soprano and Walter White demonstrate the same architecture in slightly different styles. You can map the logic of their choices and still find yourself surprised by them. That’s not a bug, it’s the whole feature.

It Was Never About You. It Was About the Season.

Fiction offers a particular kind of clarity that real life almost never does. In our own relationships, we’re inside the weather, reacting to it, unable to step back far enough to notice we’ve been living through seasons. I want to stay with that idea for a moment, because what started as a personal realization while watching Mad Men eventually reshaped how I think about the business problem at the center of this piece.

I’ve been the Faye. Not to anyone with Don Draper’s charisma, but the architecture was the same. Someone willing to sit with the broken parts, passed over for whoever felt more like a fresh start than a reckoning. I spent months constructing explanations that had more to do with self-protection than truth, asking what’s wrong with me when the actual answer had almost nothing to do with me.

I couldn’t see that from inside it. But I could watch Don choose Megan over Faye and suddenly understand what months of self-interrogation never produced: it was never about inadequacy. Faye arrived when Don needed to be known, and that was precisely why he ran. Caring about the outcome without having my actual future tied to it removed me just enough from my own story to finally see the pattern.

I don’t think that recognition is rare. Being the Faye is something close to a rite of passage: you show up willing to see someone clearly, and you learn that clarity isn’t always what people are looking for. It’s bitter in the moment. It becomes something closer to wisdom over time. But while you’re inside it, you have no way of knowing you’re in a season and not a sentence.

That last distinction is the idea I kept coming back to throughout the whole binge: when a relationship ends, what actually ended it usually has less to do with the two people involved than with the phase of life each person was in when they met. People move through phases the way weather moves through seasons. Don cycles through them relentlessly, and the people in his life encounter entirely different versions of him depending on when they arrived. Betty, Megan, and Faye each met a different Don, and the version they got had almost nothing to do with them and almost everything to do with which season he was in when they walked through the door.

Moreover, that framework didn’t stay contained to the show. I found myself applying it to my own history: the moments where I’d been completely certain something was headed in one direction and then it just... wasn’t. The endings that still carried open questions. Before Mad Men, I kept circling those questions without landing anywhere useful. After watching Don’s arc season after season, I arrived somewhere better than a clean answer. I got a way of reframing the question entirely.

Self-Sabotage on a Seven-Year Lease

Roger Sterling is the “season’s theory” wearing a better suit. Two marriages that collapsed not because he chose the wrong women, but because he was the wrong version of himself when he chose them. His first wife got the man performing stability. His second got the man performing youth. It wasn’t until he stopped performing entirely, somewhere in the final stretch, that he ended up with someone roughly his own age, his own speed, his own weight class. Not because he finally learned his lesson, but because he arrived in a season where he didn’t need the relationship to prove something about himself.

And the framework doesn’t stop at romance. It runs through every professional arc on the show with the same quiet logic.

Joan Holloway spends years navigating a workplace that reduces her to what she looks like while she quietly becomes the most operationally competent person in the building. She claws her way to a partnership at the advertising agency they all work for “Sterling Cooper Draper Price” (SCDP) and the cost of getting there is something the show never lets you forget (yes, you remember in season five what I’m talking about).

Then McCann Erickson acquired the firm and almost overnight, the title she bled for becomes decorative. Not because Joan got worse at her job, but because the environment she’d mastered no longer existed. She recognized the turn before it buried her, and she left. What looks from the outside like giving up is actually one of the sharper reads on the show.

Peggy Olson’s arc is the inverse, and it’s the one I think about most often in professional terms. She rises from secretary to copywriter to someone whose talent is no longer debatable, and then she hits the wall that talent alone can’t break through. She watches men with half her ability get compensated at twice her rate and has to decide whether the prestige of the bigger machine is worth the daily experience of being undervalued inside it. It’s the kind of career crossroads that doesn’t announce itself with a dramatic moment. It accumulates. You wake up one Tuesday and realize you’ve been weighing the same tradeoff for two years without naming it: be the star of something smaller, or be a cog in something everyone’s heard of.

And then there’s Pete Campbell, who might be the show’s most underrated case study in seasons. Early Pete is unbearable. Entitled, petty, constantly engineering conflict with Don because he mistakes proximity to power for a claim on it. But somewhere around season five, the friction eases. Not because Pete suddenly becomes likable, but because the timing and alignment shift. He stops trying to take Don’s seat and starts building something adjacent to it. The doors open, the accounts land and he even opens an office in California for SCDP for a hot minute. And yet, the one thing Pete cannot get right across all seven seasons is, his marriage. He cheats compulsively, loses Trudy, and spends the final stretch slowly arriving at a conclusion most people reach much faster: the thing he kept leaving was the thing that actually fit. Seven years of chronic self-sabotage before the weather broke.

TV’s Most Underrated Gift Is the Distance It Gives You from Yourself

You can’t get distance from your own heartbreak while you’re still inside it. But you can watch someone else’s version of the same pattern and suddenly see what proximity always obscured. Not because the emotional investment isn’t real; But because caring about a character without having your actual future tied to their decisions creates just enough room to finally name the thing you’ve been circling in your own life.

That’s what makes certain shows hold up across your entire life, not just the period when you first watched them. You watch Don Draper at 18 and you see a cautionary tale about dishonesty. At 30, you start noticing the patterns you share with him, the ones you swore you didn’t. At 45, you’re probably watching the people around him more carefully than you’re watching him, because by then you’ve been Betty, or you’ve been Peggy, or you’ve been the person who showed up in someone’s life during the wrong season and didn’t understand why nothing took root.

The Compound Interest of Character

Television, when it’s built right, gives you a safer place for exactly that energy. You’re not banking your actual life on Don Draper. You’re not waiting for him to call. But you can still do the real work of sitting with ambiguity and arriving somewhere closer to understanding. And unlike doing it with real people, in real time, with actual stakes, you can pause, rewind, and watch someone else’s analysis of the same scene at 1.5x speed.

Here’s what gets lost in most conversations about TV development: what makes these characters resonate across years isn’t that they grow. Growth implies a clean upward line, and clean upward lines don’t mirror how people actually change.

Mad Men keeps putting the same characters into new situations and lets you watch how they react differently than they would have two seasons ago, or sometimes exactly the same way. Both outcomes teach you something.

Pete doesn’t become a better man. He becomes a man in a different context, and the context reveals parts of him that were always there. Joan doesn’t evolve past the system that undervalues her. She encounters a new version of that system and makes a different calculation. That’s not growth. That’s the accumulation of reactions across enough time that a pattern becomes legible.

You can’t get that accumulation in eight episodes that are released every two years. You need the long runway, the stretches where a character plateaus because real people plateau too. When a show runs long enough and the writing is honest enough, it stops feeling like a story someone constructed and starts feeling like a life you’ve been adjacent to.

That’s also where the economics become hard to ignore. The most valuable audience behavior in streaming is not the first watch, it’s everything after.

The binge generates supplemental content. That content becomes the discovery mechanism for the next wave of viewers, people who find the show through a YouTube breakdown or a friend’s insistence five, ten, twenty years later.

Those viewers (like myself) watch, create their own analysis, and the cycle compounds. Mad Men ended in 2015. I started it in late 2025 and the supplemental ecosystem was richer than most shows that premiered this year. That’s not nostalgia, it’s a flywheel, and it runs on characters built with enough contradiction that people keep asking why long after the finale.

A show that generates this kind of long-tail engagement is not just content, it’s infrastructure. Every new subscriber who discovers it represents viewing hours the platform didn’t spend a dollar acquiring.

The question every streamer should be asking is not whether a show will perform in its first four weeks, but whether it has the character architecture to still be pulling people in a decade from now.

What the Second Season Collapse Is Actually Measuring

Everything I just described took seven seasons to become legible. The current streaming model rarely gives a show that runway. But the deeper issue isn’t duration alone. It’s that shows are being built for premises, not for people.

Bloomberg recently reported that Netflix is struggling to retain viewers beyond season one. When you understand what that flywheel requires, Netflix’s current problem stops looking like a retention issue and starts looking like a construction one.

One Piece lost more than 30% of its audience for season two. Beef dropped over 70%. The Night Agent shed 50% for its second season and another 35% for the third. Avatar: The Last Airbender dropped more than 60%. These come straight from Netflix’s own first-four-weeks data.

Netflix is reportedly studying why this keeps happening. They renewed Running Point and The Four Seasons despite both losing more than half their audiences. They’re ending The Night Agent after its next season. The company went roughly four months without a major hit after Bridgerton in early 2026.

I have written before about how the industry keeps chasing stated preference while ignoring revealed preference. Audiences say they want something new. What they actually do is return to shows where they trust the characters, where the emotional infrastructure is already built.

Without productive irresolution, you don’t just lose season-two viewers. You lose the entire downstream engine: the supplemental content, the discovery flywheel, the subscribers who would have found the show three years from now through someone else’s obsession. Netflix’s total engagement grew less than 2% last year despite the finales of both Squid Game and Stranger Things. The pipeline isn’t slow. It’s producing shows that expire on contact.

The one show that bucked the trend? Bridgerton. Characters complicated enough to argue about, relationships messy enough to obsess over, a world rich enough to return to. That’s what duration and contradiction look like when they’re built in from day one.

What This Should Mean for the Future of TV Development

Working inside a major streamer, I rarely heard someone ask the question that actually matters: are these characters built with enough depth that someone is going to spend their lunch breaks for six months trying to understand them? The conversation organizes around genre, format, talent packages, and comparable titles. Not whether there’s enough irresolution baked into these people to sustain an ecosystem beyond the show itself.

But the data is making the case the writers’ room should have made first. The second-season problem is not a marketing failure. It’s a character-depth problem dressed up as a retention metric. The answer to why audiences won’t come back is not sitting in the data science team. It’s sitting in the room where someone decided these characters were built to serve a premise rather than outlive one.

The shows that become culture rather than content are rarely the ones that would have scored highest on a pre-production data model. They’re the ones where the writing understood that the most durable competitive advantage in this business is a character the audience can’t stop turning over. That’s not an accident of craft. It’s a design decision.

Mad Men proved that better than anything I’ve watched in a long time. Don Draper is deeply flawed, perpetually self-destructive, occasionally impossible to defend and somehow always worth understanding -- is the proof of concept I keep returning to.

We Built a Machine for Producing Meaning

The media industry has always had an unusual relationship with its own importance. Unlike most industries, it cannot pretend its product is neutral. We’re not here to sell detergent, we are selling stories: the mechanism by which cultures understand themselves, process grief, recognize their own absurdity, and occasionally get warned about where they are headed.

That is no small feat. For a long time, the people inside this industry knew that. There was genuine pride in being part of something that outlasted its original moment and the mission felt real.

Then we built the machine around it. And the machine was built around the incentive of optimizing content consumption over quality content creation.

There is an art and a science to identifying, producing, and promoting great content. The science is what this industry has spent the last decade perfecting: dashboards, completion rates, 72-hour viewership windows, quarterly guidance.

The art is what it has quietly let go in the process.

The hit HBO series Succession spent four seasons inside the fictional media empire Waystar Royco, following the aging patriarch Logan Roy and his children as they battle for control of the company and its vast influence. Along the way, the show watches a family slowly lose the thread of why any of it mattered. Not through scandal or sudden collapse, but through something quieter and more insidious: the replacement of purpose with positioning.

Nobody in the Roy family wakes up one morning and decides to stop caring about what Waystar is actually for. They simply get deep enough inside the deal logic, the internal maneuvering, the performance of vision, that the original question stops feeling worth asking.

What remains is ego disguised as strategy, perfumed by an unresolved longing for their father's validation, with power serving as the only compass still pointing anywhere.

Below is a clip of Kendall (the oldest son) ousting his father to take over Waystar-Royco at the end of season two.

That dynamic is not just compelling television. It is a remarkably precise portrait of how legacy media institutions actually function. The company becomes secondary to the contest happening inside it and the product becomes a pretext. What looks like a fight over a business is really a fight over something far older and harder to name, about validation, inheritance, and the particular hunger that power never quite satisfies.

Below is a clip of the Roy children realizing their mother didn’t protect them from their father or consider their best interests for inheritance and legacy at the end of season three.

This summer, as Paramount absorbs Warner Bros. Discovery (WBD) and Fox folding Roku into its portfolio, both deals arriving wrapped in the familiar language of synergy and scale, I found myself rewatching Succession. It felt almost involuntary. WBD is now facing its third consolidation in a decade. The first, with AT&T, played out like a corporate version of My Big Fat Greek Wedding: overwhelming, absurd, but survivable. Now, it’s less like watching Succession as satire and more like receiving it as an orientation video. The genre has shifted to Squid Game with better tailoring, and nobody is safe.

The Pitch That Stopped Being a Joke

There is a scene in Succession Season 4 where one of the Roy children, Kendall (far right), pitches a new venture The Hundred as ‘Substack meets Masterclass meets The Economist meets The New Yorker.’

As described verbatim: “The global media startup is a digital hub delivering all the essential information to navigate the now. The world’s leading experts provide humanity’s most invaluable knowledge in bespoke, bite-sized parcels, designed to improve the lives of subscribers and the world in general. The antidote to the modern malaise of empty-caloried input-overload.”1

When it aired three years ago, hearing this word-salad of corporate jargon made me laugh. Now AI has made the Kendall pitch table stakes. Any team can spin up a proof of concept in an afternoon, which means any team can perform vision without actually having any. The language has shifted, however the emptiness underneath it has not.

Attach AI to an initiative and suddenly there is perceived momentum, a narrative for the board, a reason to delay harder questions. Never mind that many of these programs, when examined against actual cost structures, are more expensive to run than the human workflows they replaced.2

The market has made clear it will reward the story over the substance.

The proposed acquisition by Paramount-Skydance to acquire Warner Bros. Discovery has attracted the AI gloss that every major media transaction requires in 2026: AI-powered content pipelines, algorithmic audience development, synthetic production efficiencies. The details are ambient. The intention is atmospheric, to signal that this is not a legacy-media distress sale but a forward-looking technological transformation.3

The underlying reality is that Paramount's linear business is in structural decline, its streaming unit has never turned a meaningful profit, and its library valuation depends on licensing assumptions that may no longer hold. But that reality recedes behind a story about optimizing underutilized assets and the promise of what the combined entity might eventually become.

This season the market is overvaluing potential and underselling reality

SpaceX recently completed the largest IPO in history, raising $75 billion at a valuation of $1.77 trillion, roughly 94 times its annual sales. The company lost almost $5 billion last year, merged with xAI in early 2026, which also lost billions in 2025, and the market opened the stock up anyway. The vision is genuinely enormous: Starlink, the only profitable division, xAI and its Grok platform, and an orbital computing ambition that involves launching data centers into space to solve the power constraints choking AI infrastructure on the ground. U.S. electricity supply in 2025 was only 49 gigawatts against an estimated demand of 62. It sounds like science fiction pitched as a prospectus. None of the execution is small or close. But the market is not pricing the execution, its pricing is the narrative.4

The narrative is the through line.

The incentive in these deals is never the health of the underlying company. Semafor reported that most individual investors will not get in at the SpaceX offering price, while the ultra-rich are expected to comprise a large part of the retail allocation.5 The Paramount transaction has its own version of this geometry: the controlling shareholder structure, the legal settlements that preceded approval, each a layer of insulation for the people already inside.

Succession dramatized exactly this. Lukas Matsson (orange turtleneck in the image above), the tech billionaire acquiring Waystar, had inflated his subscriber numbers by tens of millions of fake accounts. A small circle of insiders knew the stated valuation was fiction. The deal moved forward regardless, because the narrative served too many people’s short-term interests for anyone to stop it. The Waystar board was not fooled. It was willing. That is a more honest description of how the Paramount deal, and most consolidation in this industry, actually works. Nobody at the table is confused about Paramount’s trajectory. The ratings decline is public. The debt load is public. The AI language in the deal materials is not a lie exactly. It is a courtesy, a shared vocabulary for not saying what everyone knows.

The incentive is never the health of the company. It is the momentum of the transaction itself, and the wealth it generates for the people who close it.

The Numbness Is Not Just a Leadership Problem

Media mergers are sold the way chemotherapy is sold: necessary, painful, and ultimately worth it. The patient will emerge leaner, stronger, better positioned to survive. What the announcement never covers is the eighteen to twenty-four months of nausea in between, or the fact that the treatment cannot always distinguish between what is sick and what is healthy. The side effects, existentialism, toxic territorialism, a deep and spreading numbness, are not bugs in the process. They are the process, distributed downward to the people who had no vote on whether to begin it.

Succession shows us the Roys. It shows us the boardroom, the deals, the knives out at every family dinner. What it never quite shows us is the several thousand people working at ATN and Waystar who have no seat at that table and no say in what happens to them. The chaos at the top is visible because it is dramatic. The quiet unraveling happening everywhere else in the building is harder to film.

Most criticism of the media industry aims upward. Executives walked away with enormous payouts as the creative slate hollowed out and thousands of employees lost their jobs. That is all true and it matters. But the more interesting and less discussed problem is:

What does that treatment do to the rest of the organism over time?

The Side-Effects of Corporate Chemo

The detachment at the top does not stay at the top, it distributes. It seeps into every layer of the org chart until entire companies are full of people performing engagement without quite feeling it. Inside a media company, that gets called being professional.

A more honest word is anesthetized.

Re-orgs are happening like furniture rearranged in a house where nobody has decided yet whether to sell. However, psychological safety is not a soft metric. It is the condition under which people take risks, propose things, build something beyond the minimum required to keep their seat. When that safety is gone, and in a prolonged merger it goes fast, productivity does not collapse dramatically. It hollows out quietly. The company does not fail, it just stops reaching.

Part of why that numbness is so hard to fight is structural, not personal. Most people working inside these companies have no legible line between what they do every day and what it produces on a balance sheet. Streaming broke the clean P&L logic that cable relied on and never replaced it with something coherent. The content team that green-lights a show, the product team that surfaces it, the algorithm that saves a churning subscriber: all of them contributed. None of them can prove it, because the data lives in incompatible systems with incompatible definitions of success.

That is not apathy. That is people responding sensibly to a system that has made accountability structurally impossible.

For more information about intricacies of why P&L attribution is difficult to measure inside legacy media corporations, see below!

Are my colleagues and I really that different from the Roy Children after all?

In the image above, Shiv (one of the Roy children) is vouching for Tom (her husband that was a pushover for most of the show) to Matsson as someone who won’t actually run the show is that dynamic made explicit.

The Roy children spend four seasons chasing something that keeps moving: their father’s approval, the CEO seat, some version of finally being enough. What the chase required is almost the inverse of what they wanted — every compromise, every betrayal, every time one of them stayed in the room by making themselves useful to someone who didn’t respect them.

The rest of us are running the same loop at a smaller scale. Chasing a promotion, job security, some signal that the work actually matters to the organization we do it for. When the deal closes and surviving leadership gives the all-hands about exciting new chapters, the triumph belongs to the small group who negotiated it and received the payout. Everyone else is just figuring out if they still have a job. The Roy children never got what they were really after either.

@themillionaireparadigmTom didn’t become CEO because he was the strongest. He became CEO because he was the most useful. While everyone else fought for legacy, pride, and identity… Tom just said, “What do you need me to be?” And that’s the tragedy: In the end, the crown didn’t go to the one with vision or birthright. It went to the one willing to bend the lowest to sit the highest. #mindset #inspiration #motivation #millionaire #tvshow

Tiktok failed to load.

Enable 3rd party cookies or use another browser

The tragedy of the Roy children is not that they are villains. It is that they absorbed the values of the institution so completely that they cannot locate themselves outside of it. They have no idea what they would want if Waystar were not the frame. That is what the show is really about, not the money, but the loss of a self that exists independent of the machine. People do not need to be a Roy to pay that tax. They just need to work somewhere long enough for the side effects to become the climate.

Settling Is a Choice. So Is Not Settling.

Every writer has heard some version of “don’t quit your day job.” Most of us didn’t. That is not weakness. That is how the lights stay on while figuring out what you actually want to say.

There is real value in working inside an institution, even one that sometimes feels like slow erosion. I can vouch for that directly -- if I did not work for a recognizable name in this industry, the insights I am sharing in this newsletter would carry less weight. That is just true, and I am not going to pretend otherwise. A W-2 from the right company opens doors that would otherwise stay closed. It lends credibility that takes years to build independently, if it can be built at all.

But that is exactly the trade. Leadership has a direction. Conformity is expected. Conformity wins -- and in a functioning organization, that is actually how it should work. Businesses run on alignment, not dissent. The existential question most people inside these corporations are quietly negotiating is how much of themselves they are selling to access those doors, and whether the exchange still feels worth it over time.

Part of why I started writing was to resist that logic without pretending I am above it. I am not outside this system. I am not performing rebellion from the safety of a salaried position -- or at least, I am trying not to. What I am trying to do is keep the habit of asking what the point is, even when the environment has stopped rewarding the question. Because the question is the point.

And in media specifically, the question gets harder to ask. These companies are built on two fundamentally opposed power centers: the content side, which cares about the content, and the corporate side, which cares about the dollars. Those two things have never fully been in sync. What is different now is that the gap has become structural. When two companies with this already in place merge, they are not combining strengths. They are doubling the dysfunction.

We turned our blind spots into an award winning TV series, then followed the script IRL.

The show was never really about the Roys. It was about what happens when the people who understand the mission cannot hold power, and the people who hold power do not understand the mission.

But watch it again: the children were never really fighting for the company. They were fighting for their father’s approval. The CEO title was just the only currency Logan ever accepted. That’s what makes it tragic. And that’s what makes it true.

Anyone at any level of one of these organizations faces a version of this right now. The territorial behavior when budgets compress, the reluctance to collaborate when headcount feels like a threat, the obsession with protecting a title or a fiefdom…these aren’t really about the role. They’re about the same thing Kendall and Siobhan and Roman were chasing. A sense of worth the structure keeps promising and rarely delivers.

The uncertainty makes this worse. When there’s no clear path to promotion, no guarantee that loyalty will be rewarded, the psychological default is fight-or-flight. Collaboration starts to feel like redundancy. Cooperation looks like volunteering to be cut.

The question worth asking for the Roys and for anyone in a restructuring org right now is: what you’re actually chasing?

Is it the title? Or is it what the title was supposed to represent? One of those is still available even when the org chart isn’t cooperating.

The industry produced its own warning. It streamed it, won awards for it, wrote think pieces about what it meant. The question was never whether we understood the lesson. The question is whether we are still willing to act like it applies to us.

It went on to collect over 15k likes, which, for a writer who normally produces clinical breakdowns of why media economics are broken, is not the usual response. When a post performs exponentially beyond your baseline, it is impossible to ignore. So I had to ask: why.

I sat with the comments. And what I found underneath the nostalgia was not really nostalgia at all. It was relief.

The Shuffle gave you almost no choices, and that was the point

Think about what an iPod Shuffle actually let you do. You chose which songs went on it, once, in advance. After that, the device offered you three decisions:

1. play or pause.

2. the song before or the song after

3. the volume up or down.

That was the entire interface. No library to scroll, endless recommendations, or push notification quietly reminding you that you could be doing something better with your attention.

Call it the “Magic 3”. Three inputs, and the experience resolves itself.

Now hold that against how we consume everything today. Streaming and social media handed us a near-infinite menu and called it freedom. But this freedom comes with a tax, and the tax is paid in decisions. The reason “I can’t find anything to watch” became a cliché while we literally hold the entire history of television in our hands is not a paradox. It is the predictable result of asking a human being to make an unbounded number of choices and calling that a feature.

It is easier to order from a restaurant with three things on the menu than thirty. We know this in our bones. We just built an entire entertainment economy that pretends otherwise.

I build the machine that took the relief away

Here is where I should be honest about which side of the table I sit on, because it changes what this essay is. I build first-party data products for a major streamer. I can slice and dice an audience until it confesses.

So when I tell you the abundance is the problem, I am not romanticizing a simpler past from the cheap seats. I am describing a thing I help construct for a living.

I have written several substack essays now arguing that media is broken, and the argument always lands in the same place: there is a fault line running through this industry between the people who make the work and the people who hold the data, the spreadsheets, and the infrastructure, and almost nobody is fluent in both languages at once.

Because the consumer version of that same failure is exactly what those ten thousand people were responding to. Having every signal available and no discipline about it is like keeping a commercial stove in a studio apartment, or dropping a Ferrari engine into a tow truck. It’s impressive on paper but wrong for the job.

The mismatch shows up on the business side as content engineered to within an inch of its life, too many plot-lines, too little consistency, every variable optimized and nothing left to simply return to. And it shows up on your couch as a thousand thumbnails and the slow, low hum of not being able to decide.

It’s the same broken pipe. The studio cannot tell which show is worth keeping. You cannot tell which show is worth starting. We mistook more for better at every layer, and then we handed the cost of that confusion to the person holding the remote.

From a people problem to a media problem

So let me put on the hat I actually wear and apply this to the thing I write about.

The translation is simpler than it sounds. The exhaustion is not a content problem. It is a translation problem, and it is the same one I keep circling from the supply side: the distance between what a person is feeling and the title that would meet it.

Nobody types “existential dread” into a search bar at 11pm on a Tuesday when they’re trying to find something to watch. The vocabulary of that kind of need is not clinical, it is ambient. So the platform that wins is not the one that offers the most. It is the one that closes the gap between a feeling and a title without ever making you name the feeling out loud.

Which means the first question in the green-lighting room is wrong. We start by asking what genre a show belongs to. The right first cut is not genre. It is our good friend Abraham Maslow.

Before you ask what category a show belongs to, it’s more strategic to ask what need it serves.

• Do I want to feel better?

• Do I want to escape?

• Do I want to look inward, or have something help me sit with a question I cannot answer alone?

• Do I want to feel like I am keeping up, part of the conversation?

Those are the real buckets. The genre is just the wrapper they arrive in.

This is why the classics endure in a way the optimized stuff does not. Friends, The Office, Cheers, or Parks and Rec. None of them were narratively complicated. Their power was never the plot. It was habituality, the consistency of showing up to the same emotional address every week. I watched The Office every night on a trip through Europe, because Netflix holds the international license there. It was not the writing that pulled me back. It was the familiarity. Familiarity beats almost everything, which is the same reason a twenty-year-old medical drama keeps quietly winning the year while the expensive new bets bloom for a weekend and disappear.

What handbags and TV Shows have in Common

Once you start from the human emotional need instead of genre quotas for a content portfolio, an old confusion clears up. We stop grading two different things on the same curve.

Think about it like this:

A Longchamp and a Louis Vuitton are both bags.

Both bags carry your things. That’s where the similarity ends.

A Louis Vuitton is understood by everyone involved to be more than a bag. The object is almost incidental. Longchamp is a good bag, but you’re reaching for it to solve a practical problem. They meet different emotional needs. The signal requires the contrast. Put them on the same sales report and wonder why one “underperformed” and you’ve already asked the wrong question.

Television works exactly the same way:

Silicon Valley was built for a specific person to feel seen, and when it lands, it lands deep.

The Office is not trying to do that. It is trying to give you somewhere to belong, a recurring group of people you return to out of habit and affection, a standing invitation back every week.

Both are exceptional shows with different emotional value propositions. They were never running the same race to start.

The illusion of optionality is doing something to us and it’s not serving anyone

Streaming collapsed the shelf. Prestige drama next to reality filler, niche comedy next to broad procedural, everything under the same lighting with the same thumbnail format. In the race to fill libraries, brand and value proposition got sacrificed for volume until every platform looked the same and every show competed on the same terms. When you stop distinguishing what something is for and who it’s for, you do not expand choice. You dissolve it.

There is a cost beyond inconvenience.

When we are handed this much input and not deliberate about it, the algorithm stops being a tool and becomes the thing we mistake for our own taste. What we clicked on twice gets fed back to us until it feels like a preference we chose rather than one that was shaped. The dopamine is real and the design is good. That is exactly why the loop is so easy to fall into.

To enter that flood without being swept by it, you almost need a thesis going in. A sense of what you are actually looking for before you open the app. Most people do not have the energy for that every night, which is the whole point. The illusion of optionality quietly outsources the deciding to a system whose interests are not yours.

The Magic 3, for television

So what would relief actually look like here? I can verify this, it’s not the algorithm trying to read your mind. I have argued elsewhere that trying to predict what you want is mostly a way of generating false data. The goal is not to read your mind. The goal is to stop overwhelming you.

The Shuffle’s genius was collapsing the choice to three. Television has its own Magic 3 hiding in plain sight. When someone decides what to watch, the decision really comes down to a few things:

• What emotional need does it satiate for me?

• What is it about?

• What does it cost me in time?

And, at the edges, where can I watch it and who is in it?

Strip the experience back to those, layered on top of the emotional need underneath, and you have given someone a way to choose without drowning them. That is the design question worth obsessing over. Not “how do we offer more,” but “what are the three things that actually help a person decide, and how do we get out of their way after that.”

The Truth: Why did the ad really hit so hard?

I think the Shuffle ad went viral for the same reason people defend a slower, simpler era of anything. We have spent a decade over-engineering the experience of being a person, and there is a movement, quiet but real, back toward basics. You can already see the early edges of it: people abandoning inboxes, swearing off the feed, opting out of the optimized everything. With AI now accelerating all of it, that instinct is only going to sharpen.

So the ad was not really about a discontinued music player. It was a cry, the same one I keep hearing underneath all of this. We do not want to lose our sense of humanity to the machinery of infinite choice.

Audiences are not nostalgic, they are exhausted. What they are reaching for is not the past. It is the relief of fewer, better decisions, and the iPod Shuffle is simply the cleanest artifact of that relief we ever built.

The question for anyone designing what comes next, in media or anywhere else, is whether we are willing to build for that need on purpose, or whether we will keep mistaking abundance for a gift.

Letterboxd has been put up for sale. And nevertheless, a coalition called Intrinsic Entertainment Collaborative has been organizing a community ownership bid, proposing a model where small donors, filmmakers, fans, and mission-driven investors pool resources and elect a Board of Governors to steward the platform alongside its existing team.

The idea is straightforward: a platform built by film lovers, for film lovers, should not end up inside another corporate portfolio where its value gets extracted and its users become a product line.

If the community ownership bid succeeds, it would be genuinely unusual in media. But the more interesting question is not whether it happens, it’s what creators do if it does.

Because Letterboxd already holds something most filmmakers have never had access to: honest, unsponsored, human-generated data about what their work actually means to the people who watched it.

I work as a product manager building first-party data products in streaming, which means I spend my days working with information the public never sees. That invisibility cuts both ways.

Studios that license their IP to platforms like Netflix or HBO Max are often operating in the dark about how their titles actually perform once they leave the lot. Streamers may send back high-level analytics on content performance, but receiving data and acting on it are two very different things.

Letterboxd, by contrast, is sitting on a library of organic signals that no streamer controls. Third-party services available to employers like mine scrape the internet to gauge content engagement, but Letterboxd’s edge is concentration: superfans and film connoisseurs producing user-generated lists, reviews, and ratings in one place. The audience is self-selecting, which sharpens the signal rather than diluting it.

Below I’ve used The Devil Wears Prada as an example to explain how Letterboxd can provide valuable insights to creators. While this is all theoretical and making this information open source is realistically unlikely, it’s an exercise worth exploring because it provides creators the opportunity to see how user generated content and engagement is the one asset that will never be replaced by automation.

Subscribe now

What the Devil Wears Prada Actually Tells You

The 2006 film has a full profile on Letterboxd: watch availability, cast and creative metadata, press mentions, popular reviews, related films, and the lists where it appears.

None of that is new information. What makes it interesting is the infrastructure around it. When a user adds a film to a list, saves it, likes it, or writes a review, that behavior is a data point. Not a vanity metric, but a signal about the shape of the audience.

• Which lists include The Devil Wears Prada, and how engaged are the followers of those lists?

• What percentage of people who encounter it through a curated list actually follow through to watch it on the platform where it is currently licensed?

• What is the conversion rate from discovery to platform acquisition?

Letterboxd can, in theory, track exactly that. And that number, the actual downstream conversion a title drives to a streaming platform, is the metric that should be restructuring how creators are compensated.

Right now, residuals are calculated on models that predate streaming entirely. They do not reflect the reality that a film like The Devil Wears Prada is still actively doing discovery work almost two decades after release. Every time someone adds it to a “Chick Flicks” list that gets shared and followed and interacted with, that is essentially free marketing for whoever currently holds the license.

The creator sees none of it. The streamer benefits invisibly.

The Engagement Metrics That Actually Matter

I want to be specific about what a creator should want from a platform like Letterboxd, because not all engagement is created equal.

“Likes” on a film page or list are a weak signal. They reflect sentiment, but sentiment does not pay residuals and does not tell you much about what drove the watch. Total list appearances are more interesting, but only as a starting point.

Growth rate matters more than raw totals. A niche film that appears on 50 highly engaged lists and drives a measurable conversion spike to its host platform is telling a more valuable story than a blockbuster that sits on 5,000 lists nobody actively uses.

The signal is in the velocity, not the volume.

What that means practically: if you are a filmmaker, showrunner, or studio executive sitting across a licensing negotiation from a streamer, Letterboxd data could be the closest thing to an independent audit of your title’s discovery value that currently exists anywhere in the ecosystem. Not because Letterboxd is perfect or comprehensive, its 26 million users represent a small fraction of the total viewing public, but because those 26 million are disproportionately influential. They are the people whose lists get followed, whose recommendations get trusted, whose engagement actually moves something.

Below I’ve outlined a diagram to visualize the different avenues Letterboxd can provide to drive conversions based on a title.

The “Similar Films” Section Is VERY Underrated

One piece of the Letterboxd interface that gets overlooked is the “similar films,” “themes,” and, “nanongenres” categorization. This section is underrated because it helps provide signal on a few fronts.

From a content discovery lens (aside from user sentiment), what is the key ingredient that convinced a user to watch? While there’s user intent, such as “liking” a film or “saving it” to a list, the action to convert to the streamer that’s licensed the film is the key driving factor, and while it’s almost never, “one” factor, it’s a healthy indicator.

What made the viewer decide to watch?

• Was it a particular actor?

• Was it a particular theme?

• Was it a specific genre?

Below I’ve broken down the insights that can be gathered from each section:

1. All Similar Films

Titles that appear under “similar films” are populated by a propensity score which is calculated based on how many common “tags” a title shares with The Devil Wears Prada.

The Insight: In an ideal data environment, you would see not just which films appeared in that rail, but which specific tags they share with the title and how strongly each one weighted the score. Actors, themes, genres, tone: the full taxonomy the algorithm used to draw the comparison.

Three distinct insights emerge from looking at this layer:

For positioning: Tag-level visibility tells you how a film is being placed among its peers, not by the creative team’s intention, but by the cold logic of the platform itself. Filmmakers carry a precise vision of how their work should be perceived. The algorithm and the audience carry their own. Where those three readings converge is validation. Where they diverge is information worth having.

For conversion: Tag data opens the door to a more structural question: is there a specific attribute, a shared actor, a recurring theme, a tonal register, that consistently correlates with a viewer actually choosing to watch? If one tag keeps surfacing across the titles that get clicked, that pattern is not noise. That is signal, and signal at this level is rare enough to be genuinely valuable.

For the broader picture: Together, these two lenses reframe what a content rail actually is. Not just a row of thumbnails generated by an invisible hand, but a map of how a title exists in relation to everything around it. Understanding that map, who drew it, by what logic, and what it is quietly telling the audience about your film, is its own form of competitive intelligence.

2. Themes

Below you’ll see how the “tags” are categorized by themes so a viewer can view other films that share the same themes as The Devil Wears Prada.

The Insight: Themes operate at a different register than genre. Where genre describes the container, theme describes what is actually inside it: the overarching tone, the emotional core, the idea the story keeps returning to. A genre label tells you a film is a romantic comedy.

A theme tells you it is about the specific exhaustion of wanting something you are not sure you deserve.

Three distinct insights emerge from looking at this layer:

For creatives: Themes reveal the gap between intention and reception. A filmmaker has a vision for how their work should land. The algorithm and the audience have their own. Seeing which themes surface in the films placed alongside yours is a way of reading that gap honestly, understanding not just how the film was made, but how it settled in the minds of the people who watched it.

For buyers and sellers: Themes reframe a single title as part of a portfolio of reception. If a particular emotional core keeps appearing across titles that audiences are gravitating toward, that is not coincidence. That is a market signal. Take “laugh-out-loud entanglement” as an example. On the surface it lives inside the romantic comedy bucket, but it says something more precise: the audience wants to laugh, and they want it to feel genuinely romantic, not just categorically so. That granularity points toward an appetite in the market that shows up in behavior before anyone has named it.

For the broader picture: Themes are ultimately the most useful lens for packaging, the layer at which a film communicates its identity to the world. They answer the question not just of what a title is, but of what it means to the people who chose it.

The distinction between themes and nanogenres is ultimately a distinction between packaging and anatomy. Themes tell you how a film faces the world. Nanogenres tell you what it is made of.

3. Nanogenres

Then a viewer can view a section called “nanogenres” which categorizes films at a level deeper by individual traits of the film. Nanogenres are where the tags gets surgical.

The Insight: Nanogenres are where the taxonomy gets surgical. Hyper-specific micro-categories built around exactly what appears on screen or how a viewer emotionally responds, they sit a full layer beneath themes in specificity. A theme tells you a story is built around wit and romantic tension. A nanogenre tells you there is fashion in the frame, that the chemistry reads as intelligent, that the humor never tips into cheap.

Three distinct insights emerge from looking at this layer:

For the viewer: The appetite is real, even if the language is not. Nobody opens a streaming app consciously searching for “intelligent chemistry with fashion.” But that combination keeps pulling people back anyway, expressed through behavior rather than words.

For creatives: Nanogenres function less as a blueprint and more as a mirror. You would not engineer a film around a checklist of attributes. But in the aftermath of analysis, patterns surface that even the creator may not have consciously intended, specific textures that kept resonating, combinations that gave the work genuine pull. Think of it as a scene-level audit of your own writing.

For buyers and sellers: Where themes help position a film within the broader market, nanogenres add precision to the question of what a title actually is and who it is actually for. That granularity matters when licensing decisions hinge on more than a genre label.

The distinction between themes and nanogenres is ultimately a distinction between packaging and anatomy. Themes tell you how a film faces the world. Nanogenres tell you what it is made of.

Subscribe now

The Bigger Argument at Play

The data has always existed. The question is who it belonged to.

Streamers have known for years which titles drive subscriptions, which ones keep households from cancelling, and, which discovery pathways convert casual browsers into committed viewers. That intelligence stays inside the platform.

Creators, the people who made the thing that is quietly doing all that work, have operated largely on instinct, reputation, or whatever their representation could negotiate in the dark.

Letterboxd changes the geometry of that problem, modestly but meaningfully.

It is not a perfect dataset. Its 26 million users skew devoted and discerning, which means the signal is sharpened rather than representative. But that selectivity is the point. What Letterboxd captures is not passive consumption. It is active curation, a community of people building lists, following recommendations, and converting to platforms to actually watch something. That chain of behavior, from discovery to intent to action, is the closest thing to an independent audit of a title’s discovery value that currently exists outside a streamer’s walled garden.

Each data point Letterboxd offers is a different answer to the same question: what does this film actually mean to the people who chose it?

That question is not just creative. It is economic.

Creatives who learn to read audience data as both creative signal and contractual leverage will negotiate from a position that those who ignore it simply cannot. That advantage, however, compounds most when it moves beyond individual deals. If guilds and unions were to build formal benchmarks around engagement metrics, conversion velocity, list growth, downstream traffic to licensed platforms, the residuals conversation gains something it has lacked since streaming made the old models obsolete: evidence.

An organic funnel is still a funnel. A film that is twenty years old and still actively pulling new viewers toward a platform it has no formal relationship with is still doing work. That work has value.

The industry has just never had the infrastructure, or the incentive, to account for it honestly.

That infrastructure may be closer than it appears. The community already knows what it cares about. The data is already there. The gap that remains is not technical. It is a question of will.

Subscribe now

There is a version of this story where streaming saved Hollywood. Lower barriers, global distribution, a direct pipeline from creator to audience. The early pitch was genuinely exciting.

That version did not account for what streaming would do with the data.

In my day job, I have the privilege of building first party data products for a major streamer.

But here’s the uncomfortable truth: every major streamer treats first party data like trail mix, handing the studios the nuts while quietly keeping the M&Ms for themselves. And we all know that’s not a fair trade.

After developing an unwilling expertise in patience, courtesy of Looker loading times that make dial-up look like a competitive product, a theory started forming in the space where productivity used to be:

Hollywood is not in a recession because of a creativity problem.

It is in a recession because the people making content cannot see the numbers that would tell them whether their work is actually building anything. And the platforms sitting on those numbers have every incentive to keep it that way.

This is the piece where I want to trace how we got here, what data is actually available, what is not, and what it costs everyone below the platform level when the answer is mostly the latter by using Seinfeld as the poster child.

The Seinfeld Standard: What We Lost When the Language Broke

Picture a Thursday night in November 1992. It is 9:00 PM on the East Coast, 6:00 on the West. Across the country, somewhere close to 29 million people are settling onto couches, balancing dinner plates on knees, reaching for the remote to turn up the volume because the opening bass line has just hit. “The Contest” is about to air.

In living rooms from Manhattan to Sacramento, 29 million people were about to watch the same episode at the same moment, and these civilians were about to be asked the same quietly titillating question about the body and its loyalties.

Seinfeld had always known that the truest comedy lives in what we do not say out loud. “The Contest” simply made the unsayable into a national referendum, and by Friday morning the verdict was in. The whole country was contemplating the challenge of restraining from humanity’s most sensational pleasures.

That kind of shared national comradery is not something an algorithm can manufacture.

The Morning After: What 29 Million Meant Before Anyone Had Finished Their Coffee

The thing about a Thursday night number is that it did not stay locked in some data center or spreadsheet. By Friday morning it had already traveled.

Consider the chain of people who needed that number to do their jobs:

• The network (NBC) needed it to price inventory.

• The studio (CastleRock) needed it to value a library asset.

• The advertisers needed it to justify a media plan to a client.

• The showrunner (Larry David) needed it to negotiate a back end.

• The agent (Rob Prinz) needed it to push for a raise in the next cycle.

• The syndication buyer in a mid-sized market needed it to decide what to pay for the strip.

• The international licensor needed it to pitch Channel 4 in the UK on a show that had already proven itself somewhere else.

A writer negotiating residuals and a network negotiating ad rates could sit on opposite sides of a table, pulling that 29 million in opposite directions, and still leave the room having had a coherent conversation. The number was the floor they both stood on while they fought. It was the reason a fight could happen at all, rather than two parties shouting past each other about whose proprietary dashboard was more accurate.

They were all in their separate corners of the business looking at the same number.

That is the part worth pausing on.

Twenty-nine million was a fact on the ground. The kind you could put on a deal memo, hand across a conference table, and watch the other side nod. Not because they loved the number. Because they recognized it.

This is what scaled Hollywood.

Not the number itself, but the collective agreement to treat it as real. Not production efficiency, not distribution reach, not even the primetime schedule. Just the existence of a single, flawed, universally accepted metric that every participant in the ecosystem could point to and say: this is what happened last night, and therefore this is what my work is worth this morning.

Measuring Seinfeld in Two Centuries

Now imagine trying to measure Seinfeld today.

Not the Seinfeld of 1998, pulling 29 million viewers on a Thursday night. The Seinfeld of now, quietly archived in a Netflix library, the same show that reportedly commanded $500 million for exclusive global streaming rights for a five year deal starting in 2021.

Same show. Completely different measurement problem in a streaming world.

Thirty years ago, a studio could answer almost any question about Seinfeld by picking up the phone.

• Who watched last night? Nielsen had it, imperfectly but legibly.

• What time? Eight o’clock Thursday, the only time it aired.

• What ran before and after? Mad About You, then Suddenly Susan.

• What ads ran in the breaks? The network knew. The advertiser knew.

That lineup alone told you something real about the household on the other end of the signal. The time slot, the lead-in, the ad load, the simple fact that someone had chosen to be in front of a television at that specific hour.

All of it signal. All of it shared.

On Netflix, Seinfeld exists as a permanent artifact in a room with no windows.

Somewhere in a cluster of spreadsheets and analytics tooling, there is a granular picture of who is watching it, at what hour, on what device, after what other show, for how many minutes before they stopped. That picture exists. It is just not shared. Not with the studio. Not with the creators.

Not with anyone trying to decide whether to green-light the next show about nothing.

This is the quiet restructuring that happened beneath the headline numbers. For that $500 million licensing fee, Netflix acquires something worth considerably more than streaming rights:

What Streamers Know and Neglect to Share

Streamers acquire the exclusive privilege of knowing what the show actually does once it is on the shelf. The rights holder receives a wire transfer and an invitation to guess.

This is what basically happens:

Think of it as leasing a storefront on Fifth Avenue while being contractually barred from seeing the foot traffic reports, the sales data, or even which products are displayed next to yours. You would accept the check, cash it, and spend the next five years wondering whether you had underpriced the lease. The tenant, meanwhile, has learned exactly what your asset is worth and is using that knowledge to negotiate the next one.

The data is real. The data is proprietary. And that distinction is the entire game.

There is a screenshot I compiled together of all the ways Seinfeld is impressed on my Netflix account that I keep returning to.

It shows Seinfeld appearing across five distinct curations on a single Netflix profile: two personalized rails, Today’s Top Picks for You and Gems for You, and three content affinity rails, Critically Acclaimed TV Comedies, 30-Minute Laughs, and Award-Winning Witty TV Shows.

On the surface, it reads as a small victory for the title. Broad visibility, multiple algorithmic entry points, and a comfortable presence across several content categories. The show is doing the work it was licensed to do.

But in reality, that screenshot is actually a map.

The five rails it shows are not just placement. They are a record of algorithmic decisions, metadata judgments, and behavioral signals that Netflix has been quietly accumulating and acting on.

Every one of those curations tells a story:

1. Who is watching

2. How they got there

3. What the show means to the platform’s broader content ecosystem

Minimal information from that story is shared with the people who actually own the IP.

Our friends Jerry Seinfeld and Larry David probably do not lose sleep wondering whether more viewers found the show under the rail “30 Minute Laughs” or “Gems for You.” But Sony Pictures Television and Warner Media, which holds the Castle Rock catalog, have every reason to care.

To understand what is actually being withheld, it helps to break the data architecture into three distinct layers. Each one generates a different category of insight which is largely invisible to the licensor.

Layer One: How Audiences Find the Title Before They Even Open the App

This is where third-party data earns its keep, and where the customer journey begins long before the platform has any say in the matter.

Before a viewer lands on a streaming platform, a discovery journey has already started. For a show like Seinfeld, some portion of that journey is entirely external: people searching by name, encountering a clip on TikTok, following a cultural resurgence triggered by a meme or a podcast reference. Search volume trends, social listening, SEO traction, these signals are publicly accessible and can tell you whether a title is building organic momentum outside the platform walls.

Partners like Parrot Analytics, Luminate, and Nielsen can at least gauge that general appetite. They are imperfect instruments, but they measure something real: intent, before it converts.

External signals only measure intent, though. They do not tell you what happens when the viewer actually arrives. The full customer journey, from search to click to play, is something Netflix sees in its entirety and shares in fragments, if at all. For a studio trying to understand whether their IP is being rediscovered by a new generation or simply rewatched by its original audience, that journey data is not a nice-to-have.

It is foundational.

Layer Two: How the Platform Has Already Decided What Your Show Means

Before any human viewer opens the app, the algorithm has already made curatorial judgments about your title. You just cannot see them.

Every piece of content inside a streaming library gets tagged, categorized, and positioned relative to other titles. These are not neutral administrative decisions. The metadata schema, the clustering logic, which titles share a neighborhood and why: these are interpretive judgments that shape how a show’s identity is perceived long before any viewer interaction occurs.

See the image below of how each title is tagged and categorized under “genre” and “this show is.”

Look at what Netflix has done with Seinfeld. It appears across three distinct affinity rows alongside Suits, Gilmore Girls, and Sex and the City.

On the surface, a peculiar constellation: a legal drama built on fraud, a mother-daughter story in a Connecticut town, a Manhattan meditation on love and ambition, and a show that famously declared itself about nothing. Pull up their individual tags and the organizing logic snaps into focus. Witty is the only descriptor all four titles share.

Not comedy. Not prestige. Witty.

That is not a genre. It is a tonal promise.

However, not all of those judgments work the same way. There is a meaningful difference between a title that surfaces because it is popular and one the algorithm has decided belongs next to you specifically.

Layer Three: Why Personalization and Popularity Tell Completely Different Stories

Content affinity rails operate on a propensity for similarity of tags associated with each title. These rails are served to you with a personalized touch in mind but for this discussion we’ll focus on how tagging similarities pertaining to genre and tone. How the algorithms are designed could be its own article.

Critically Acclaimed says: trust is pre-built; your taste is safe here.

Award-Winning Witty says: external recognition has already absorbed the risk of a bad pick. Three different value propositions.

30-Minute Laughs says: low commitment, built for a tired Tuesday.

Three distinct viewer psychologies. One catalog asset serving all of them simultaneously.

Now compare “Today’s Top Picks” and “Gems for You” operate on cultural gravity: brand recognition doing its own persuasion before the thumbnail loads.

A click driven by brand recognition and a click driven by behavioral matching look identical in aggregate.

They are not.

Every title on a streaming platform accumulates a popularity score, and that score does significant work inside the curation engine. Popularity is, in many ways, the platform’s preferred instrument for managing risk.

“Today’s Top Picks for You” is built on exactly this logic: titles with high popularity scores are paired against your watch history and preference signals to produce a recommendation that reads, internally, as low-risk and high-probability. The algorithm is not simply guessing what you might like. It is hedging.

And the hedge works because brand recognition does part of the persuasion before the viewer has made any conscious decision. Surfacing a well-known title to an unfamiliar viewer is a fundamentally lower-stakes bet than surfacing something obscure, because the title arrives pre-loaded with cultural credibility. The platform borrows trust it did not earn.

In the cable era, you could read a grid. In the Blockbuster era, you could walk the aisles. In the streaming era, the shelf is encoded, the adjacencies are invisible, and the key belongs entirely to the platform.

A studio that could read those clusters would hold genuinely actionable intelligence: not just what their show is doing in isolation, but what structural role it plays inside the platform’s content ecosystem.

Below you’ll find a mocked up version of some of the insights of first party data streamers collect around the content analytics on their app.

The dashboard above is theoretical, every number modeled rather than measured, but the architecture of what it describes is real.

Netflix holds some version of the layers of these insights or at least the raw analytics:

• which row framing drove the click

• which thumbnail variant won the A/B test and was quietly served to 94% of impressions

• which behavioral cohort anchors the platform’s lowest churn segment

• which creator-loyalty thread running through the exit flow amounts to a development brief for the next greenlight.

The studio licensing Seinfeld holds unlikely any of it.

What can creatives and studios do with these insights?

The tag cluster is where it begins.

When “witty” becomes the connective tissue between Seinfeld, Suits, Gilmore Girls, and Sex and the City, the platform is not filing paperwork. It is surfacing the emotional frequency a specific audience is already tuned to when they open the app.

For a greenlighting conversation, that is a development brief the algorithm has already written: a sharp, language-forward register earns disproportionate discovery leverage because it anchors an entire content neighborhood rather than competing inside a single genre lane. That structural advantage lives in the metadata, invisible to anyone locked out of the curation logic.

The engagement data tells a quieter and more consequential story.

A comfort re-watcher returning to Seinfeld at 10pm on a Sunday, mid-series, three episodes deep, is not making an entertainment choice. They are performing a ritual. Some titles do not compete for attention. They compete for routine.

Understanding which of your titles have crossed that threshold reframes the licensing conversation entirely: you are not selling entertainment, you are providing retention architecture. The shows that earn a standing appointment rather than a one-time viewing event are a categorically different asset, and they should be priced that way.

The thumbnail finding is the one that should intrigue creative teams most directly.

The platform ran an A/B test, discovered a single-character conflict frame outperformed ensemble art by more than double, and served it to 94% of impressions without a conversation.

But that testing data is also a ranked index of the show’s most viscerally legible moments, a guide to which visual language the current audience associates with the IP right now, not at the time of original release. Useful for merchandising. Useful for future licensing. Useful for any adjacent production brief.

The platform ran the study. The studio could be reading the results.

The exit flow is the most forward-looking layer.

When 22% of sessions move from Seinfeld to Curb Your Enthusiasm (this is an obvious one), the algorithm has quietly detected a creator-loyalty thread that no viewership report has ever surfaced for the studio. That is a pre-validated development brief, demand already mapped and measured before a single pitch exists. The platform holds it. The studio could be acting on it.

So now what’s next?

What this analytics mock-up is really arguing is simple: placement logic, tag assignment, thumbnail testing, and audience behavior should be negotiating variables in any rights deal involving a title the platform has already identified as retention infrastructure.

These are not exotic requests. They are questions the platform answers for itself every Tuesday morning.

The data exists. The questions are knowable. The barrier is not technical. The barrier is that most studios are not even asking.

Part of that is an awareness problem. Streamers have this data siloed so aggressively that even their own teams do not always know what signals are being captured or why. But the more interesting problem is that different parts of a studio want entirely different things from the same black box:

• A development executive wants to know which tonal registers are gaining gravity with emerging audiences.

• A marketing team wants to understand which visual language viewers actually associate with the IP, not the one the original campaign decided on fifteen years ago and has never revisited.

• A rights negotiator wants to know whether a title functions as a satellite or an anchor in the platform’s curation architecture, because those are not the same asset, and they should not carry the same price.

All of this intelligence already exists, generated continuously, at scale, using IP the studio created and the platform is quietly using to get smarter. The platform is earning a graduate degree in your content. You are still reading the prospectus.

Platforms have built extraordinary fluency in behavioral data.

Most studios are still learning the vocabulary. And until IP owners can articulate what they are entitled to know, and what knowing it is actually worth, the black box stays intact. Not by necessity. Not by technical limitation. Asymmetric information sustains itself until the less informed party decides the cost of ignorance outweighs the discomfort of demanding more. In streaming, that moment is overdue.

The harder problem is that most IP owners cannot identify what to ask for because they have not built a framework for understanding which insights exist and which ones map to their actual business interests.

The gap rarely announces itself as a disadvantage. It presents as a measurement problem, a temporary friction, something to revisit next cycle. The fog is comfortable enough that most parties settle into it, while the data stays inside the platform doing compounding work and the people who built the underlying IP negotiate on instinct.

What makes this durable is that the failure does not live in one place.

From what I have observed working inside legacy media and speaking with people across the industry, three structural problems tend to reinforce each other: most teams lack a clear picture of what data exists even within their own organization;

• a data is siloed by design and by inertia inside corporate structures that rarely reward cross-functional curiosity

• a licensing team and a data team can share the same building while operating as though they are in entirely separate industries.

Sometimes these teams sit on the same floor.

The result is not malice. It is a failure of shared language, and no one has been given a reason to build one. Until that changes, the platform keeps the intelligence, and the people who created the underlying IP keep negotiating from a position they could strengthen, if only they knew where to start.

So this is that.

Ten articles worth of questions, patterns, and conversations with writers, producers, executives, and readers, all trying to make sense of the same shifting ground. I’ve distilled the things I get asked most about the state of the industry, through a creative, technical, and financial lens.

What follows is ten of them, organized across five categories:

1. The Money Problem — who built what audiences are watching, and why they are seeing almost none of what it earns.

2. The Measurement Problem — why platforms swimming in data still cannot tell which content is actually worth keeping.

3. The Creativity Problem — how risk appetite quietly collapsed, one defensible decision at a time.

4. The Structural Problem — why the mergers, release strategies, and systems designed to fix things keep recreating the same conditions instead.

5. The Audience Problem — what viewers are actually telling the industry, and why the signal keeps getting misread.

Quick Note: Everything here reflects patterns observed across the industry broadly, not one single company. Where something is a documented fact, I’ll say so. When it’s editorial read, I’ll say that too.

Part I: The Money Problem

Where did the revenue go, and who decided the people who made the content shouldn’t see it?

1. Writers helped build shows that stream millions of hours a month. Why are their residual checks sometimes smaller than a car payment?

What the record shows: In the broadcast era, viewership data moved through a public channel. Nielsen numbers were verifiable, contested, and negotiable, and residuals were calculated against a real number that everyone could see. Today, when a studio licenses IP to a streaming platform, there is no legal or financial requirement to share performance data with the people who made the content. Studios and creators often have no way to track how their titles are performing across multiple platforms. Streaming residuals currently average a one-time payment of $50 to $200 per title, regardless of how many millions of hours that title accumulates.

My Observations:

1. The residual check is a diagnostic, not just a grievance.

When it’s smaller than a car payment on a show streaming millions of hours a month, that gap is not a rounding error. It reflects a deliberate structural choice: platforms built their business models around proprietary data and flat fees because subscriber growth rewarded scale, not equity.

2. The labor problem runs deeper than opacity alone.

The shift from weekly network seasons to binge-release streaming didn’t just change how audiences consume content. It dismantled the employment rhythm the industry ran on. A show that aired over nine months out of twelve wasn’t only a viewing ritual for audiences. It was a labor infrastructure. Writers, editors, and post-production crews depended on that cadence for consistent, sustainable work. Streaming disrupted this by introducing shorter seasons, extended production hiatuses, and minimal one-time payments, transforming writing from a reliable middle-class profession into precarious gig work. You staff a room for six weeks, drop everything at once, and the workforce disperses. The residual system was never updated to account for any of it.

3. Capital that doesn’t circulate becomes parasitic.

Platforms spent years optimizing for minutes watched rather than revenue that could circulate back through the ecosystem. When capital fails to circulate and strengthen the broader ecosystem, it becomes an early sign of systemic parasitism. That isn’t hyperbole. It’s a measurable outcome: the people who built what audiences are watching have the least visibility into how it performs and no structural leverage to negotiate otherwise. The frameworks for fixing this already exist. What’s missing isn’t a mechanism. It’s the will to treat the people who made the content as participants in its long-term value rather than line items on a closing budget.

Read more: “The Kryptonite Inside Legacy Media Mergers Keeping Hollywood Hostage: Part I” | “The Business Casual Method” | “Content as a Product, Part II” | “If Audiences Want Fresh Content, Then Why Is Grey’s Anatomy Still Dominating Views in Streaming?”

2. Studios are among the most powerful publishers on the internet. So why are they still treating social media content like a marketing expense instead of a revenue stream?

What the record shows: the creator economy is currently valued at $191.55 billion and projected to reach $528 billion by 2030. Netflix’s YouTube channel has 31.5 million subscribers. Warner Bros. has 12.6 million. Both rank in the top 0.5% of all YouTube accounts globally, placing them among just 2,639 channels worldwide with over 10 million subscribers. YouTube itself generated $42.5 billion in revenue in 2024.

And yet, streamers are leaving ad revenue on the table. The content they post to their own social media accounts generates real money, and most of them are not collecting it in any systematic way. That gap is worth paying attention to, because a well-structured approach to that revenue could become one of the more reliable paths to funding consistent residuals around their IP.

My Observations:

1. The gap is structural, not creative.

The distance between what these channels could be and what they actually are is not a failure of imagination. It is rooted in how Hollywood has historically separated the money that makes a show from the money that sells it. For decades, the studio paid to produce the IP and the network paid to market it. Two budgets, two objectives, two organizational cultures that rarely spoke the same language. That arrangement made sense when the only downstream revenue came from licensing finished content to other platforms after syndication. The incentive to build anything beyond the show itself simply did not exist.

2. The distribution channel is now the asset.

What changed is that the channels themselves are now worth building, and no one has updated the deal structure to reflect that. Netflix and Warner Bros. are not just studios with YouTube channels. They are among the most powerful publishers in the entertainment industry, sitting on audiences that independent creators would build entire companies around. Every week, fans finish an episode and immediately migrate to TikTok, Instagram, and YouTube to extend the experience, generating views and ad revenue that flows entirely to third-party creators while the studio that funded the original captures nothing.

3. The fix is a licensing question, not a creative one.

If the studio retains ownership of supplemental content and licenses it to the network or platform the way it would license any other IP, then ad share revenue becomes a real and recurring income stream rather than an afterthought. The actors already have base rates for production. The marginal cost of extending that into owned digital content is not zero, but it is far lower than greenlit programming, and the upside compounds in a way that one-time residual payments never could. The reason this was never architected before is that studios never had to. Syndication handled the downstream economics. But syndication assumed a world where other platforms were the destination. In a world where the studio’s own YouTube channel has tens of millions of subscribers and YouTube is generating over $40 billion a year in ad revenue, licensing your supplemental content to yourself and collecting the ad share is not a creative strategy. It is simply closing a door that has been left open for years.

Read more: “The Business Casual Method: How Legacy Media Can Turn TV Budgets into Multi-Million Dollar Digital Empires”

Part II: The Measurement and Attribution Problem

Why are streaming platforms are swimming in data and still making decisions in the dark?

3. Platforms are swimming in data. So why does no one seem to know which shows are actually worth keeping?

What the record shows: Streaming platforms can measure retention and completion rates. What they cannot do, as an industry-wide structural problem, is cleanly attribute which content drove a subscriber to sign up, stay, or cancel. Marketing teams track acquisition across external channels while product teams measure only in-platform engagement. These datasets rarely reconcile. Free ad-supported streaming is growing at roughly twice the rate of premium subscription services, partly because ad-supported attribution is simply cleaner and it’s economically scalable for the consumer and business.

My Observations:

1. The cable era had a legible P and L. Streaming broke that and never replaced it.

In the cable era, attribution was straightforward. If a show drew more eyeballs, a network could charge advertisers more per slot. More viewers meant higher rates, higher rates meant more revenue, and more revenue justified renewal. You didn’t need a sophisticated data infrastructure to understand whether a show was earning its keep. The P and L told you. Streaming broke that model and never fully replaced it with something coherent. Instead of selling eyeballs to advertisers, platforms moved to a subscription model: pay one flat rate, consume as much as you want. The problem with an unlimited buffet is that some people eat once and leave, and some people are there every night. If you can’t tell the difference, and can’t price accordingly, you’re not running a sustainable business. You’re running a subsidy.

2. Nobody owns the full customer journey, and the customer journey is the only thing that matters.

Take Stranger Things. It’s reasonable to assume that the show brought millions of people to Netflix and kept millions more from leaving. But how do you actually measure that? How much of any given subscriber’s lifetime value do you credit to Stranger Things versus everything else they watched? What about the person who signed up because of Stranger Things but stayed because of The Crown? The data exists in fragments across teams that don’t talk to each other: one team owns acquisition, another owns in-platform engagement, another owns churn. Nobody owns the full journey, and the full journey is the only thing that actually answers the question.

3. Platforms default to measuring the surface of attention, not the depth of it.

The result is that platforms default to the metrics they can measure cleanly: minutes watched, completion rates, opening-weekend numbers. These are real signals, but they’re measuring the surface of attention, not the depth of it. A show can score well on all of them and still not be the reason anyone subscribed. A show can score poorly on all of them and still be the reason a household hasn’t cancelled in four years. The industry doesn’t yet have the infrastructure to tell the difference, and until it does, cancellation decisions will keep feeling arbitrary, because in a meaningful sense, they are. The one place attribution still works cleanly is in ad-supported streaming, which is precisely why it’s the fastest-growing segment of the market.

Read more: “The Kryptonite Inside Legacy Media Keeping Hollywood Hostage: Part II” | “If Audiences Want Fresh Content, Then Why Is Grey’s Anatomy Still Dominating Views in Streaming?” | “What Fight Club Can Teach Us About the Future of Audience Data”

4. Streaming platforms track hundreds of millions of viewing hours. So why do they still not actually know what their content is worth?

What the record shows: Every title in a streaming library generates signals well beyond viewing hours: social media mentions, search volume, fan-created content, merchandise demand, and ongoing discovery value that can persist eighteen months or more after release. Platforms currently measure viewing activity as the primary performance variable. Meanwhile, FAST platforms with clean ad attribution are growing twice as fast as premium subscription services, and YouTube has surpassed Netflix and Amazon combined in U.S. television viewing share.

My Observations:

1. “Minutes watched” is a measure of activity, not a business model.

Think about what a show actually produces. A hundred million viewing hours is one number. But alongside it: hundreds of millions of social impressions, fan-created content that extends the IP’s life across platforms the studio does not own, merchandise signals, search volume spikes, discovery traffic that keeps arriving long after the marketing budget ran out. Platforms are currently measuring one variable and leaving the rest on the table, unmonetized and largely untracked. “Minutes watched” is the streaming equivalent of a trading floor counting transactions without tracking whether any of them were profitable.

2. A show’s cultural footprint is a data asset. Treat it like one.

Sophisticated financial operators already understand this logic. Hedge funds pay significant premiums for alternative data and sentiment signals precisely because those signals capture value that conventional metrics miss. The parallel for streaming is direct. The social conversation around a show is an asset. The fan ecosystem it generates is an asset. None of them show up in a minutes-watched report. The reason FAST platforms are outpacing premium subscription services is not because audiences prefer ads. It is because ad-supported models have clean attribution. YouTube surpassing Netflix and Amazon combined in U.S. viewing share is not a cultural shift. It is what happens when a platform with perfect attribution competes against platforms with broken attribution.

3. Content is a portfolio of financial instruments, not a uniform asset class.

The fix requires treating content as what it actually is: a portfolio of financial instruments with different return profiles, different shelf lives, and different downstream revenue potential. A catalog title generating steady rewatch traffic eighteen months after release is not the same instrument as a prestige limited series engineered for one weekend. Managing them with the same metric is the equivalent of running a bond portfolio and an equity portfolio off a single spreadsheet and wondering why the returns don’t make sense.

Read more: “If Audiences Want Fresh Content, Then Why Is Grey’s Anatomy Still Dominating Views in Streaming?”

Part III: The Creativity Problem

The risk appetite is gone. Here’s exactly how that happened and where it leads.

5. Why are studios playing it safe and doubling down on franchises? Where did the big swings go?

What the record shows: FilmLA reported a 16% decline in shooting days in Los Angeles County in 2025, down nearly half from the 2018 peak. The Directors Guild reported guild-member employment down 35 to 40% as of early 2025. Disney and Fox have both reduced their annual film output. Programming budgets are being quietly reallocated toward sports and live events, a direct substitution of creative risk for the safety of pre-sold audiences. The proof is in the green-light: HBO, the same network that once green-lit The Wire without a guaranteed audience, is now producing a Harry Potter series. Pixar, the studio that built its reputation on original worlds, is releasing Toy Story 5.

My Observations:

1. The best content wasn’t made by the studios with the most resources. It was made by the ones with enough breathing room to be wrong.

HBO didn’t green-light The Sopranos because the data told them to. They green-lit it because they had enough stability to take a chance on something that had never quite existed before, and let it find its audience over time. That audience, famously, didn’t know it wanted that show until the show existed. You cannot optimize your way to that outcome. You can only create the conditions for it.

2. Debt doesn’t kill creativity through malice. It kills it through accumulation.

What’s changed is that those conditions have been systematically dismantled, not through any single decision, but through the accumulated weight of mergers that promised scale and delivered debt. When a studio is carrying that kind of financial obligation, something quiet but consequential happens to every creative conversation inside the building. The question stops being “is this worth making?” and starts being “can we afford to be wrong?” Those feel like similar questions. They produce completely different answers. The first question leads you toward The Wire. The second leads you toward Toy Story 5.

3. The measuring tools are optimized for what people did, not what they might love.

The problem isn’t that executives have lost their taste, or that audiences have lost their appetite for something new. The problem is that the metrics now in use, opening weekend numbers, completion rates, minutes watched in the first 72 hours, are all built to measure what people did, not what they might love. Succession nearly got cancelled after its first season. By every short-term metric, it was underperforming. By any honest cultural measure, it was becoming something irreplaceable. The infrastructure we’ve built cannot tell the difference. So what we’re left with is a slow narrowing that doesn’t feel dramatic from the inside: the original pitch that loses the room, the mid-budget risk that gets replaced by a franchise extension, the singular voice that never gets a second meeting because it doesn’t come with a pre-existing audience attached.

Read more: “Not Another Hollywood Eulogy” | “The Kryptonite Inside Legacy Media Mergers Keeping Hollywood Hostage: Part I” | “If Audiences Want Fresh Content, Then Why Is Grey’s Anatomy Still Dominating Views in Streaming?” | “What Fight Club Can Teach Us About the Future of Audience Data”

6. Is AI actually coming for writers’ jobs, or is the panic being overstated?

What the record shows: AI language models optimize mathematical parameters to predict likely continuations based on patterns they have already seen. They do not form beliefs through lived experience. The widely publicized incident of an AI-generated Seinfeld stream producing racist content illustrated what happens when mimicry operates without grounded meaning. Meanwhile, independent creators on YouTube and Spotify are integrating AI into their workflows not to replace creative voice but to increase output and reduce time spent on manual production tasks.

My Observations:

1. The real question isn’t replacement. It’s adoption curves.

Both creatively and corporately, the dividing line is figuring out where automation can carry the load and where human judgment still has to step in. On the creative side, that looks like offloading the mechanical work: script formatting, structural drafts, metadata tagging, the kind of friction that eats time without adding meaning. More of that gets handled, and in a healthier world, more work actually gets made.

2. On the corporate side, the human touch isn’t decorative. It’s what makes the experience feel built for someone.

Content curation on a streaming platform can be fully automated, but something gets lost when it is. You stop feeling like the app knows you and start feeling like you are being processed. The human element in that context is what makes the experience feel like it was built for someone rather than optimized for everyone.

3. The honest answer: anyone claiming to have solved it is moving too fast to be truthful.

The real question is not whether AI takes creative jobs wholesale. It is whether people who know how to use these tools will be expected to carry workloads that previously required larger teams. That outcome is already visible in certain corners of the industry, and it deserves a more serious conversation than the binary of “jobs saved” versus “jobs gone.”

Read more: “AI Tooling is Viagra for Creatives: It’s an Enhancement, Not the Organ”

Part IV: The Structural Problem

Mergers, release strategies, and the slow collapse of the systems that once made television great.

7. Every merger promises creative reinvention. Why does everything just keep getting slower?

What the record shows: The pattern predates any single deal. AOL Time Warner, AT&T WarnerMedia, Warner Bros. Discovery, and now maybe “WarnerMount”: the same logic, the same promises, strikingly similar outcomes, roughly every four to five years. When Warner Bros. Discovery formed in 2022, content impairment charges reached $3.5 billion in the first year alone. Industry-wide, entertainment and media layoffs were up 18% in 2025, with over 17,000 jobs cut. The deals keep getting larger. The results keep arriving late.

My Observations:

1. Consolidation feels like strategy. It isn’t always one.

Mergers in media have become comfortable by default. Not because they work, but because they are the move the industry already knows how to make. When the landscape feels unstable, consolidation feels like control. It has the shape of a strategy even when it isn’t one.

2. The first years of any merger are spent just trying to establish a shared language.

Here’s what actually happens inside these organizations after the ink dries. You now have two companies, each with their own content management systems, their own definitions of what a successful show looks like, their own data infrastructure built for a completely different business. Merging those things is not a clean handoff. It is years of reconciliation work: figuring out whose metadata schema wins, whose performance benchmarks apply, which teams are duplicated, which institutional knowledge walks out the door when the restructuring begins. HBO Max inherited Discovery’s tech stack, HBO’s prestige brand, Discovery’s reality TV playbook, and CNN’s news infrastructure simultaneously. Each division defined success differently. None of those definitions were designed to talk to each other. By the time the culture is even partially stabilized, the financial pressure has usually already arrived, and the window for genuine reinvention has quietly closed.

3. The outcome isn’t a surprise. The question is why the industry keeps choosing it.

Part of the answer is structural: when your competitors are consolidating, staying independent feels like exposure. Part of it is that the people who architect these deals are rewarded for completing them, not for what they produce five years later. And part of it, honestly, is that no one has yet shown the industry what a merger done differently could actually look like: not just combining assets, but building a shared operational vision before the deal closes, with a clear answer to the question of whose technology, whose culture, and whose definition of success is going to govern the combined company from day one. Until that changes, the pattern will keep repeating. The names and numbers will change. The plot will not.

Read more: “Not Another Hollywood Eulogy” | “The Kryptonite Inside Legacy Media Mergers Keeping Hollywood Hostage: Part I” | “The Kryptonite Inside Legacy Media Keeping Hollywood Hostage: Part II”

8. Binge-dropping a full season used to feel exciting. Why does it now feel like the fastest way to kill a show’s cultural life?

What the record shows: Reed Hastings once said Netflix’s biggest competitor was not HBO or Disney. It was sleep. He meant it as a compliment. In hindsight, it reads more like a confession.

Think about what Lost actually was. Not just a show, but a weekly obsession. Every Thursday, 23 million people watched the same 42 minutes and spent the next six days deep in it. Was the smoke monster alive or mechanical? What happened if the button was never pushed? Those questions had a whole week to breathe, and that breathing room turned strangers into a community. At its peak, fan site Lostpedia had over 7,000 articles written by people who could not stop thinking between episodes.

That shared anticipation was not a side effect of the show. It was the show.

Media scholar Scott Eric Kauffman said, “Binge-watching fundamentally changes the basic unit of cinematic storytelling.” The container shapes the content, and dismantling the weekly model had consequences nobody fully priced in.

The Nielsen data from 2025 is where those consequences show up. For the first time since Nielsen began tracking streaming, not a single new show cracked the top ten most-watched titles of the year. Every title on that list was built for waiting. And waiting, it turned out, was the whole point.

My Observations:

1. Weekly television was a social contract. Binge-dropping dissolved it.

Picture 9 PM on a Sunday in 2007. Millions of Americans are settling in to watch The Sopranos, knowing that tomorrow they will gather around office water coolers to dissect every moment. That wasn’t just appointment viewing. It was a shared cultural contract. The week between episodes wasn’t dead air. It was where the show actually lived: in the arguments, the theories, the anticipation, the way a story kept working on you even after the screen went dark. Now picture the Friday a new streaming series drops all eight episodes at once. By Monday the conversation has peaked. By Thursday it’s over. The show existed, but it never quite landed anywhere. Nobody had time to miss it. Nobody had to wait for anything. And without the waiting, there’s no shared moment to gather around.

2. Streaming solved for the individual viewer and accidentally dissolved the audience.

This is the hidden cost of the binge model that rarely gets named directly. It’s not about quality. It’s about architecture. Weekly television was structurally designed to create communal rituals, the same way sports are. You watched on the same night as everyone else, at roughly the same time, and the experience of watching was inseparable from the experience of talking about it afterward. That social layer wasn’t incidental to the show’s cultural life. It was the show’s cultural life. When you optimize entirely for immediate gratification, you fragment the synchronized experience that turns a show into a cultural moment. Viewers watch in isolation, on divergent timelines, and what was once a unifying phenomenon becomes a series of private ones.

3. The platforms winning this back are the ones reintroducing the space between episodes.

Think about what The Bear, Severance, The Last of Us, and The Pitt have in common. They are all hour-long episodes released weekly. Not because weekly is some inherently superior format persay, but because it gives a show room to breathe, and more importantly, room to mean something.

Bingeing quietly eroded that. When everything is available at once, you are not watching a story unfold, you are processing one. The space between episodes, that week where you argue with a friend about what happened, where a scene stays with you on your commute, where anticipation has time to build into something real, that is not dead time. That is where a show earns its place in the culture.

The titles people still talk about years later are rarely the ones they consumed in a weekend years later.

Read more: “If Audiences Want Fresh Content, Then Why Is Grey’s Anatomy Still Dominating Views in Streaming?” | “Content as a Product, Part I” | “Content as a Product, Part II”

Part V: The Audience Problem

What audiences are actually telling us, and why the industry keeps misreading the signal.

9. Streaming platforms spent a decade and billions of dollars chasing the next cultural moment. So why did a 2005 medical drama just win the year?

What the record shows: For the first time since Nielsen began tracking streaming data in 2020, not a single new show cracked the top ten most-watched of 2025. The list was dominated by Grey’s Anatomy (2005), Bluey (2018), Squid Game (2021), and Stranger Things, now in its ninth year. FAST platforms like Tubi and Roku are growing at twice the rate of premium subscription services. YouTube has now surpassed Netflix and Amazon combined in U.S. television viewing share.

My Observations:

1. The instinct is to read the fact that not a single new show cracked the top ten most-watched titles as an “audience problem”. That reading is wrong.

Viewers are not nostalgic, risk-averse, or allergic to novelty. Framing it that way lets the platforms off the hook for a structural failure of their own making. Grey’s Anatomy didn’t win 2025 because audiences stopped wanting something new. It won because it was built for a completely different economic era: 22 episodes per season, weekly releases over nine months, and decades of syndication revenue as the baseline assumption. You could discover it in season four. You could rewatch it indefinitely. It was designed, both narratively and financially, for infinite return visits. The streaming-native model inverted every one of those principles.

2. Platforms built a market almost entirely composed of growth stocks, with almost no catalog depth underneath.

They spent the better part of a decade engineering shows to dominate a single weekend of conversation, then quietly shelving them before they could take root. When the growth stocks underperformed, there was nothing underneath them. So audiences drifted back to the oak tree while the platforms kept planting flowers designed to bloom for two days.

3. Catalog content was doing the essential work all along. Platforms largely ignored it.

Think of a content like stocks. The shows that “move markets” are the high-growth stocks everyone is excited about at the same time. They spike, generate headlines, and make executives look visionary at conferences. But high-growth stocks are volatile. They need constant attention, constant capital, and they tend to disappoint the moment the excitement fades.

Catalog content is the index fund. Unglamorous, quietly compounding, never the thing anyone brags about at dinner. But it is what actually keeps the lights on. It anchors subscribers who might otherwise cancel. It generates rewatch revenue without new marketing spend. It justifies the monthly fee on a slow Tuesday in February when nothing new is dropping.

Netflix spent $17 billion on content in 2023 chasing the spike. Meanwhile The Sopranos, which premiered in 1999, still pulled in viewers on HBO Max with brand equity alone. No campaign. No buzz. Just a great show, doing what durable assets do.

Read more: “If Audiences Want Fresh Content, Then Why Is Grey’s Anatomy Still Dominating Views in Streaming?”

10. Tourists are literally flying to Croatia because of Game of Thrones. Why hasn’t HBO built a theme park there yet?

What the record shows: Disney’s Experiences segment generated $10 billion in quarterly revenue in Q4 2025, with parks delivering 8% higher domestic operating income year over year. Streaming’s operating margin sat at 8.4% in that same period. The Harry Potter franchise demonstrates a cleaner alternative: Warner Bros. collects IP revenue without bearing infrastructure costs, and the franchise is expanding globally with new locations including Tokyo.

My Observations:

1. The Croatia effect is a revenue leak at scale, not necessarily a feel-good fandom story.

Tourists didn’t need HBO to tell them to go. They went anyway, spending money in markets HBO never touched. That is not a testament to fandom. It is a measurable gap in how the industry conceptualizes the downstream value of its own IP.

2. The first question isn’t “should we build?” It’s “where does this IP actually live in the world?”

Global viewership data shapes everything downstream, including whether it makes sense to own infrastructure at all or license the right to build it to a hospitality operator who already knows how to run one. The second question is durability. There is a meaningful difference between an IP with generational staying power and one riding a cultural moment. A theme park is a long-horizon capital commitment. The fanbase has to be the kind that passes the show down, not the kind that moves on after the finale discourse fades.

3. Market fit is the most under-appreciated question in the entire conversation.

A Game of Thrones experience built for a US audience and one built for European markets are probably not the same product. One market might want the immersion and the fandom, the feeling of actually standing in Westeros. Another might be drawn to the craft: production history, set design, the behind-the-scenes texture of how the world was constructed. Collapsing those into one global template is likely to underserve both. This is also where the build-versus-license decision gets more nuanced. Markets with deeper fandom and higher capital potential might support owned experiences with direct ROI accountability. Others might be better served through licensed pop-ups: lower commitment, shared risk, generative revenue without the balance sheet exposure. The Warner Bros. and Universal arrangement around Harry Potter is the relevant proof of concept, not because it is the only model, but because it cleanly separates who owns the IP from who knows how to operate an experience at scale.

4. The subscriber-to-theme-park-visitor crossover is not a given.

The audience willing to pay for content and the audience willing to buy a plane ticket are overlapping circles, not the same one. Understanding that gap, rather than assuming fandom translates uniformly across formats, is what separates a real strategy from an expensive bet placed on brand enthusiasm.

Read more: “Why Haven’t We Experienced a Game of Thrones Theme Park Yet?”

There is a version of Fight Club for everyone.

That is not a throwaway claim. Think about it practically.

• Have you ever yearned to see a young Brad Pitt at the absolute height of his magnetism? Watch Fight Club.

• Have you ever tried to understand why men are so lonely? Not the surface kind of lonely, but the kind that quietly metastasizes into something far darker? Watch Fight Club.

• Have you ever questioned your existential purpose inside the fluorescent hum of corporate America? Watch Fight Club.

• Have you ever wondered where male rage actually comes from; what feeds it, what shapes it, what eventually breaks it open? Watch Fight Club.

• Have you ever wanted to feel something with real weight to it, something that lingers long after the credits roll? Watch Fight Club.

Every one of those is a legitimate entry point. Every one of those is a real person, sitting in front of a screen, with a real reason to click play.

And yet when Fox released the film in 1999, they were genuinely uncertain there was an audience for it at all.

That tension is the whole argument.

The Problem Was Never the Content. It Was the Packaging.

Fight Club’s marketers experimented, almost accidentally, with what we would now recognize as multi-modal positioning. The early trailers were released as fake public service announcements, letting the characters breathe outside the frame of “movie promotion.”

The DVD release leaned into parody, a kind of winking irreverence that made Tyler Durden (Brad Pitt at the height of his magnetism) feel like someone you had already met. These were not savvy data-driven decisions. They were instinctive moves toward a truth the industry still struggles to formalize: the most durable content is not bound to a single genre label or a single type of viewer. It is bound to a deeper human question that can be entered from many directions.

That question, in the case of Fight Club, is not really about underground boxing or anarchist soap companies. At its root, the film is a story about the slow suffocation of meaning inside modern life.

• What do we do?

• Why do we do it?

• Who decided that this, all of this, was the point?

That is not a niche question, it is a universal one. And it is the kind of thematic gravity that, when it exists in a piece of work, makes the surface-level genre packaging almost beside the point. The depth is already there. The job is identifying all the different doors into it.

For a title with that kind of thematic range, the most revealing exercise is not tracking who watched, but reverse engineering how they arrived. As studio budgets show no signs of expanding anytime soon, this kind of inquiry becomes more than academic. It becomes a lens for two practical questions that anyone working in development or content strategy cannot afford to ignore:

When evaluating the value of IP, whether as a buyer or acquirer, who is the audience and what is drawing them there? And what are the universal ingredients we can keep repeating to develop the potential of an unknown idea?

These questions only have answers if the content was built with enough depth to generate them. You cannot reverse engineer a world that was never really built. That is why world-building and thematic depth matter not as creative luxuries but as operational assets, in two distinct places: at the development stage, where foundations are laid, and at the acquisition stage, where value is being measured.

Existing IP: The Architecture Hiding Inside Titles You Already Own

This is where legacy streaming data becomes genuinely powerful, and where the industry is leaving significant value on the table.

I opened this piece with a simple provocation: there is a version of Fight Club for everyone. Not as hyperbole, but as a structural argument about how great content actually works. The same film that pulls in a viewer chasing a young Brad Pitt at the height of his magnetism is the same film that meets a burned-out analyst somewhere in the fluorescent hum of his ninth consecutive hour of spreadsheets. It is also the same film that articulates male loneliness better than most therapy, and functions as a pure visceral thriller for someone who has never once thought about its philosophy.

Was the entry point talent-driven? Did the viewer come for Brad Pitt and stay for David Fincher? Or was it theme-driven, surfacing in ways the original marketing never anticipated or intended?

Below are the kinds of tags, what the industry calls taxonomies, that could be assigned to Fight Club and matched to a viewer query to surface the film.

The tags that actually surface Fight Club in 2026 tell a more complicated, and more honest, story about what the film means to the people finding it now.

• Person A searches “why are men so lonely.”

• Person B is two years deep into corporate burnout, craving something existential but unable to name it.

• Person C is 19, has never heard of David Fincher, and clicked because the key art looked cool.

Three different doors. The same film behind all of them.

That is the latent architecture of a title with genuine thematic density. Each tag is not just a descriptor. It is a different viewer’s reason for showing up. And here is the part that rarely gets acknowledged: most viewers will never hand you that reason directly. It’s likely rare to find someone typing“existential dread” into a search bar when they’re looking for a movie to watch.

The vocabulary of that kind of discomfort is not clinical. It is ambient. It is the feeling of sitting at a desk on a Wednesday afternoon wondering, quietly, if this is all there is. But that same person might pause on a content rail that seems to know them. They might click a key-art that suggests someone else has been exactly where they are.

This is where metadata stops being administrative and starts being strategic, and where packaging stops being a cosmetic decision and becomes a structural one.

Consider two rails built from the same five films.

Existential Classics is a curatorial statement. It signals genre, taste, cultural literacy. Feeling Existential? is an invitation, maybe even a quiet admission. It does not describe a category. It describes a state of being. For a viewer who cannot quite articulate why they opened the app, that distinction is the difference between scrolling past and staying.

The right tag, the right rail, the right thumbnail at the right moment can reach a viewer without ever making them name what they are looking for. For a title like Fight Club, the question is not what genre it belongs to. It is how many dimensions might lead someone there. Each pathway is a viewer the platform either catches or loses. That is not a small distinction. That is the actual value of investing in deeper metadata infrastructure: not just better prediction, but better translation, closing the space between what someone is feeling and what the platform has to offer.

And that gap, the one between a feeling and a title that meets it, is precisely where long-form streaming keeps losing ground to platforms that have been quietly solving this problem for years. That is the difference between someone pressing play and someone opening YouTube instead.

Why Spotify Knows You Better Than Your Streaming Platform Does

Platforms like YouTube, TikTok, and Spotify have already mastered the art of anticipating what you need before you know you need it. Long-form streaming has not caught up, and the gap is more structural than it first appears.

Spotify does not just know what you listen to. It knows when:

• Classical music at 7am.

• A podcast about the markets on your commute.

• Indie rock on your way home.

• Sad songs at midnight on a Tuesday that make you nostalgic about your ex after a few drinks.

The platform has built an emotional profile around the rhythms of your day, one you never consciously handed over. TikTok and YouTube are doing the same thing, just faster, with more repetitions. Every scroll, every pause, every abandoned video is a data point. Short-form platforms run thousands of small experiments on your attention every single day.

Long-form streaming does not have that luxury. If you open a platform, watch one episode, and log off, the platform learns almost nothing about you. Viewing sessions are longer, less frequent, and more deliberate. The volume of behavioral signals is simply not there in the way it is on short-form, and without external data partnerships to fill those gaps, the picture any platform holds of you stays structurally incomplete.

An incomplete picture means the system is guessing.

The Gap Between a Tag and a Feeling

Genre tags hit their limit here. Calling something a “romance” points you in a general direction, but there is a real difference between a show about falling in love and a show about the specific ache of loving someone who is already leaving. Both are technically romance. Only one is what you actually needed at 11pm on a Tuesday.

That emotional texture, the particular register a title is built around, is what closes the distance between a film and the right viewer at the right moment.

There is something worth sitting with in that distance.

Long-form content asks far more of you than a track or a short video. A song costs three minutes of your time. A series costs an entire evening, sometimes more, and that difference in investment changes everything about how a recommendation has to work.

Part of the reason is structural. Most legacy streamers do not have the infrastructure to map viewer intent at any meaningful depth, and the ones that do have little incentive to share those insights back with the original IP holders. The data lives inside the platform. The intelligence stays there too. That is not an accident of technology. It is a consequence of how these organizations were built and what they were built to protect.

I have written about the deeper architecture of that problem elsewhere, but the short version is this: the pipes connecting insight to action were fractured long before anyone thought to ask what the data was actually saying.

Which is why the more interesting signal does not require anyone’s permission to receive.

Social listening captures the emotional language audiences actually use about a title, not genre labels but feelings. Search trends reveal what someone was looking for when the algorithm served them a film. Letterboxd reviews and Reddit threads are messy, unsolicited, and remarkably honest records of why someone actually watched something and what it did to them afterward. None of that lives behind a platform dashboard. All of it is navigable.

The studios that figure this out stop waiting for the streamer to hand them a report. They treat audience behavior in the wild as infrastructure. Imperfect and partial, but pointing somewhere. And that is the shift worth making, because this is not just a diagnostic tool for old titles. It is a design principle for new ones.

New IP: How to Reverse Engineer a Cultural Argument Before Anyone Has Written the First Draft

If you can map the multiple pathways into a title that already exists, you can start asking a harder question from day one:

Are we building something with enough depth to generate those pathways in the first place?

The question to keep returning to at the ideation phase is not “what genre is this?” It is: what is the irreducible core of this story, and how many different people does that core speak to?

Take The Summer I Turned Pretty and Heated Rivalry. Stack them side by side and the surface read is nearly identical. Both romantic, both emotionally driven, both adapted from book series with devoted readerships. A traditional genre breakdown would file them next to each other without a second thought.

But when you run a tagging analysis on each title, something more interesting appears.

Both Stories are about Desire. But You Should Know Upfront, Desire Is Never Really About the Other Person.

The Summer I Turned Pretty is, on its surface, a love triangle set against beach house summers and the particular ache of standing at the edge of adulthood. Belly Conklin spends three seasons caught between two brothers and two versions of herself, and the show’s built-in readership, drawn from Jenny Han’s bestselling novels, arrived with strong opinions already intact. But the audience that actually showed up was much larger and considerably harder to explain.

The #TeamConrad versus #TeamJeremiah debate was not constrained to Amazon’s expected audience demo when the show began in 2022. Instagram and TikTok were flooded with answers from everywhere: McDonald’s, Delta Airlines, the Empire State Building, Malala Yousafzai, and the Cincinnati Bengals all weighed in.1 That is not a teen romance fandom. That is a cultural argument wearing the costume of one.

The show worked because it did not actually require you to be a teenager to feel it. It required you to remember being one, and more specifically, to remember the version of yourself who had not yet made all the choices that foreclosed the other possibilities. Conrad versus Jeremiah was never really about the brothers. It was the audience nostalgic of their own crossroads, arguing about who they were at seventeen with the benefit, and the grief, of knowing who they became.

Every person who clicked on that show brought their own unresolved adolescence to the debate. The conflict gave them somewhere to put it.

Heated Rivalry operates on an entirely different frequency, though a traditional genre breakdown would file it in the same section as everything it is not. When a colleague from our engineering team came to me asking what in our algorithm had caused that show to break out, my honest answer was: nothing algorithmic. The algorithm did not manufacture that moment. The show earned it by being, at its core, a story about yearning and desire between two equals (and lots of soft porn); that is a distinction worth sitting with.

The show tells the story of two secretly gay athletes in the highly masculinist world of professional hockey: Shane Hollander, a half-Asian Canadian who plays for Montreal, and Ilya Rozanov, a Russian who plays for Boston, carrying on an intermittent love affair while nurturing an intense rivalry on the ice. The premise sounds contained. The audience it found was anything but. It became the most-watched non-animated show acquired by HBO Max since the platform’s launch in 2020.2

What that number reflects, at least in part, is a genuine hunger the genre had stopped feeding. Audiences, particularly the female demographic, had grown quietly starved for something that understood desire as a force with weight and consequence.3

Heated Rivalry understood that the most charged version of wanting is the kind that costs something. Two people who are supposed to be opponents, navigating attraction through half-hidden signals: shared grins during face-offs, meaningful glances across crowded rooms, feet pressing together under a table during press conferences. The show pulled in viewers who would never describe themselves as “romance audiences” because it was not offering them a romance. It was offering them yearning, specifically the kind that the rest of the content landscape had quietly stopped providing. That premise does not stay inside a genre. It leaks out into people who did not come looking for it and finds them anyway.

The algorithm registered the breakout after it happened. It had no role in causing it. That is the part the engineering team’s question quietly revealed:

Content recommendation algorithms are sophisticated infrastructure for measuring and amplifying what audiences resonate with. The conditions that make a story resonate in the first place live entirely outside that infrastructure. They live in the writing room, in the casting, in the decision to let two characters want each other in a way that costs them something real. No recommendation engine creates that. It can only, at best, make sure the right people find it once it already exists.

Same shelf. Completely different architecture underneath.

What Soapy Young Romances Reveal About the Architecture of Lasting Television in the Age of Streaming

When a piece of work is built with enough depth, the genre label becomes almost incidental. The real structure is underneath it: the emotional entry points, the thematic tensions, the character contradictions that give audiences something to fight about long after the credits roll. The Summer I Turned Pretty was, first and foremost, a word-of-mouth hit, evidenced by viewing parties that sprang up across continents from Sydney to London to New York by its finale. That kind of reach does not behave like attention. It behaves like investment that compounds.

But here is what the industry most needs to hear right now:

Both The Summer I Turned Pretty and Heated Rivalry arrived with a structural advantage most original development does not get: a pre-built world. Jenny Han spent years constructing Cousins Beach before a single frame was shot. Rachel Reid had spent hundreds of pages constructing the emotional infrastructure of that rivalry before a writers' room ever convened. The depth that made those shows resonate was not invented for television. It was inherited.

That inheritance is real. But it is not a prerequisite.

World-building is treated as something that happens before the show, inside a novel, a comic run, decades of franchise mythology, rather than something that can be built into development from the first conversation in the room. That assumption is costing the industry more than it realizes.

The architecture underneath both shows did not require a pre-existing fandom. It required someone asking harder questions at the ideation stage than the industry typically demands:

• What are the contradictions inside these characters that audiences will want to resolve long after the finale?

• What is the argument underneath the plot, the one different kinds of people will take different sides of, not because the writing told them to, but because it touched something unresolved in them?

These questions belong at the beginning, not the end. The world does not expand after the story is told. It expands because of decisions made before anyone typed page one.

The shows that arrived with novels behind them succeeded not because of source material, but because someone had already built a world with enough internal coherence to mean different things to different people without falling apart. There is no reason that work cannot begin in a writers’ room, on an original idea, from day one.

The tools are not the obstacle. The permission is.

The Secret Sauce an Algorithm Can’t Predict

The tools available to development executives today are genuinely impressive. You can model cast configurations, optimize release windows, cross-reference Nielsen data with streaming engagement curves, and build what amounts to a fantasy league of talent, theme, timing, and setting.

The shows that actually became culture were rarely the ones that would have scored highest on a pre-production data model.

The Sopranos debuted in 1999, at the exact moment America was beginning to feel the first tremors of a male existential crisis it did not yet have language for. The postwar social contract, work hard, provide, succeed, was quietly coming apart, and men were sitting inside lives that looked correct from the outside while something unnamed hollowed them out from within. Tony Soprano did not invent that feeling. He just put it in a therapy chair and lit it dramatically. Fight Club arrived the same year is not a coincidence.

Both were tapping the same undercurrent, a generation of men doing everything they were supposed to do and finding themselves at a desk on a Tuesday afternoon wondering, with genuine bewilderment: what is the point of existence?

Sex and the City arrived a year earlier in 1998 on HBO, catching a different but equally unspoken shift. Women were becoming financially independent at a scale the culture had not quite processed. Marriage was being delayed, careers were being chosen, desire was being discussed openly among women in ways that had no real television precedent. None of this was new as lived experience. What was new was seeing it named, centered, and refused to be treated as a problem that needed solving.

Both shows were, in a meaningful sense, “first to market” on ideas that were already circulating in the undercurrent of the zeitgeist but had never been expressed with that kind of scale, confidence, or formal ambition. The Sopranos was a premium cable drama about an anxious mob boss in therapy at a moment when prestige television barely existed as a category. Sex and the City was premium cable dramedy about four women talking frankly about desire and ambition in a way that was not supposed to find that size of audience.

Neither would have survived a data environment optimized for precedent, because there was no precedent. What made them endure was not their genre mechanics. It was that they named something audiences were already quietly feeling but had not yet seen reflected back at them from a screen. That is not a formula. It is an act of cultural recognition. And no amount of tagging infrastructure, however sophisticated, can manufacture the conditions that make it possible.

That is the gap no algorithm has ever closed and will never close. Algorithms are extraordinarily good at identifying and amplifying what audiences have already responded to. They are structurally incapable of predicting what audiences are ready to feel for the first time. The data is always reading yesterday’s output. The development team’s job is to be one step ahead of the feedback loop, operating on signal before it becomes consensus.

The fear underneath Hollywood’s current dysfunction is the fear of that gap. IP proliferates because it arrives with pre-existing audience proof. All of it is risk mitigation dressed up as strategy, which is why, despite an unprecedented volume of content, genuinely new pop culture moments feel increasingly scarce. The system is optimizing for the wrong output.

Those conditions have a name. And building them is the actual job.

The Deeper Principle: From a writers’ room to marketing, it’s the craft of World Building

The future of content development is not just about making good shows.

It is about making shows with enough depth that they generate their own interpretive universe. Characters who feel like they exist outside the episode. Themes layered enough that audiences can enter from different angles and still find something real waiting for them. A world with enough texture that it invites excavation, not just consumption.

This is what the tagging work is actually measuring. Not genre or demographic but the number of real doors a story has built into itself. And that number is not a creative metric, it is a business one. The same depth that gives a tagger twelve entry points gives a marketing team twelve campaigns, a platform twelve audience segments, and a story the conditions to become a franchise, a pop-up, a fan theory, a Reddit thread still running years after the finale.

Which is why creative development and marketing need to converge earlier than they typically do. Not at the trailer brief. Not even at the kickoff meeting in the writers’ room. At the idea itself, and as an ongoing conversation rather than a handoff that happens once the creative work is considered done.

This may be the most controversial argument in this piece. Marketing sitting inside the creative process from day one carries real risks, the fear that commercial instinct contaminates creative vision before it has had time to form. That fear is not unfounded. But it rests on a false premise: that world-building and audience-building are sequential rather than parallel. The conversation being proposed here is not “will this work?” That question poisons the well.

It is something closer to: given what we are building, who else might find a door into it, and have we built enough world that they will find something real once they arrive?

That is a different tone and a different mandate entirely. Not validation. Excavation.

The question is not “will audiences like this?” It is: have we built something with enough internal coherence that it can mean different things to different people without falling apart?

When the answer is yes, the campaigns stop being promotions. They become proof of concept.

For Severance Season 2, Apple TV+ placed the cast inside a glass-walled replica of Lumon Industries at Grand Central Terminal, actors performing in character for two hours while cryptic Lumon-branded emails hit subscribers’ inboxes the same week.

For The Drama, A24 placed a fictional engagement announcement in the Boston Globe, then opened a functioning wedding chapel in Las Vegas so fans could walk into the film’s world before it opened.

The press covered both as news. That is the tell. When a world is built completely enough, the marketing and the storytelling become indistinguishable from each other. From Fight Club’s fake PSAs in 1999 to Lumon colonizing Grand Central in 2025, the logic is the same: not a promotion for a story, but an extension of one.

None of that is manufacturable after the fact. It has to be designed in.

Which returns everything to the same question it always returns to: does this story say something that has not been said yet? Does it name a feeling already alive in the culture, waiting for someone to give it form?

The titles that last are the ones where someone answered yes before there was any evidence to justify it. Someone in a room, with taste and conviction and enough professional scar tissue to trust their instincts, decided to believe it.

Fight Club is still here. The rules still apply.

I’ll be transparent: I could be directly affected by the outcome of the merger this essay examines. That proximity is worth naming, because it shapes why I wrote this the way I did. It would have been easy to write something angrier. The numbers invite it.

But anger at individuals misses the more durable problem. With yet another merger looming above us, my intrigue kept growing to understand:

What keeps producing these outcomes is not a shortage of cautionary tales or a surplus of bad actors?

It is the M&A cycle itself, and the near-total absence of accountability for what a merger leaves behind? Deals get structured, approved, and rewarded in ways that treat the industries they move through as abstractions rather than ecosystems with their own economies, labor forces, and long-term logic.

That is what this essay is really about: not who is to blame, but what keeps the cycle turning, and what it would take to interrupt it.

The most seductive quality of a vision deal is how reasonable it sounds in the room where it is made. In January 2000, when AOL and Time Warner announced their merger, they were not lying. They were describing something real: a future in which the internet’s reach would be married to Hollywood’s depth.

That future did arrive.

It simply arrived through Netflix, YouTube, and a generation of builders who had no legacy infrastructure to protect and no integration politics to navigate. The $99 billion write-down that followed was not merely a financial catastrophe. Think of it as an unintentional clearance, a making of space in which an entirely new media order took root and grew.

But here is what that history reveals, and what the current WarnerMount deal repeats: the failure was not a failure of vision. It was a success of incentive. The deal was designed to reward itself, not what it produced. That distinction is the key to understanding why the same story keeps being told with the same assets, the same logic, and strikingly similar outcomes, three times in less than a decade, with nearly $80 billion in debt now attached to the fourth attempt.

The question this moment demands is not whether Gerry Cardinale, the managing partner of RedBird Capital and chief architect of the Paramount bid, has studied the right cautionary tales, or whether David Ellison is sufficiently clear-eyed about the TV decline curve. It is whether the people inside these institutions will be given the conditions to do something worth doing.

This year, Hollywood entered Oscars weekend with guild employment down 35 to 40 percent and a generation of agency assistants quietly debating whether the movie business will still exist by the time they reach it, the answer to that question carries weight that extends far beyond a spreadsheet.

The Three-Deal Pattern: Same Assets, Repeating Cycle

To understand why this pattern keeps repeating, it helps to stop thinking about these deals as isolated business decisions and start seeing them as chapters in a single, ongoing story where the plot points are almost identical and only the names and numbers change.

In 2018, AT&T’s CEO Randall Stephenson bought Time Warner for $109 billion, a bet so large, it made AT&T the most indebted operating company in America overnight, carrying more borrowed money than any non-bank in the country. Stephenson was not reckless in his own mind. He saw, correctly, that the future of media would be won or lost on the strength of content libraries and distribution reach.

What he underestimated was the cost of being right about the destination while being catastrophically wrong about the route.

The stock never recovered to where it was the day the deal was announced. While the broader market rose roughly 30 percent over the same period, AT&T’s shares fell. A major ratings agency downgraded its debt. Stephenson retired in 2020 with $16.3 million in final compensation plus a $12 million consulting exit agreement, the customary arrangement by which a person is paid to remain available for advice about decisions that did not work out. Two years later, the assets he had spent $109 billion to acquire were sold to Discovery for roughly $43 billion, about half of what he paid. The market, to its credit, did the math.

The assets then passed to the CEO of Discovery, David Zaslav, who stood before investors in 2022 and promised that the merger of Warner Bros. Discovery would generate $14 billion in annual profit by 2023.

The stock fell of course, and in 2024, the company posted an $11.5 billion net loss.

And now the same assets are changing hands a third time, at $111 billion, carrying $79 billion in debt, with a promise that $6 billion in cost cuts will unlock $18 billion in annual profit and finally produce a streaming platform capable of competing with Netflix. As Gerry Cardinale acknowledged to Puck in March 2026, on the surface that amount of debt is a large amount of debt for any company. His defense is that the combined cashflow is also unusually large. The cold math of whether those two facts can be reconciled will determine everything that follows. It is, at minimum, the third time someone has been confident it will.

How M&A Rewards the Deal, Enriches the Few, and Abandons the Aftermath

The pattern is not incidental.

A landmark study in the Journal of Finance examined 370 mergers and found something that should stop us cold: even in deals where the acquiring company’s stock underperformed the broader market by roughly 52 percent, three-quarters of the CEOs involved ended up personally wealthier than before the acquisition.

A companion study from the University of British Columbia confirmed the asymmetry: after a merger, CEO pay is largely unaffected by poor stock performance but rises in step with positive performance. And a Journal of Financial Economics study found that what actually predicts the size of a CEO’s merger bonus is not how well the deal performed. It is how much power the CEO has over the people deciding the bonus.

Three deals. Three generations of executives. Three sets of promises that arrived with fanfare and departed with write-downs. The assets kept moving. The debt kept growing. And through it all, the executives kept getting paid, not in spite of the outcomes, but in complete indifference to them.

The deal, in other words, functions as a one-way bet. Heads, the executive wins. Tails, the executive wins less, but still wins.

From 1978 to 2023, top CEO pay grew by over 1,000 percent, compared with roughly 24 percent growth in typical worker compensation over the same period. That gap does not reflect a commensurate expansion in value creation. It reflects, in significant part, the proliferation of this architecture: deals that reward scale, deals that reward completion, deals that reward the act of acquiring regardless of what the acquisition actually produces.

None of this is illegal. That is precisely the point.

The problem is not bad actors making cynical choices. It is a well-functioning machine producing entirely predictable outputs that happen to be misaligned with the health of the companies, the industries, and the people working inside them.

How Debt Rewrites the Creative Playbook

Here is something that rarely makes it into merger press releases: debt has a creative opinion.

It is conservative, risk-averse, and allergic to the unconventional. When a studio is carrying tens of billions in borrowed capital, the question of whether to green-light something gets quietly replaced by a different question: Can the company even afford to green-light it?

Those are not the same question, and the distance between them is where culture goes to die.

Think of it this way: a healthy studio operates like a diversified portfolio, with some franchise bets, some mid-budget gambles, and some genuinely strange swings that occasionally become cultural moments. A heavily indebted studio operates like a company in survival mode, doubling down on what it already knows works, because the cost of being wrong has become existential. Franchise sequels. Legacy IP. The commercially legible over the artistically necessary.

The creative middle class is always the first casualty of this shift.

Not through any single dramatic decision, but through a thousand small ones: the green-light that does not come, the development deal that does not renew, the executive with taste who loses the argument to the executive with the spreadsheet. Consolidation does not kill creativity through malice. It kills it through math.

When Warner Bros. Discovery formed in 2022, the company’s first creative act was effectively an act of demolition. A completed Batgirl film was scrapped. Westworld was cancelled. Dozens of series were axed, not because audiences had rejected them, but because the balance sheet needed the write-down. Content impairment charges reached $3.5 billion in the first year alone. The CFO described the carnage as the “rectification of content exuberance.”

That is what the finance department calls creative ambition when the debt comes due.

The numbers now document what that creative compression has cost. FilmLA reported a 16 percent decline in shooting days in Los Angeles County in 2025, down nearly half from the recent peak in 2018. The Directors Guild of America president, Christopher Nolan, reported guild-member employment down 35 to 40 percent at a February press conference, calling it a very worrying time for the industry.

Industry tracker ProdPro found that sound technicians and dolly grips were averaging six months of downtime between productions. Studios are spending less on film projects since the 2023 actor and writer strikes, with programming budgets being reallocated toward sports and live events, a direct substitution of creative risk for the safety of pre-sold audiences.

The contraction has now reached the exhibition layer.

Multiple theater chains, including Cineworld, Alamo Drafthouse, iPic, Metropolitan, CMX, and Look Dine-In, have all filed for bankruptcy in the past five years. AMC Entertainment, the world’s largest cinema chain, has lost 99.7 percent of its value since its stock hit an all-time high in June 2021. The chain is not collapsing in isolation. It is collapsing because fewer diverse films are reaching screens, and fewer diverse films are reaching screens because debt-burdened studios can no longer afford to be wrong.

HBO, the same network that once greenlit The Wire without a guaranteed audience, is now producing a Harry Potter series. Pixar, the studio that built its reputation on original worlds, is releasing Toy Story 5. These are not creative decisions. They are financial ones dressed in creative clothing.

When the cost of being wrong becomes existential, no one chooses to be wrong. And so the entire system tilts, slowly but unmistakably, away from the new and toward the familiar, not because audiences have stopped wanting to be surprised, but because the institutions capable of surprising them can no longer afford to try.

The Damage Runs Deeper than Content Slates

What these deals produce, almost without exception, is a particular kind of organizational uncertainty that does not announce itself in a press release but spreads quietly through every closed-door conversation of a company mid-merger.

It is the uncertainty of people who cannot answer the most basic questions about their own lives: whether their role will exist in six months, whether the work they did this year matters to anyone making decisions.

Consider what Paramount employees were navigating in the days immediately following the WarnerMount announcement. The company told eligible employees that their 2025 year-end bonus multiplier had been cut to 94 percent, down from 136.7 percent the prior year, a decline of more than 30 percent. More pointedly, it removed individual performance differentiation entirely, assigning every employee the same multiplier regardless of how they had actually performed.

In a single memo on March 2nd, the company delivered two messages simultaneously: we had a worse year, and it no longer particularly matters how you personally performed. For anyone who had stayed through prior rounds of uncertainty because they believed their individual contribution was visible and valued, this was not merely a financial adjustment.

It was a statement about what the institution values during transformation.

Then consider what the executive suite was signaling at that same moment.

Then on March 3rd, WBD CEO David Zazlav, who presided over billions in losses filed to liquidate $114 million in personal stock holdings in the company he was nominally stewarding forward. No employee needs a financial analyst to interpret that signal.

It is immediate and instinctive: if the person at the top is cashing out, what exactly are the rest of us being asked to believe in?

The Talent Equation Nobody Accounts For

The research on talent attrition during mergers is consistent: high performers leave first. They do not wait for a layoff notice. They read the signals, and they begin, quietly, to make other plans. Entertainment and media layoffs were already up 18 percent in 2025, with over 17,000 jobs cut across the industry, before WarnerMount’s $6 billion cost mandate had even begun to land.

The financial cost of that broken compact is larger than most merger models ever acknowledge. Gallup estimates voluntary turnover costs U.S. businesses roughly $1 trillion annually. Replacing a single employee runs between half and twice their annual salary; for executive roles, that figure climbs to 213 percent. Two-thirds of those costs, according to SHRM, are intangible: lost institutional knowledge, fractured team dynamics, the informal systems that make organizations actually function.

A Wharton study tracked 50 million smartphones over four years, tracing each device to the exact factory conditions of its assembly. Each percentage-point increase in weekly worker turnover raised product failure rates by nearly 0.8 percent. Even in a highly standardized manufacturing environment, losing experienced workers degraded what the company produced. Now apply that insight to an industry where the work is, by definition, not standardized, where what gets made depends on relationships, accumulated judgment, and a shared sense of what the company stands for.

Talent is not a line item.

The institutional knowledge that walks out when uncertainty becomes intolerable, the creative relationships that dissolve when slates shrink, the experienced executive who quietly decides the risk is no longer worth it: these losses do not appear in an EBITDA forecast. They appear later, in the thinness of what gets made, in the sameness of what gets greenlit, in the slow hollowing of the creative culture that was, in theory, the entire reason the acquisition was worth making.

The Bonuses Reward Cost Cuts for Productions, The Business Card Says “Content Company”

To understand the scale of what that means in practice, consider a single year’s worth of executive compensation at Warner Bros. Discovery. In 2024, the five named senior executives collected a combined $123.2 million: Zaslav at $51.9 million, Bruce Campbell at $19.8 million, JB Perrette at $19.7 million, Gunnar Wiedenfels at $17 million, and Gerhard Zeiler at $14.8 million.

So I couldn’t help but wonder…

I calculated the hypothetical: if each executive had been capped at $10 million, the company would have retained $73 million in a single year. Enough for a one prestige drama series, three to four mid-budget films, or twenty writers’ rooms. In one year, from five people.

When WBD restructured executive pay in 2023, it tied Zaslav’s bonuses directly to free cash flow and debt reduction. This precisely rewards the behaviors that hollowed out the creative slate: cost cuts, cancelled projects, and workforce reductions. To be fair, some of that debt was not of Zaslav’s making. He inherited a balance sheet AT&T had already pushed past its limits.

But inheriting a fire and being paid handsomely to let it burn are not mutually exclusive conditions.

In 2024, his compensation rose to $51.9 million. Nearly 60 percent of shareholders voted against it. The vote was advisory, the corporate governance equivalent of a strongly-worded handwritten letter to someone who has opted out of snail-mail. It changed nothing, which was, strictly speaking, within the rules.

Between 2021 and 2024, Zaslav collected over $390 million while the company missed its own EBITDA projections by $5.3 billion, cancelled finished films, and shed tens of thousands of jobs.

This Linked-in Post by just proves the point:

This is not a story about one executive’s character.

It is a story about a system that has quietly divorced performance from reward while maintaining the paperwork that suggests. Moreover, it’s a performance review where the criteria shift to match the outcome, and the outcome always finds its way to the same number.

The gap between what was promised and what was delivered was not absorbed by the compensation committee. It was absorbed by the people who made the work.

Boards have the tools to do this differently. They can clawback provisions that activate when promised synergies fail to materialize, compensation tied to long-term outcomes rather than deal completion. None of it requires new legislation.

What it requires is something considerably harder: compensation committees willing to impose consequences on people whose professional circles largely overlap their own. These are not villains twirling mustaches over quarterly write-downs. They are peers, operating inside the same incentive architecture, making the choices that architecture consistently rewards.

The problem isn’t moral failure. It’s a closed loop, and closed loops have a well-documented allergy to self-correction.

Asking a compensation committee to cap executive pay is a bit like asking a dinner party to vote on whether to end the dinner party. The motion rarely passes. The wine keeps flowing. And somewhere in the balance sheet, the creative middle class picks up the tab.

That’s not a mystery. It’s the whole design.

The Root Cause of Hollywood’s Recession: When Wealth Stops Moving, Everything Stagnates

In finance, market makers are the institutions that keep money flowing. Their entire business model depends on circulation, not stasis. Media, at its healthiest, works the same way. The most generative periods in Hollywood history were not produced by the studios with the largest balance sheets. They were produced by the ones with the highest velocity of creative bets, where capital moved fast enough, and decision-makers felt secure enough, to be wrong occasionally.

The Warner Bros. of the 1970s green-lit The Exorcist and All the President’s Men alongside mainstream blockbusters. HBO in the early 2000s spent lavishly on serialized drama before any algorithm could have predicted an audience for The Sopranos, because that audience did not yet know it wanted one. Mid-period Netflix green-lit so promiscuously across genres and geographies that it accidentally invented new content categories entirely.

In each case, the breakthroughs shared a single underlying condition: money was moving, and the people deciding where it went had enough security to be wrong.

Then the debt arrived.

What $79 billion in debt does to a studio is structurally simple, even if its consequences are not: it converts a market maker into a holding pattern. Capital that once circulated stops moving. It pools at the top, servicing obligation. The bets narrow. The franchises multiply. The meeting never happens.

A creative ecosystem, like a coral reef, requires tidal flow. When the major studios collectively carry hundreds of billions in debt and funnel available capital toward servicing that obligation rather than circulating it through the ecosystem that produced them, the structure remains visible. The life within it calcifies.

The One Thing Debt Can't Acquire

Here is what is genuinely hopeful about this moment: consolidation has never succeeded in consolidating creativity. In every cycle, it has tried and failed in the most instructive way possible.

The AOL Time Warner collapse did not kill internet-distributed media. It cleared the ground for Netflix, YouTube, and a generation of builders unburdened by the politics of a $350 billion integration. The AT&T WarnerMedia era did not suppress prestige television. HBO kept making extraordinary work because its creative culture proved stronger than its corporate parent’s identity crisis. The pattern is consistent, if not quite a law: when incumbents compress under the weight of their own debt, space opens at the edges.

There are early signs this is happening again. In February 2026, Iron Lung — a $3 million film by YouTube creator Mark Fischbach — took in $51 million at the box office by mobilizing an audience the major studios no longer know how to reach. One film is not a movement. But it is a signal worth reading: creator-native work, built outside the studio system, finding cultural traction precisely where the center has retreated.

The audience side offers a more tentative counterpoint. The National Research Group found that 59 percent of post-2013 audiences prefer the theatrical experience, a higher rate than the generations immediately before them. Whether studios give them something worth showing up for is the open question.

Which is, ultimately, the only question that matters. Not whether WarnerMount succeeds on a spreadsheet, but whether the people inside it will be given the conditions to do something worth remembering. And whether this consolidation, like every one before it, quietly opens a door its architects never intended to unlock.

The Default Is Not a Destiny

There is a particular kind of learned helplessness that settles into industries after enough cycles of the same story. The deals get made. The promises dissolve. The write-downs come. The executives exit wealthy. And somewhere in that repetition, an assumption hardens into consensus: this is simply how it works.

Consensus is not correctness. Three generations of executives described futures that did not materialize and departed with compensation that bore no relationship to what they left behind. Shareholders protested through the only formal mechanisms available and were ignored. Thousands of jobs evaporated while the architects of that evaporation negotiated their exits.

That shrug deserves examination. What it conceals is not inevitability. It is a choice.

The machine is real. The incentives are real. The pattern is real. But so is the capacity, inside every one of these organizations, to make work that outlasts the deal that funded it. The institutions that emerge from consolidation cycles with their creative cultures intact share one consistent trait: they treated their people as the constant, not the variable.

The deal is designed to reward itself. What happens to everyone else is what we choose to make happen.

Then the Citrini 2028 Global Intelligence Report landed on February 22nd. It read, to me, like science fiction wearing the clothes of research. It was precise enough to feel authoritative, sweeping enough to feel inevitable, and just unsettling enough to refuse to leave. Days later, Citadel Securities released the 2026 Global Intelligence Report which felt like a quiet, yet well articulated rebuttal: more grounded, more honest about what the data actually shows and doesn't.

Reading both in succession, I kept returning to the same image — the grad students I used to teach at Columbia, the coffee chats I’ve had with all the students graduating this spring, holding AI and disruption not as theoretical frameworks but as live variables, trying to build careers on ground that keeps shifting beneath their feet.

That image wouldn’t leave me alone. So I wrote this, not as an expert dispensing advice from a safe distance, but as someone who empathizes with the feeling of uncertainty and is trying to hand back something useful rather than something frightening. I wrote it as if I were the one receiving it.

We’ll return to media chaos shortly.

I’ve been out of college for a few years now. Getting here required an identity crisis or maybe two. I came in having run my own startup, been my own boss, treated my work as an extension of myself in the way that early-stage environments quietly demand. Joining corporate America felt a little like getting married too young. But it gave me something I didn’t expect: the ability to hold my identity separately from my job title. Startup culture, for all its virtues, rarely teaches you that. The health insurance doesn’t hurt either.

What’s kept me thinking, though, is a pattern I keep observing. Across every professional context I've inhabited, the same fault line keeps appearing. It showed up when I served on faculty at Columbia, in guest lectures at Boston University, in intern office hours at Warner Bros. Discovery, in work with BU's Women's Leadership Council, and in conversations with hundreds of students navigating the distance between their education and the workforce waiting for them.

The gap is consistent enough to feel structural: between how students are being prepared and what the job market actually rewards right now. This essay is my attempt to close it, at least a little.

First: Don’t panic. But don’t sleepwalk, either.

The loudest voices online would have you believe AI has already made your degree worthless before you’ve even used it. The data tells a more complicated story. Software engineering job postings are up 11% year over year as of early 2026. Data from the St. Louis Fed shows that actual day-to-day AI usage in the workplace has remained largely stable.1 The displacement apocalypse being sold to you on social media is not what’s happening in the labor market and I think it’s worth saying that clearly, because anxiety makes for bad decisions.

Every major technology, from the PC to the internet, followed a gradual S-curve of adoption, not an overnight explosion. Before Microsoft Office arrived, people feared it would replace office workers. Instead it made them more productive and created entirely new categories of work. AI is more likely to follow that pattern than the sci-fi version we've been sold, and there's a structural reason for that beyond historical analogy.

Displacing white collar work at scale requires orders of magnitude more compute intensity than current utilization levels. As automation expands, demand for that compute rises with it, driving up its marginal cost until a natural economic boundary forms: the point at which replacing a human simply costs more than keeping one. Even recursive improvements in algorithmic capability run into the same physical reality that data centers, energy grids, regulatory timelines, organizational inertia. The intelligence may scale. The infrastructure around it doesn't.2

That said, something is shifting, and it’s worth sitting with honestly.

Output expectations are rising. The people who learn to use these AI agents to think bigger, not just faster. They will pull ahead of those who don’t. Understanding which side of that line you’re on is probably the most important career question at this time.

That’s not a threat. It’s an invitation.

The degree still matters, just not for the reason you think.

For decades, a college degree signaled a specific, credentialed skill set. You learned the thing, got the paper, the paper got you the job. AI has quietly disrupted that equation. If what you’re selling is the ability to follow instructions, organize information, and execute predictable tasks, a machine now does that faster and cheaper.

Here’s the thing about school that nobody really says out loud: the system trained us to be great order takers. Show up, follow the syllabus, hit the rubric, get the grade. That was the job. And for a long time, it translated reasonably well into the workforce. It just doesn’t map as cleanly anymore.

The new question isn’t whether you can execute the task. It’s whether you can figure out which task is worth doing in the first place.

Think of how corporations signal credibility. A compliance certification, being listed on some Bloomberg index, these aren’t proof of excellence so much as proof of a threshold crossed. A signal that says worth a second look. Your degree works the same way: a social passport that lubricates first impressions and eases entry into rooms that might otherwise stay closed.

But let’s be precise about what that means… and what it doesn’t.

The degree is an access point. A potential booster. At its best, it loosens the latch on a door that a relationship still has to open. Mistaking the access point for the relationship itself is where a lot of people get quietly stuck.

Here’s what’s changed and why it matters more now than it did ten years ago.

A decade ago, a recognizable degree could cut through the noise on its own. Today, that noise has become structural. Inboxes are flooded, LinkedIn is saturated, and AI has made polished, personalized outreach trivially easy to produce at scale. The credential that once distinguished you is now table stakes.

And yet, paradoxically, the access point your institution provides has become more valuable. Not because the brand means more, but because genuine human connection has become rarer. When a fellow alum forwards your name before a role is posted, when a relationship formed in a seminar room resurfaces years later as an open door, that’s not the credential working. That’s the relationship the credential made possible, doing the thing only relationships can do, and it’s the one thing AI cannot replicate.

It can optimize a resume, simulate the language of networking, surface an opportunity. What it cannot manufacture is the quiet credibility that accumulates when real people vouch for how you think.

A human recommendation will always outperform an algorithm.

We see the same dynamic everywhere: on TikTok, Instagram, every major streaming platform, the most powerful discovery signal isn’t the algorithm, it’s what someone you actually trust told you to watch. Platforms spend billions trying to simulate that experience precisely because they know they can’t reproduce it. Hiring is no different.

A university degree tells someone you’re worth considering. A genuine endorsement from someone who has watched you think, problem-solve, and show up that tells them you’re worth trusting. The brand gets you in the building. The people are what keep you there.

The rubric for getting promoted is not the rubric you learned in school.

This one surprised me when I first experienced it, and I think it’s worth being honest about.

In a startup, the question is “can you get me those numbers?” In corporate America, it’s closer to “can you play nicely with others?” You are more likely to get promoted when your product fails but your colleagues love working with you, than when you ship the best-performing feature and everyone dreads your calendar invites. That’s not cynicism, it’s just the actual system, and understanding it early is genuinely useful.

This is where soft skills stop being soft. They become the thing that is genuinely irreplaceable about you: emotional intelligence, the ability to make people feel heard, knowing how to read a room and make someone want to work with you again. Not nice-to-haves. The actual moat.

And here’s an important nuance worth sitting with: even as AI capabilities improve, economic adoption doesn’t follow the same curve.

Organizations are slow. Integration is costly. Trust takes time to build. The tasks that are hardest to automate are almost always the ones that are fundamentally relational. That’s not an accident, it’s a structural feature of how humans actually work together. It’s something to lean into, not take for granted.

A note on being a student, right now.

Here’s something I don’t think gets said enough: being a student is one of the few windows in your life where the world is genuinely not expecting you to have it figured out. And that is a gift, even when it doesn’t feel like one.

If you walk into a room at the age of 22 with total certainty about exactly where you’re headed, the most experienced people there are quietly thinking, give it two years. That’s not an insult, it’s just pattern recognition from people who’ve lived it.

And I say that with real compassion for that twenty-two-year-old, because the certainty isn’t their fault. We’ve been oversold on it. Society rewards the person who projects confidence and knows the answer before anyone asks. But certainty is a trap as much as it’s a comfort. It optimizes for the path that already exists, and makes it harder to notice when a better one appears.

The people who learn to savor uncertainty, and to treat new information with curiosity rather than anxiety, tend to be the ones who adapt well as the world changes around them. Right now, while nobody is expecting you to have the answers, that uncertainty is working in your favor. It’s worth savoring, even a little.

Use this window to find the things that genuinely make you curious. The things that make you forget to check your phone. Don’t dismiss something because it doesn’t seem professionally relevant. You’d be amazed at the multidimensionality of the people you’ll eventually want to work alongside: the finance person obsessed with theater, the engineer who makes ceramics, the product manager quietly co-writing a TV series. People with professional careers also have interior lives. That’s often how real connection happens.

Whatever project comes out of your curiosity becomes your portfolio, proof that you can think beyond a tool, not just operate one. Passion, follow-through, and the willingness to keep showing up: those are harder to replicate than any technical skill.

Every tool in the world can amplify what you bring. None of them can manufacture why you care.

Following your passion might be the most authentic way to build your network in the age of AI… here’s why:

If you’re actively looking for a job, I highly recommend you stop cold-applying into the void and start building something. A YouTube channel. A research paper. A side project. A creative experiment. Whatever it is, do it in public and use it as your reason to reach out to people.

Two examples from my own life, because the abstract version of this is easy to dismiss.

When I was building my startup as a full-time student, I knew almost nothing about marketing, software development, or finance. So I used the project as a reason to reach out to people who did. I needed help, and I said so. That honesty gave people a reason to engage and gave me a reason to follow up. By the time I wound the company down, I had built a network I never would have had otherwise. Not because I was especially well-connected, but because the project gave every conversation a real purpose.

Years later, the same principle played out in a completely different context. I had no screenwriting background. But I had an idea for a television series, and I used that project as my reason to learn, ask questions, and build relationships with people who knew far more than I did. Two years later, that project has two incredible co-creators who wrote the pilot with me, a well-connected executive producer helping us get meetings, and an entertainment lawyer who took us on contingency. None of that happened because I had the right credentials. It happened because a real project gave me a genuine reason to connect with real people.

That’s the model. Not networking for its own sake, a project you actually care about, used as the vehicle to build relationships with people who can see how you think and how you work.

How economic trends favor following your curiosity.

Here’s the broader context worth holding onto, and it’s actually kind of hopeful: if rising productivity from AI does what productivity gains have historically done — lower costs, expand real incomes, give people more time — we’re likely heading toward more space for the liberal arts, for pursuits that once seemed impractical.

Even in the Citadel Securities report, they notably mention how Keynes predicted a 15-hour work week in 1930. He was wrong about the timeline, but what followed was an enormous expansion of what people chose to spend their time and money on. More industries. More niches. More ways to build a life around what you actually care about. We may be at the beginning of something like that again. The students building skills and relationships around their genuine interests right now are not wasting time. They might just be early.

If you’re still not sure, think of it this way:

Some very tactical advice to anyone trying to expand there network.

Essays like this are possibly helpful, but I personally find myself at the end thinking: so what do I actually do now? Here’s what has genuinely worked for me and for the students I’ve been lucky enough to talk with.

How to have a good informational interview.

Think of informational interviews as a funnel. Volume creates surface area, and surface area creates luck. But here’s what most people miss: attention is the scarcest resource in the modern economy. AI has made content infinite and transactions frictionless, which means a real human giving you 30 unscripted minutes is quietly becoming one of the most countercultural acts in professional life. When someone offers you that time, they’re giving you something worth more than they probably realize. So are you.

Before the call: Come in honest, not rehearsed. You don’t need a perfect story, you just need a real one. Tell them where you are, what you’re trying to figure out, and if you’re hoping for a specific introduction, just ask. You’d be surprised how willing people are to help when you’re upfront about what you actually need. Do your homework, too and come with something real to explore, not questions you could have asked anyone.

In the conversation: Ask the things that Google can’t answer and ChatGPT can only simulate. What did they get wrong? What would they do differently? What actually makes someone succeed in that environment versus wash out? That’s the kind of intelligence that only lives inside a person.

The impression you leave: The questions you ask say more about you than anything on your resume. They signal how you think, what you’re genuinely curious about, and whether you’re someone worth investing in. If you walk away knowing a little more about what lights you up and what doesn’t, that’s the whole cake. Anything else is just a cherry on top.

A call to action for everyone.

If you’re a student: Find the thing you’re genuinely excited about, build something with it, and use that project as your reason to reach out. Be specific: ask to be introduced to hiring managers and directors in your target field, even if nothing is open. Those relationships compound over time in ways that applications never will. And start before your final year. The best time to build is before you need anything.

If you’re a hiring manager or senior leader: It’s worth being specific about why your role in this is different from anyone else’s.

A hiring manager isn’t just a networking contact. They’re structurally the closest human being to an actual job offer, the person who can decide, before a resume ever enters a system, that someone is worth a conversation. That proximity is rarer than it sounds. Most early-career candidates never access it until they’re already desperate, submitting into applicant tracking systems designed to filter, not discover.

Thirty minutes of honest conversation, followed by a door left open, is not a small thing. In a hiring landscape increasingly mediated by algorithms and keyword filters, a human being who already knows a candidate’s name before the resume arrives is an almost unfair advantage. You may not remember the exchange six months later. They will carry it for a decade. The relationship is theirs to maintain, that’s on them, not you. But the door is yours to open, and it’s one of the few that actually leads somewhere.

If you’re an academic institution: The reframe worth making is this: your alumni network is not just a fundraising asset. It’s placement infrastructure, and most institutions are leaving it almost entirely unused.

Career fairs and panels create ambient exposure. Ambient exposure doesn’t convert. What converts is a student who enters their final year already in a genuine relationship with a hiring manager in their target field, someone who has had the conversation, followed up, and built enough credibility that a referral feels natural rather than transactional. That pipeline doesn’t assemble itself, but the institutions that build it deliberately, with the explicit goal of connecting students to hiring decision-makers before graduation, will produce the placement outcomes that every other metric is ultimately trying to approximate.

The relationship is the shortcut. Everything else is theater.

For the first time since Nielsen began tracking streaming data in 2020, not a single new show cracked the top ten most-watched of 2025. Not one. Based on Bloomberg’s analysis, what filled that vacuum instead was Bluey (2018), Grey’s Anatomy (2005), Squid Game (2021), and Stranger Things, now in its ninth year. And yes, this is what audiences chose when given access to the entire history of television at their fingertips. If your instinct is that audiences have simply lost their appetite for novelty, you’re reading the symptom and missing the disease.

The appetite for new stories is real and enormous. Social media proves it every week, with fandoms erupting around cultural moments worth dissecting. But somewhere between the promise of infinite content and the reality of what actually gets watched, something broke.

Platforms spent the better part of a decade pouring billions into new originals — each one engineered to dominate the conversation for a single weekend, then quietly fade. It’s the entertainment equivalent of planting flowers that bloom for two days and then pulling them before they can take root. What grew in their place were the shows nobody was actively betting on: the long-running catalog titles built for a slower, steadier era of television, when stories had room to breathe and audiences had time to find them.

When you flood a garden with flowers designed to peak and disappear, don’t be surprised when people keep returning to the oak tree. So if audiences genuinely crave new content, yet shows from 2005 keep winning, the question worth asking isn’t what’s wrong with the audience. It’s what’s wrong with the model.

The Framework: If Streaming Platforms Are like Stock Exchanges, then TV shows are Stocks

Like any exchange worth its trading floor, platforms are perpetually hunting for the hottest new listings: the prestige drama that triggers a cultural moment, the breakout original that moves the market. More listings, more activity, more data generated from every transaction.

But here’s where the analogy breaks down in practice and the economics quietly unravel:

To understand why the economics are broken, we should stop thinking about content as just movies and television series.

Every title in a streaming library is a financial instrument, and its value depends entirely on what it was designed to do and whether the platform holding it understands the difference between activity and return.

While I realize this sounds sterile, this mindset is intended to be applied on the corporate side of media when thinking through things such as IP licensing, acquisitions and content strategy.

Below I’ve outlined how I’d view different tv shows as if they were stocks.

What this reveals is the fundamental portfolio error streaming platforms have repeated at scale: building a market almost exclusively around growth stocks, then expressing shock when the volatility proves unmanageable.

Catalog depth was doing the quiet unglamorous, essential work all along. It was a rewatch-friendly foundation of any durable library.

Why Familiar Content Wins… it’s Not that Deep

Grey’s Anatomy didn’t dominate streaming because audiences stopped wanting something fresh. It dominated because it was built like Coca-Cola in an era when everyone else was trying to launch the next Rivian.

It was designed for broadcast: 22 episodes per season, released weekly over nine months, engineered for ritual and sustained emotional investment. You could discover it in season four and catch up at your own pace. Syndication created decades of rewatch revenue. It was architected, economically and narratively, for infinite return visits.

In a prior article I wrote about the “hidden cost of gratification” which explains how the customer journey of binging content versus weekly drops over a span of months yields an entirely different cultural impact.

The streaming-native model inverted every one of those principles.

Releasing all eight episodes simultaneously optimized for binge completion over a single weekend. The cultural conversation peaks Monday and dies by Thursday. There’s no discovery window, no sustained ritual, you either watched it opening weekend or you effectively missed it. Platforms designed growth stocks to behave like penny stocks: one earnings call, then delisted forever. Of course audiences fled back to the dividend payers.

The “Blitzscaling” Era of Streaming is Over

History has a uncanny way of repeating itself in other sectors, hence- this reckoning has a precise historical parallel in tech not too long ago.

The 2010s model was straightforward: raise millions, burn cash for growth, subsidize users with cheap rides and free software, and promise profitability “later.” Until the early 2020s, Millennials inhabited a lifestyle subsidy sponsored by venture capital to fuel astronomical growth rates.

Then the interest rate environment shifted, growth-at-any-cost ended, and unit economics became the only metric that mattered. The subsidy era collapsed and Uber finally makes money by charging what rides actually cost.

Streaming is living through an identical correction. From 2015 to 2020, every major platform was spending billions annually on originals, green-lighting expensive bets on the implicit promise that profitability would come later. It didn't.

Now platforms must prove content ROI while simultaneously lacking the measurement infrastructure to know which content actually earns it.

They’re cutting budgets blindfolded.

The Content Portfolio Problem

Netflix’s model is not irrational, it’s blitz-scaling applied to the zeitgeist. They lead in subscribers, dominate minutes-watched globally, and have made “Netflix” a verb in the same way Google once did.

The math they initially were optimizing for is: profit follows when viewing time scales faster than the cost of content.

But attention captured is not the same as value compounded. The VC model works when growth stocks eventually mature into dividend payers. When Netflix spends $17 billion annually commissioning originals measured by opening-weekend views, then canceling under performers before they can season into catalog assets, and immediately rolling that capital into the next moonshot, it is not building a portfolio. It is running an infinite series of isolated wagers dressed in the language of strategy.

This is compounded by binge culture itself: a show consumed in a weekend becomes a memory by the following one, with none of the slow accumulation of meaning that weekly appointment television once generated. Without stable, evergreen programming beneath the originals, the dividend payers absorb losses and generate baseline viewing. And as a result, every expensive bet is a naked gamble with no portfolio to break the fall.

Here’s this cycle in practice: a platform commissions a $200 million drama, measures success by opening-weekend views, cancels it when it doesn’t immediately perform, then commissions the next multi million dollar bet. Then it repeats until billions have been spent annually with nothing compounding beneath it. It is structurally a venture capital firm that only makes moonshot bets and never builds sustainable companies.

Here’s a couple of examples:

1. Mindhunter

David Fincher’s serial killer procedural built a devoted, critically acclaimed following across two seasons before Netflix quietly let it slip into “indefinite hold,” the streaming equivalent of ghosting. The reported cost per episode (~$10M) and Fincher’s exacting production pace made it economically inconvenient, so a show with genuine long-tail catalog potential was shelved before it could compound into anything.

2. Marco Polo

This show followed the same logic in reverse: a $90 million budget for a second season, canceled anyway because the numbers never justified the premium on Netflix’s internal benchmarks.

Views Are Vanity, Revenue Is Sanity, And Streaming Still Can’t Tell the Difference

Every piece of content generates signals far beyond viewing hours: social media mentions, search volume, fan-created content, merchandise demand, cultural conversation that functions as earned media. These signals have real monetary value. And they can be monetized through supplemental content, used to understand multi-platform engagement, or treated as data products revealing what drives cultural conversation at scale.

This is how sophisticated financial operators already work. Hedge funds pay millions for sentiment signals and alternative data that predict price movements before the market reflects them. The signals themselves are the edge.

Streaming could operate in a similar practice:

A show generates 100 million viewing hours plus 500 million social impressions plus two million pieces of fan content plus measurable merchandise demand plus ongoing discovery value eighteen months after release. Platforms are currently measuring one variable and ignoring the rest. Here’s an example of applying this concept to Sex and the City.

The structural consequence is legible in the data: free, ad-supported streaming (aka FAST) that are served on platforms like Tubi or Roku, is growing twice as fast as premium subscription services. When you can’t measure content ROI, ad-supported models work better because attribution is clean, it’s ad generated this revenue. YouTube has now surpassed Netflix and Amazon combined in U.S. television viewing share.

The platform with perfect attribution beats the platforms with broken attribution. This is not a coincidence.

The Uncomfortable Truth

Audiences want new content. The demand is unambiguous. But streaming platforms have engineered a model where new content structurally cannot be profitable, and the math was always going to catch up with them.

The economics are working against each other simultaneously: production costs have ballooned while release windows have compressed, leaving no room for the slow accumulation of meaning that turns a show into something people return to. You can’t build a dividend payer by designing everything to peak and disappear.

The fix isn’t a content strategy. It’s a measurement attribution one. I’ve been contemplating this for a while and here’s my recommendation of a new equation streamers could use to evaluate the valuation of content:

Right now, they’re optimizing the first variable and ignoring the other three. It’s like a fund manager tracking only trading volume while remaining indifferent to whether anyone is actually making money. That’s not a strategy problem. It’s an infrastructure problem; one baked into the organizational DNA of every post-merger media conglomerate, which I examined in depth in an earlier piece called “The Kryptonite keeping Hollywood Hostage Part II.”

The alternative is portfolio thinking. It’s less a reinvention than a return to a logic the industry already understood. Prestige dramas balanced by catalog depth. High-margin reality programming subsidizing expensive bets. Kids’ content anchors the whole structure like a foundation that never gets credit for holding everything else up.

When risk is distributed across asset classes rather than concentrated in a single breathless opening weekend, something quietly remarkable happens: content gets the chance to season into something people actually return to.

The future belongs to platforms willing to run a real stock exchange: growth stocks generating excitement, dividend stocks paying the bills, bonds providing stability, and measurement infrastructure sophisticated enough to capture value across the entire attention economy — not just watch time within a single platform.

That’s why Grey’s Anatomy, a show that premiered twenty years ago, is still one of the most-watched programs on television. It was never supposed to still matter. The fact that it does isn’t a nostalgia story. It’s a balance sheet, and it should concern everyone building the next one.

This shift crystallizes a question the industry has avoided confronting: If the future of entertainment resides in physical experiences rather than content creation, why do we continue developing prestige television as though streaming subscriptions, not theme parks, represent the ultimate revenue destination?

Why hasn’t HBO built a Game of Thrones theme park when Disney has already proven the model works?

The World-Building Deficit

The answer reveals a fundamental misalignment between how we develop content and how audiences increasingly value it. The entertainment industry faces a peculiar contradiction. We’ve mastered producing prestige television that demonstrates complex narratives with production budgets akin to major theatrical releases. Yet we’ve simultaneously abandoned the systematic world-building that transforms stories into enduring properties.

Consider the shows that generate sustained cultural presence: Harry Potter, Game of Thrones, The Lord of the Rings. Each originated from extensively developed fictional universes with internal logic, geography, and social structures. These weren’t stories that happened to include interesting locations; they were worlds that happened to contain compelling narratives. The difference is architectural.

Most development treats world-building as production design rather than foundational discipline. Writers’ rooms meticulously architect emotional arcs yet leave the physical and cultural ecosystems conspicuously schematic. For example, take the iconic coffee shop “Central Perk” in the sitcom Friends. This gathering place functions as narrative infrastructure for six friends without economic logic, neighborhood context, or cultural particularity beyond what any given episode demands.

This shift toward experience-driven revenue will reshape how creative teams approach development from the outset.

Writers’ rooms may soon find themselves tasked with a different kind of imagination: one that extends beyond narrative arcs into the material logic of fictional worlds. Interrogating whether Central Perk could sustain itself as an actual business, envisioning its menu evolution or relationship to shifting neighborhood demographics, becomes foundational creative exercise rather than marketing afterthought.

When studios begin evaluating IP potential during ideation, before writers’ rooms convene and before pilot scripts are commissioned, the criteria will expand beyond “Is this a compelling story?” to include “Is this a world audiences will want to inhabit?”

The execution may fall to marketing teams or licensing partners, but the creative foundation must be established upstream. Should the market later reward that depth with theme parks or immersive hotels, the infrastructure exists organically rather than as desperate retrofit.

The Revenue Realization Problem

This creative deficit cascades into economic consequence. Traditional media’s collapse has created uncomfortable silence around a fundamental question for legacy media survival: Where, precisely, will the money come from?

Cable’s decline eliminated the advertising foundation subsidizing content production for generations. Streaming generates revenue, but at far thinner margins. Netflix pioneered the cost-plus model, guaranteeing full production budgets while seizing worldwide rights for extended periods. Studios celebrate eliminating deficit financing, yet they’ve simultaneously surrendered the downstream revenue streams that historically generated actual profits: syndication, international licensing, ancillary markets.

As we continue to cycle through mergers and corporate breakups, media companies will be forced to choose between becoming pure content producers (the Sony model, creating IP to license) or building vertically integrated ecosystems extracting value across multiple touchpoints. The latter demands treating content not as finished product but as foundational infrastructure for extended commercial exploitation.

Can any company truly excel at both creating transformative IP and delivering five-star hospitality experiences at scale?

When Spreadsheets Speak Louder Than Scripts

Disney’s recent quarterly earnings don’t just answer this question; they reveal which bet the industry’s most sophisticated operator has already placed. Disney’s Q4 2025 earnings reveal the Experiences segment reached $10 billion in quarterly revenue, dwarfing Entertainment’s compressed margins. While content grapples with elevated production costs, parks delivered 8% higher domestic operating income. Even streaming’s respectable 8.4% operating margin pales against physical experiences generating billions with healthier returns.

This isn’t a coincidence. It represents a strategic culmination. D’Amaro’s appointment merely formalizes what capital allocation has long telegraphed: Disney is becoming an “experiences company” that happens to own content IP, rather than a content company that happens to operate theme parks. Content now serves primarily as intellectual property feedstock for the more lucrative business of selling immersive, physical storytelling.

These numbers raise an uncomfortable question Disney has answered but other companies still evade.

The Impossible Duality: Can any company master both generational storytelling and industrial-scale hospitality?

Disney has walked this tightrope longer than anyone, attempting to function simultaneously as both IP originator and experience curator. But these are fundamentally different competencies requiring divergent organizational cultures. Creating IP with sufficient depth demands obsessive attention to narrative architecture and environmental specificity. Operating immersive physical experiences demands an equally obsessive focus on hospitality operations and infrastructure maintenance. The creative mind that conjures a world and the operational mind that maintains one are rarely the same, and the organizational gravity that nurtures one often suffocates the other.

Disney’s positioning suggests they’ve already placed their bet. As reporter Julia Alexander articulates in an interview with Business Insider: “People will continue to spend because it is an experience that they’re not going to get elsewhere, which means that [Disney] can monetize the customer at a higher rate than they’re going to get in the content business.” The parks business isn’t just more profitable. It’s defensible in ways streaming never proved to be.

For companies unwilling or unable to master both competencies, an alternative model beckons.

The Licensing Alternative: Separation as Strategy

Warner Brothers has already demonstrated this alternative model’s viability through their licensing deal with Universal Studios for Harry Potter attractions. Universal operates the experiences; Warner Brothers collects licensing revenue without shouldering infrastructure costs or operational complexity.

This bifurcation (IP companies versus hospitality companies) may ultimately prove more sustainable than vertical integration’s false promise. A Game of Thrones themed boutique hotel in Dubrovnik, operated by a hospitality company but licensed from HBO, might generate better returns than HBO attempting to build that capability internally.

The Customer Lifetime Value Question

Disney’s customer-for-life strategy assumes seamless conversion between content consumption and physical experiences. The evidence suggests otherwise.

The question isn’t whether Disney+ subscribers or park visitors exist; both clearly do. The question is conversion efficiency between these channels. How many streaming subscribers translate into cruise passengers? Which audience segment generates the highest average revenue per customer, and should resource allocation reflect that reality?

This tension becomes visible in marketing expenditures. Streaming platforms routinely spend millions on immersive promotional experiences such as elaborate installations for season premieres and interactive pop-ups. But these investments are categorized as temporary acquisition costs rather than retention infrastructure. Netflix’s permanent Stranger Things installation in New York City represents a notable exception. The economics shift dramatically when amortizing costs across years rather than weeks, particularly when there’s sufficient crossover between fans who watch the show and demonstrate enough engagement to justify bringing that world to life.

The challenge is that streaming platforms have optimized for subscription metrics such as conversions, churn rates, and minutes watched, without developing competency in physical operations. This explains why most platform experiences remain temporary despite their proven engagement value. The skills required to program compelling content differ fundamentally from those needed to operate profitable entertainment venues consistently.

So Will There Be a Game of Thrones Theme Park After All?

The absence of a Game of Thrones theme park today reflects an industry that hasn’t reconciled its creative traditions with contemporary economic reality. D’Amaro’s ascension formalizes what the company’s capital allocation has long telegraphed: experiences, not content, represent Disney’s future revenue engine. The question facing Warner Brothers, Netflix, and every other content company is whether they’ll reach the same conclusion, and whether they’ll act on it before Disney’s head start becomes insurmountable.

The absence of a Game of Thrones theme park today reflects an industry that hasn’t reconciled its creative traditions with contemporary economic reality. We continue green-lighting prestige television using frameworks inherited from an advertising-subsidized broadcast model, even as that foundation crumbles.

For creators, this evolution presents both constraint and opportunity. The constraint is accepting that content value increasingly derives from properties supporting sustained engagement across multiple formats. The opportunity is receiving resources to develop fictional universes with unusual depth, creating environments rich enough to reward both narrative immersion and physical exploration.

For audiences, it means potentially reclaiming something lost in streaming’s fragmentation: spaces of collective experience where shared enthusiasm manifests in physical gathering.

Adult audiences deserve that experience. We’ve demonstrated our appetite for sophisticated narratives and willingness to invest emotionally in complex fictional universes. The question isn’t whether we’d visit Westeros or King’s Landing. The question is whether the industry will build worlds worthy of the visit, and more crucially, whether they’ll build them themselves or license that privilege to companies whose DNA is hospitality rather than storytelling.

The answer may determine not just where entertainment revenue comes from in the coming decades, but what kind of companies survive to create the stories worth experiencing.

It’s an understatement to say there’s “never a dull moment” working on the corporate side of a media company right now. The rise of AI has sparked both fear and excitement across the industry. News outlets spike their views with scarcity headlines about job elimination, while creatives grapple with something deeper: what happens to the future of art and storytelling itself.

I get paid to uncover what goes on inside the black box of algorithms, I work on building out the “Bloomberg Terminal of Streaming Analytics” to enhance optimizing content curation on a major streaming app. However, for months I’ve been thinking about the best way to describe AI’s impact on developing the content itself:

AI tooling for content creators is like Viagra. You still need the main part to perform on its own. Without it, the enhancement is useless. But with it? The results can be transformative.

I know this sounds provocative, especially in Hollywood, where “AI” has become synonymous with existential dread. But during my time in tech, from advising companies on AI before it was fashionable to now building tools that explain how algorithms decide what content gets seen (basically uncovering the black box).

So here’s my honest take: AI isn’t going to replace human creativity. In fact, understanding why it won’t tell us something fundamental about what makes great stories in the first place.

The Data Processing Difference

For tech folks, this will sound familiar. For everyone else uneasy about AI, here’s the key distinction: AI can simulate human responses, but it form those responses the way humans do and that’s the difference.

Humans gather data through lived experience which provides filters on how we process and evaluate situations around us: emotionally charged moments, filtered personally, accumulating into intuition, judgment, and creativity.

AI systems are also trained on data and process information, the difference is they don’t associate emotions with “data” the way we do. Feed a large language model enough text and it learns statistical relationships between words, concepts, contexts. With more data, those relationships become more precise, allowing the system to produce fluent, convincing responses.

But there’s a fundamental difference in how that data is processed and what it’s grounded in:

AI doesn’t form its own beliefs through “lived experiences” that shape it’s world view in the same way humans do.

It optimizes mathematical parameters to predict likely continuations based on patterns it has seen before. This enables impressive mimicry of reasoning, emotion, and creativity, but the process itself is probabilistic, not experiential. That’s why it also produced a racist AI slop episode of Seinfeld, which proves this point.

Human thinking isn’t separate from lived experience. Human characteristics such as emotions, memory, biology, provide context that’s inseparable from how we process information. Grief feels different depending on timing and circumstance. Creative leaps happen not because they’re statistically likely, but because something resonates, contradicts, or suddenly matters.

Despite breathless claims of sentience, these systems lack consciousness or subjective awareness. Humans aren’t purely rational processors, we’re creatures of intuition, contradiction, emotional resonance. And that’s exactly what makes human creativity different.

Some refreshing examples why human creativity will exist inevitably:

The Revised Ending of When Harry Met Sally: Originally, the ending was written where Harry and Sally don’t end up together. In an interview cited in the Hollywood Reporter, the director, the late Rob Reiner said, “I had been married for 10 years. I’d been single for 10 years, and I couldn’t figure out how I was ever going to be with anybody, and that gave birth to When Harry Met Sally. I met my wife Michele, who I’ve been married to now 35 years. I met her while we were making the film, and I changed the ending.”

You can program AI to have a cynical outlook on love and even prompt it and say have it simulate a story where they meet their perfect match, but for an AI agent to make autonomous emotionally creative driven decisions like changing the ending of a story.

Let alone, organically writing a line such as, “I came here tonight because when you realize you want to spend the rest of your life with somebody, you want the rest of your life to start as soon as possible".

Larry David’s process of taking mundane situations and thinking of the catastrophic outcomes: When the New Yorker published a profile on Larry David, he attributes his humor to the experiences of his upbringing. “It has to do with Brooklyn,” David said of his humor. “It has to do—I think—with growing up in an apartment, with my aunt and my cousins right next door to me, with the door open, with neighbors walking in and out, with people yelling at each other all the time.””

You can simulate experience in AI by inventing characters and feeding it memories. What you can’t simulate is the organic stitching of those moments into a meaningful origin story; one shaped by lived continuity rather than design.

What I realized when I started Screenwriting

About two years ago, I wanted to understand the other side. I had a story to tell and screenwriting somehow seemed like the logical answer… maybe it was the fact that I paid my bills in exchange for understanding how the algorithm serves content to consumers like you and me.

My co-creators and I spent a good year and a half developing a tv show series concept and pilot script. I even took a screenwriting class to learn the technical craft that informed me about proper pilot structure, dialogue formatting, the whole thing.

The writing and original ideas were entirely human, created by my co-creators and me. But AI helped us generate ideas to help us think about the vision of the show and the world around it.

That distinction matters: AI wasn’t a writing tool so much as a lens, one that helped us see further, imagine more, and scale the vision beyond what was immediately on the page.

Where AI can enhance the Creative Process, not replace it

It all comes down to how you prompt your AI agent and what you’re actually asking it to do.

But that doesn’t mean it’s gospel. This brings us back to the Viagra analogy. The enhancement only works if the main function is there. If you don’t have the creative vision, the storytelling instinct, the emotional intelligence to understand what resonates with audiences—AI isn’t going to magically provide it. Viagra without the thing it enables doesn’t do much of anything.

And that’s exactly how we need to see AI in creative work: as an enhancement, not an answer. It’s never been the answer, and I genuinely don’t think it will become the only answer. It’s a tool we’ll have to start incorporating into our workflows but it won’t wipe out the process entirely.

Below I’ve outlined a three use cases where AI applications for the script and series development process was incredibly helpful:

Use Case #1: Tracking Pilot Plot Structure

At one point, I uploaded a Netflix pitch structure slide (like the one above) and my pilot script draft. I wasn’t asking AI to rewrite anything. I was trying to answer a much more basic question: what was I missing structurally that I couldn’t see anymore because I was too close to it?

AI helped me map my story against known story frameworks such as things like act breaks, escalation, and character motivation. It helped me visualize my pilot as a storyboard instead of just a linear script. I could ask it to write descriptions of every scene and list which characters were in it or even give me an analysis on how each scene helps the plot move forward.

That alone saved me a lot of time. Not because it gave me answers, but because it gave me a center point when I was wandering in an objective manner.

Use Case #2: Character Development

When I was writing my series bible, I didn’t want to just have one-paragraph descriptions of my characters. I wanted to create not just a backstory, but I wanted them to feel real. I had the idea, if some content creator is going to watch my show and make a character analysis video of one of my characters someday, I had to make the characters complex and deep enough to invite that level of scrutiny. So I found in-depth character analyses of TV’s icons’ The Sopranos, Tony Soprano and Mad Men’s, Don Draper.

Then using the transcripts of the two videos above, I asked ChatGPT to make me a character guide to follow themes and topics to cover for each of my characters. Here’s the document it produced for me. It’s pretty comprehensive.

Once I had filled out the template for my own characters, I uploaded the files and began using AI agents to explore adjacent ideas such as, what songs these characters might have on a Spotify playlist, or what would show up in a “starter pack” that captured their vibe. That exercise wasn’t about accuracy; it was about expansion. It became the seed that helped me think bigger and more holistically about the world of each character.

The Intersection within AI tools between Creatives and the Streaming Business

One useful exercise I’ve learned to pass on to creators is this: think backward from the moment a potential viewer is deciding whether to watch your content for the first time.

When a user is scrolling through a streaming platform looking for something to watch, the real question isn’t “Is this good?” — it’s “What information do I need to convince you to spend your time on something new?” And there are infinite combinations of “data points” about what will convince a user to watch a title, let alone one they’ve never seen before.

While it’s ultimately the algorithm’s job to answer that question at scale (what data points will convince me to watch something new), creators can meaningfully influence the outcome from day one.

I started approaching my own work by thinking of every title as having a kind of DNA: the tags, themes, tone, genre signals, and metadata that travel with a title as it moves through the media supply chain. And now that major streamers are using AI to generate those tags, who’s to say creators can’t get ahead by ensuring the proper tags are associated with their content?

As a practical exercise, I began running my script drafts through large language models and asking them to generate tags and thematic descriptors for each scene. I did this to help me understand the complexity of our writing and think bigger about:

• What themes were coming through the most from my writing?

• Which audiences would resonate with these themes?

For example, let’s take a look at the MadMen pilot script. I uploaded it to ChatGPT with the following prompt:

Evaluate this script and generate metadata tags of all the themes it could be associated with for an algorithm to retrieve and rank.

This is all the tags it generated for me:

TL;DR- Why we Shouldn’t Fear the Machine

I could spend several more posts explaining the under-estimated value of the metadata associated with a title on any streaming platform, and at some point, I will. However, as more folks start to become more bullish on AI tooling, our attitude as creatives and corporate professionals is just as key to leveraging the best of what it has to offer.

AI isn’t the death of creativity. It’s an amplifier of it, and only when real human vision is already present.

The industry panic comes from mistaking mimicry for meaning: AI can generate fluent outputs and simulate emotion, but it can’t originate insight because it doesn’t create from lived experience, memory, or intention.

For creators, the most powerful way to use AI is not as a replacement writer, but as a tool that reduces friction and expands possibility; helping map structure, deepen character work, surface themes, and translate a story into the metadata “DNA” that determines how content gets discovered.

The takeaway is simple: the future belongs to creators who treat AI like enhancement for the organ we uniquely possess, not its entire replacement.

The contradiction isn’t technological. It’s organizational, cultural, and catastrophically expensive.

Finance learned to separate creative bets from execution precision. Legacy media only learned the first half.

It’s not a data problem. It’s not a creativity problem. Streaming platforms cannot measure who creates subscriber value which makes optimization structurally and literally… impossible.

Think about that.

The business side of media operates like a hedge fund that never tracks which pod generates returns.

They can’t attribute value because their systems were Frankensteined together through decades of mergers, leaving insight permanently disconnected from action. Curation teams don’t know if their rails drive conversions. Algorithm teams don’t know if their models directly retain subscribers. And product teams don’t know if their features create value or just noise.

This isn’t a bug. It’s a systemic, organizational and politically ego-driven problem.

This article explains why streaming will remain stuck no matter how many mergers occur in this expensive limbo until it adopts what finance figured out in the 1980s: an operating model where value creation is attributable, owned, and continuously recalibrated. Not as an aspiration. As infrastructure.

The P&L Attribution Reality

Let’s analyze the core user action every streamer anticipates. There’s a new season of your favorite show coming out so you subscribe to the platform with the intention just to watch that one show and cancel when the season is over.

While there are various strategies to convert you to stay, from save offers to getting archaic algorithms to predict what you’re most likely to want to watch next, it’s almost impossible to attribute to what was the cause on why you stayed.

The content team that green-lit the show? The merchandiser who placed it on the homepage? The algorithm that recommended it after the user almost churned? The product team that reduced buffering during playback? All contributed. None can prove their incremental impact because the data measuring each contribution lives in incompatible systems with different definitions of success.

After every merger, the pattern repeats: synergies announced, headcount frozen, integration dragged out. Legacy teams survive on incompatible tech stacks. Functions multiply without clear accountability.

The problem is that legacy streaming platforms can’t tell who moved the P&L.

Finance solved this problem decades ago. When a quant improves a pricing model, the P&L is measurable. When a trader reduces slippage, the basis points are tracked. When a strategist finds new alpha, returns are attributed.

Citadel Securities seems like an absurd example for a media conglomerate to analyze, until you look at the operating principles. Bloomberg profiled their annual “goal book”: a 50-pound, 5,000-goal blueprint where every initiative has an owner, metric, timeline, and visible strategic integration. Nothing moves without accountability or a decision tree. That’s the standard legacy media needs in the streaming division, even if the intensity doesn’t translate.

While the creative side operates by different principles (akin to early stage VCs vs a Private Equity Firm), Citadel’s underlying framework has a foundation for legacy media to follow if it not only wants to survive on the business side, let alone thrive for decades to come. However, more than anything is a change in mindset around how we view content in an algorithm driven world: manage titles like positions, own them through their full lifecycle, continuously recalibrate.

The Holding Company Trap: Why the Current Merger Structure Guarantees a Slow Death

When two major streamers merge, the integration process becomes the first critical juncture. And yet, despite decades of legacy media mergers, no one has attempted a fundamentally different approach. Post-merger, streaming divisions operate like private equity rollups: the balance sheet gets monitored but rarely optimized. PE firms can justify this because they’re playing a longer game—hold for five years, improve EBITDA marginally, exit at a multiple.

But streaming is a real-time market. You’re competing for attention every day. Subscriber decisions compound weekly. Legacy streaming platforms operate like holding companies. They keep departments staffed because they existed before the merger, tolerating duplicative data sources because integration is hard, avoiding questions about revenue attribution because restructuring is painful, and will lose to a platform built for speed.

Streaming platforms have raw proprietary data that could yield a competitive advantage in the market. Everything from viewing behavior, engagement signals, preference indicators and churn warnings, just to name a few. But they can’t operationalize it because the pipes connecting insight to action are fractured beyond recognition. So they don’t. They add more dashboards. More cross-functional syncs. More performance reviews. More analytics layers on top of broken pipes.

This isn’t a missing feature. It’s a structural impossibility. Not because people lack competence, but the transfer of data pipelines makes this tracking nearly impossible.

Let’s understand why:

Why Legacy Media Mergers keep re-building the “Franken-stack”

Like we tell friends getting back with toxic exes: nothing changes unless you learn your lesson and do some introspection. Legacy media hasn’t learned. The employees re-stitching antiquated data sources are the collateral in this case, but at least we get a paycheck and can individually grow our knowledge base and gain first hand exposure. I’m one of them.

Here’s what we’re Stitching

A decade of consolidation glued together systems never designed to talk to each other. The infrastructure exists. The intelligence vendors exist. The signals exist inside platforms. But the pipes connecting those signals to the teams responsible for action remain fractured. Most legacy media companies can’t fix this without ripping out and rebuilding their entire data infrastructure, a three-to-five-year, multi-billion-dollar commitment that would require admitting the current system is unfixable.

Warner/Discovery. Paramount/Showtime. Disney/Hulu. Every merger created new layers of technical debt. Teams inherited overlapping content management systems with different metadata schemas, different performance definitions, different ways of tracking the same user behavior.

Take HBOMax. It inherited Discovery’s tech-stack, HBO’s prestige brand marketing, Discovery’s reality TV playbook, and CNN’s news infrastructure. Each division defines “success” differently, let alone content types. If we focus on the streaming business during the merger of Warner Bros. and Discovery, the ongoing debate was: what’s better- higher growth rates with higher churn or lower growth rates and lower churn?

When it comes to optimizing algorithms akin to algo-trading, Netflix isn’t streaming’s Jane Street because it’s smarter. It’s Jane Street because it never had to retrofit a trading system onto a floor-trading culture. It was born algorithmic. Everyone else is still translating.

Then the age-old question is why did Netflix avoid this entirely…for now (we shall see if the merger goes through and they inherit HBOMax). It’s one of the few long-form content streamers built from scratch as a unified data system. It never absorbed a legacy studio, never integrated incompatible tech stacks, never reconciled theatrical windowing with streaming economics.

The real kryptonite that’s parasitic to a legacy streaming division

Based on my observations immersing myself in this industry and my professional experience, the core dysfunction boils down to two structural failures:

1. Cable Metrics in a Streaming Infrastructure

Legacy streaming divisions never truly migrated from cable-era thinking to streaming-native operations. In the cable era, the attribution model was clean: retention meant advertisers or ongoing subscriptions (for premium channels), while completion rates determined renewals. More renewals = more retention = more revenue. The P&L attribution was clear.

In streaming, we can measure retention and completion, but we can’t cleanly attribute who or what drives them. The measurement infrastructure exists, but the attribution layer doesn’t. Legacy media still struggles to define content success in a multi-platform world. We default to metrics like “number of minutes watched” or “hours watched per sub” because corporate politics create inverse incentives: marketing tracks acquisition and fandom experiences across external channels while core product only measures in-platform engagement. This misalignment breaks down when the customer journey spans multiple platforms. We should be measuring the ROI of IP engagement across all platforms—an approach I explore in Content as a Product: The Future of Consuming Television Part II.

2. One-Size-Fits-All Benchmarks Across Incompatible with Premium Channels

Different content serves different strategic purposes. Prestige dramas build brand credibility. Reality TV drives daily engagement. Kids’ content increases household stickiness. Catalog fills libraries and reduces churn during content gaps.

But post-merger, legacy streamers apply uniform benchmarks and objectives across all content types, as if every show should perform like a blockbuster tentpole. Like Netflix, other streamers tried to become the Cheesecake Factory of long-form content: sprawling menu, no clear identity, operational chaos underneath.

While platforms like Roku or Tubi have become like a supermarket, the lesson legacy media hasn’t quite grappled with is that premium networks like HBO or Showtime attract different customer bases and those markets want the delineation and prefer the add on. I even cited this in an article I wrote a year ago.

In 2018, this paradox became evident when HBO was acquired by AT&T. Warner Media’s leadership, determined to outperform Netflix, learned a costly lesson: prestige content isn’t always compatible with the metrics-driven pursuit of mass audience engagement and increased watch time.

The Consequences: Why it’s almost impossible for Insight to Become Action

This is why HBO Max can deliver Succession and The Last of Us. It’s marquee content that performs phenomenally by any creative or engagement metric and still struggle with retention relative to investment. The content performs. The platform can’t capitalize on that performance because the underlying system isn’t architected to translate user intent into persistent engagement.

You can build a unified recommendation engine on top of that infrastructure. You can polish the app interface. But underneath, you’re managing incompatible measurement systems stitched together from multiple acquisitions. And incompatible measurement means attribution is structurally impossible. You can’t optimize what you can’t measure. You can’t measure what you can’t attribute. And you can’t attribute when your benchmarks were designed for a different business model entirely.

Regardless of how this merger shakeup pans out, every media merger announcement follows the same script: promises of scale, synergies, and a strengthened position against tech platforms. Yet as the years go on, the same operational paralysis returns. Different companies, identical problems. The question isn’t who consolidates next—it’s why consolidation never fixes what’s actually broken.

For a living, I build the products that separate signal from noise inside a major streaming platform, identifying which metadata, curation strategies, and metrics actually drive user behavior.

But to understand why media companies remain stuck, you have to look beyond engagement dashboards and examine what happens at the operational level after companies merge. There’s a recurring pattern, almost parasitic, that emerges after each consolidation. The same technological and organizational failures resurface because the industry treats symptoms instead of diagnosing the underlying disease.

Two Industries, One Evolution—Until the End

The stock market of the 1970s and the golden era of television operated under identical constraints: slow measurement, opaque pricing, and a handful of gatekeepers controlling outcomes. But finance treated these constraints as a crisis to solve. Floor trading became electronic markets. Electronic markets became algorithmic execution. Inefficiency transformed from an accepted reality into something to eliminate.

Streaming should have been the media industry’s identical evolution, its electronic trading moment. Instead, it rebuilt Nielsen in high resolution. Hours viewed. Minutes streamed. Completion rates. New wrappers around the same backward-looking metrics that cable used.

Yet the structural parallels remain undeniable. Both industries allocate capital: one in dollars, the other in attention. Both manage power-law portfolios where a few hits carry everything else. Both bet forward on uncertain outcomes. But only finance built the machinery required to operate in that environment: granular attribution, unified data architecture, and continuous recalibration.

Media never finished that final step. And the consequences cascade everywhere. Below I mapped out a table to show the parallels between the media and financial landscape throughout the decades.

The evolution tracks perfectly—until the final transition. Finance crossed the bridge. Media got stuck on it.

The Infrastructure Paradox

Streaming platforms already operate exactly like trading engines. They use real-time data to decide what content to surface, to whom, at what time, and for how long. Their goal is continuous retention optimization—functionally identical to a hedge fund’s continuous optimization of portfolio exposure.

Netflix’s recommendation engine behaves more like Citadel’s order routing algorithm than like a traditional network scheduler. YouTube’s feed is essentially a momentum-trading strategy applied to human attention. TikTok’s “For You” page is pure quant: short signals, rapid prediction cycles, and feedback loops optimized for velocity.

The platforms have the infrastructure of a trading floor. They just don’t operate like one.

That’s the structural problem the industry refuses to name: development operates like venture capital, but streaming and networks don’t operate like a hedge fund—and that gap is costing billions.

In development, executives place early bets on talent and IP, nurture risky projects, and accept that most will fail. This is exactly how venture capital works: portfolio approach, high variance, taste-driven decision-making. It’s the right model for the creative layer.

But once a title is produced and enters the streaming ecosystem, the operating model should flip entirely. A finished series isn’t a startup anymore—it’s a liquid asset that needs active management, dynamic allocation, and continuous optimization around measurable signals.

That transition never happens because of legacy media’s legal incentives from the cable days that haven’t been updated to reflect the consumer needs and financial incentives in a streaming world.

The viewership data remains locked inside a closed system, and when that same title moves to a different streaming service six months later, there is no continuity of measurement. Without consistent access to engagement metrics, it becomes nearly impossible to understand audience trends or assess long-term performance for the owners (the Studio) and creators of the IP.

When a studio that owns the IP for a film or TV series licenses it to a streaming platform, the streamer has little incentive to share performance data. Unlike the old broadcast era, when producers could simply check weekly Nielsen ratings, studios and creators today often have no way to track how their titles are performing across multiple platforms, unlike independent content creators that are not tied to a major studio or network.

I work with streaming data regularly, and the current model of withholding access does not serve the industry. While streaming platforms may benefit financially in the short term, restricting information prevents creators and studios from developing more relevant content and contributes to inequitable residual compensation.

When capital fails to circulate and strengthen the broader ecosystem, it becomes an early sign of systemic parasitism within the industry.

The Content Market That Already Exists

Source: The Modern Meme Lifecycle

Returning to the financial analogy, streaming already behaves like a real-time market. Every platform repeatedly cycles through the same three conditions:

Content behaves like an asset class. Attention moves like capital. Titles exhibit volatility, momentum, and decay curves.

When the underlying data is fragmented, it is the equivalent of a company losing visibility into its stock performance every time its shares are traded on a different exchange. That is essentially what happens when content moves from one streaming service to another. And because of that fragmentation, platforms struggle to operate with the precision the market demands. They lack the operating system that finance is already built to connect signals with action. Slow access to performance data, incompatible technology stacks, and organizational structures inherited from pre-streaming eras make real-time response nearly impossible; unlike traders responding rapidly to market movements.

The result is the media industry in the world of streaming is running a high-velocity attention market on infrastructure designed for linear television, leaving most companies acting more like post-merger holding entities than active operators.

And traditional executive thinking and standard cable practices increasingly lag behind the technology and consumer expectations of an algorithm-driven, continuously curated world.

What Real Execution Looks Like

In well-run trading operations, every role has clear revenue attribution. The quant who builds a better pricing model gets measured on the incremental P&L their model generates. The trader who executes with lower slippage gets measured on basis points saved. The strategist who identifies new alpha gets measured on returns attributable to that signal.

Legacy media doesn’t work this way—and structurally cannot, given its frankenstein of a data infrastructure. Let’s break this down:

How Weak Corporate Organs Become Media Merger Kryptonite

The deeper truth is that streaming doesn’t have a content problem—it has a data-plumbing problem and a P&L problem. The equivalent of a Bloomberg Terminal for content already exists. Intelligence vendors exist. The signals are already inside the platforms. What is missing is the infrastructure that connects those signals to the teams responsible for acting on them. The data pipes are fragmented, inconsistent, and often inaccessible, making it nearly impossible to tie performance signals to operational decisions.

Why?

• A decade of mergers glued together incompatible content management systems, metadata schemas, and user databases

• Teams inherited overlapping definitions of success with no unified language for performance

• Platforms own the metrics, studios want the metrics, and no one has incentive to share them transparently

• Attribution is impossible when infrastructure itself obscures the signal

The insights exist. The market moves. The data flows.

But organizations cannot see it cleanly, measure it cleanly, or act on it cleanly.

Finance learned to separate the creative layer (venture bets) from the execution layer (real-time trading). Legacy media learned only the first half. And that gap between having trading-floor infrastructure and operating with cable-era processes is the kryptonite keeping Hollywood hostage.

This brings us to Part II: The Data Pipeline Reality and the P&L Attribution Problem.

Part II examines the core structural failure in detail:

• Why platforms can’t see which teams actually drive subscriber behavior

• Why no one owns the lifecycle of a title from greenlight through decay

• Why streaming companies are effectively flying blind in a market that demands precision

• What the data architecture would need to look like to actually function like a trading operation

This is the failure beneath every other failure. It’s why the execution layer breaks. It’s why human insight can’t translate into machine action. And it’s why no amount of consolidation will fix what’s broken until the industry rebuilds its operational foundation from the pipes up.

The Multi-Million Dollar Opportunity Media Companies are Missing

When Sex and the City (SATC) premiered in 1998, its modest ~3 million viewership grew to 10.6 million by the finale—a surge of more than 250% that cemented the show as a cultural phenomenon with decades-long value, transforming four fictional characters into generational icons still generating millions through syndication nearly three decades later.

Today's television landscape presents an unprecedented paradox. Netflix invested $17 billion in content in 2024, yet it still grappled with subscriber declines in mature markets. Meanwhile, creators like Alex Cooper transformed herself from collecting unemployment checks to building a $160 million media empire in just six years with "Call Her Daddy," commanding 10 million listeners per episode through deeply engaged audience monetization. The difference between these outcomes reveals a fundamental truth: we now know exactly how passionate and devoted audiences are the foundation that creates lasting value -but most studios ignore this truth in their production and financing models

This blueprint explains why the creator economy points to the future of sustainable entertainment, while traditional television financing actively prevents shows from developing these revenue engines around their own intellectual property. By examining the structural barriers preventing innovation—from the legacy deficit TV financing models to streaming's broken residual systems, it will be evident how proven TV success stories, developed without ecosystem thinking, leaves millions in untapped digital value on the table. More importantly, -a practical solution will be outlined: the Business Casual (BC) Method, a low-cost, high ROI production strategy that channels just 0.32% of an episode’s budgets into permanent, revenue-generating digital assets that bypass traditional profit waterfalls entirely.

Using SATC as our case study (as if it came out in 2025 with the same audience growth trajectory), the BC approach could generate $15-45 million in supplemental revenue over 20 years from a $1.27 million investment—transforming traditional marketing expenses into decades-long cultural and financial assets. When studios green-light their next prestige show, they're not just funding television—they're potentially creating a digital empire that can generate revenue for decades while solving streaming's fundamental compensation crisis through transparent, performance-based residual payments that benefit everyone from showrunners to technical crew.

The State of Streaming: Growth Without Profit

The streaming industry is steadily growing but churn is always a battle: global subscriber growth increased by 11% last year, and monthly churn averages 5–7%. Yet while platforms fight to retain users, the money tells a different story: in 2024, YouTube generated $42.5 billion in revenue—outpacing Netflix’s $39.2 billion—by monetizing over 2 billion users through advertising and premium subscriptions. Projections for 2025 show Netflix pulling ahead with $46.2 billion, largely from subscriptions, but YouTube will remain nearly as large at $45.6 billion with the majority coming from ads.

The overlap is striking: in the U.S., 57% of YouTube users are also Netflix subscribers, which means these audiences don’t just watch—they toggle between ecosystems.

Alex Cooper’s leap from a $70,000 Barstool Sports contract to a $100+ million SiriusXM deal illustrates why this matters: she captured the full engagement journey across platforms, growing her audience by 160% and monetizing it directly. The lesson is clear: the real competitive advantage isn’t just in streaming hours—it’s in owning the engagement ecosystem where emotional connection gets converted into revenue.

Why the Content Creator Economy is Sustainable: Total Ecosystem Control

Cooper’s exit from Barstool wasn’t just about salary—it was about IP and platform ownership. By controlling distribution, she turned content into long-term equity instead of one-off paychecks. The margin contrast is stark: traditional marketing spends heavily with little residual value, while diversified digital content—cheap to produce but with infinite shelf life—delivers 2,000%+ ROI potential.

Prestige TV Already Proved the Audience Exists

This isn't disruption—it's restoration. Remember The Sopranos? It wasn't designed for binge-watching—it was crafted as a weekly ritual spanning years, creating shared social experience akin to following a sports team. Prestige TV grows its audience through primary storytelling, cultural conversation, and ongoing engagement. As SATC already proved with its 250% audience growth, Cooper’s Model drives loyalty with authentic personal content, behind-the scenes moments, and direct community relationships in the same 5-6 year time frame.

The crucial insight: Both prestige television and Cooper's model create sustained engagement through different but equally powerful emotional value propositions:

Prestige TV Journey:

• Primary Content: Complex, serialized storytelling requiring investment

• Supplemental Engagement: Theory discussions, character analysis, cultural commentary

• Community Building: Shared interpretation of artistic meaning

• Long-term Value: Cultural significance and rewatchability

Authentic Content Journey (Cooper Model):

• Primary Content: Personal, direct conversation requiring trust

• Supplemental Engagement: Behind-the-scenes authenticity, extended conversations

• Community Building: Shared life experiences and advice

• Long-term Value: Relationship depth and life applicability

Both models work because they understand that content consumption is ultimately about human connection—whether connecting to compelling fictional characters or authentic personalities. Yet, most studios fail to monetize these connections at every engagement point.

The Missing Gap: Why Streamers Are Leaving Money on the Table and Studios Lose Billions

Studios are already positioned in the top percentiles of the creator economy—they just aren’t capitalizing on it. Today the creator economy is worth $191.55 billion and growing 22.5% annually toward $528 billion by 2030.

Netflix’s 31.5M-subscriber channel and Warner Bros.’ 12.6M-subscriber channel both rank in the top ~0.5% of YouTube globally—among just 2,639 accounts with 10M+ subscribers—and at this scale could generate annual revenues in the tens of millions (Netflix $30–50M, Warner Bros. $10–20M). Even if monetized at a fraction of Alex Cooper’s efficiency, they could unlock an additional $25–50M per year.

The critical difference: Cooper captured every moment of her growth journey through supplemental content, while traditional shows left millions in engagement value on the table. When Sex and the City aired in 1998, there was no infrastructure to monetize the intense fan discussions, theory crafting, and cultural commentary it generated.

Today, that infrastructure exists—but streamers still treat their YouTube channels as marketing afterthoughts rather than revenue engines.

This is where the ROI of emotional connection comes into focus. Fandom is no longer a byproduct—it is the customer journey. SpongeBob SquarePants has generated over $13 billion since 1999. Disney understood this long before streaming existed by transcending their content into theme parks, video games, and Broadway: in Q2 2025 its Experiences division grew 6% year-over-year to $8.9B in revenue, with operating income up 9% to $2.5B and company margins climbing to 19.3%.

Even Netflix is now trying to model this Disney playbook. Co-CEO Ted Sarandos has framed “Netflix Houses” in Philadelphia, Texas, and Las Vegas as recurring destinations—spaces fans might visit multiple times a year, with rotating experiences he envisions eventually spanning dozens of global locations.

Consider the viewer behavior that studios completely miss: every Sunday, millions binge The White Lotus on HBO Max and feel deeply connected, but it is fans who create memes, analysis, and commentary on platforms outside studio control—enriching someone else, while the studios gain nothing.

Sunday – Binge on HBO Max, deeply connected

Monday – Scrolls X for memes to extend the experience

Tuesday – Watches unauthorized breakdown videos on YouTube

Wednesday – Joins conspiracy threads on Reddit

Thursday – Shops for Thailand vacation rentals, inspired but untracked

This entire emotional journey generates millions in engagement and revenue, yet outdated financing and profit structures leave no budget, incentive, or clear legal path for studios or talent to capture it. To understand why studios are walking away from billions in supplemental content revenue, we need to examine how the current financing structure systematically prevents ecosystem thinking.

Subscribe now

“If It’s This Obvious, Why Isn’t Hollywood Doing It?…Because Traditional TV Financing Actively Blocks Innovation.”

Let's understand why:

1. How TV Financing Budgets Work Today

According to John Wells’ comprehensive breakdown of TV financing and production budgets, the economics made it clear why studios never prioritized supplemental digital content in the early days.

Here's how a typical $5 million episode gets funded (according to Wells’ article linked above):

1. Network pays license fee: $3.75 million (75% of budget)

2. Studio covers deficit: $1.25 million (25% out of pocket)

3. Studio waits years for profit through syndication, international sales, DVDs

This 60-year-old model existed because networks needed content but couldn't afford full production costs. Studios covered “deficits” in hopes of -eventual syndication profit from "ancillary markets"—selling reruns to other networks, international broadcasters, and DVD distributors.

Wells' budget breakdown reveals the fundamental constraint: most costs are locked in—office rental ($5,000), department heads ($143,411), crew salaries ($115,200), equipment ($125,946). Studios have virtually no discretionary spending flexibility within the approved budget.

Here's what the current structure doesn't address: The existing budget creates a structural disincentive for supplemental digital content, trapping it in an economic dead zone. Networks allocate $500K-$1M per season for promotion that generates temporary awareness spikes but builds zero lasting assets.

Meanwhile, studios own the intellectual property but lack economic motivation to create promotional content they can't directly monetize. This ownership-incentive misalignment persists because traditional revenue streams—advertising, licensing, and syndication—provided sufficient returns without requiring measurable marketing ROI.

Ted Lasso is a textbook case of the supplemental content opportunity studios overlook. Warner Bros. owns the IP, while it has been rumoured Apple TV+ paid ~$10 million per season plus marketing—yet neither captured the massive secondary engagement. Independent creators now rack up millions of views with reaction videos, analysis, and commentary, building six-figure businesses off the show’s cultural impact. The result: a double loss—Warner Bros. gave up syndication rights in a cost-plus deal, and Apple failed to monetize the phenomenon they bankrolled.

Today this legacy budget framework flat-out fails because it was built for a world that no longer exists. It locks tens of millions into promotional campaigns that vanish after premiere week, leaving studios with sunk costs, networks with no enduring assets, and talent completely excluded from upside. Worse, it actively prevents innovation—treating supplemental content as disposable scraps instead of the foundation for lasting residual value.

Bottom Line: This structural flaw doesn’t just waste money—it actively blocks supplemental content, turning millions in promo spend into lost value instead of residual income, and setting the stage for the deeper crisis now crippling the industry: the collapse of residuals in the streaming era.

2. Why Residuals barely exist in a streaming world

The Profit Distribution

Understanding why this opportunity exists requires understanding how television economics fundamentally shifted. John Wells Productions' detailed financial breakdown of a hypothetical successful show reveals the complex mathematics of modern TV production: a 5-season series with a $5 million per episode budget generates $500 million in gross receipts, but after distribution fees (15%), overhead fees (15%), and interest charges (4.25%), the actual profit participation becomes dramatically smaller.

With a total production cost of $303 million for 60 episodes, the traditional model leaves studios dependent on licensing deals and backend participation.

The Streaming Revolution That Broke Everything

When Netflix introduced the "cost-plus" model, it seemed like studios finally won. Netflix pays 100% of production costs plus a premium (typically $5.2 million for our $5 million episode). No more deficits!

But there's a catch: Netflix seized worldwide rights for 10+ years. No syndication. No international sales. No ancillary markets. The traditional profit model just... disappeared.

The Residuals Problem Gets Worse: As the Deadline article reveals, streaming residuals are broken. Traditional residuals are based on licensing fees paid to studios, but supplemental digital content creates new revenue streams (YouTube, social media) not covered by existing residual structures. Talent contracts don't typically cover additional promotional content appearances, creating legal complications that further discourage innovation - so neither talent nor studios have a financial reason to invest in or optimize digital content.The net effect is that studios treat digital content as a cost rather than an investment, despite its proven, compoundable ROI.

The Profit Waterfall That Kills Innovation

Even when shows succeed, Wells' example exemplifies how complex the current system has become. A hypothetical series generating $500 million in gross receipts faces this deduction sequence:

• Distribution fees: $41.25 million (15% of non-domestic revenue)

• Overhead fees: $45.45 million (15% of production costs)

• Interest charges: $12.87 million (4.25% on production investment)

• Profit participant payments: Multiple backend deals reducing studio take

Total deductions: $99.57 million before studios see meaningful profit.

The Legal Structure That Prevents Progress

IP ownership confusion between studios and networks, technology exclusivity deals that "freeze" distribution rights, and residual formulas that don't account for social media revenue create a perfect storm of disincentives. Everyone knows supplemental digital content works, but nobody knows how to make money from it under current deal structures.

Why this Structure is Lethal for the Digital Age

This waterfall structure means supplemental content falls through the cracks. Studios don't want to spend money on content they can't clearly monetize. Networks don't want to pay for assets they won't own. The result? A $50 billion industry that treats supplemental content like craft services—a necessary expense rather than a revenue opportunity.

The Bottom Line: Promotional content gets trapped in post-production budgets as an afterthought because it's not planned during pre-production. Networks treat it as a marketing expense, studios treat it as someone else's problem because there's no clear ROI incentive.

So what's the solution?

Introducing the Business Casual (BC) Method: Transforming Existing TV Budgets into Lasting Digital Empires

The Problem: Wells' budget structure locks most costs into fixed categories, while networks spend millions documenting shows but can't monetize that material long-term. Meanwhile, studios own the IP but lack direct distribution control. This disconnect is what Wells calls the "royalties issue"—a structural misalignment that wastes $500K-$1M per season on promotional content that generates temporary awareness instead of lasting value.

The Solution: The Business Casual Method is a B-crew production strategy that costs just 0.32% of episode budget while establishing studio-owned IP for all supplemental content. By leveraging natural downtime in existing production schedules, studios can create permanent revenue-generating assets that bypass traditional profit waterfalls entirely, transforming wasted promotional spend into decades-long cultural and financial assets that deepen fan engagement while building IP longevity.

1. Budget Integration: Working Within Wells' Production Framework

Wells' budget breakdown reveals the fundamental constraint: most costs are locked into fixed categories—office rental ($5,000), department heads ($143,411), crew salaries ($115,200), equipment ($125,946). Using Wells' $4.29M per episode baseline for a 13-episode season, studios have virtually no discretionary spending flexibility within the approved $55.77M total season budget.

Budget Efficiency:

The Business Casual Method circumvents these constraints entirely by fitting within what Wells identifies as "discretionary areas that you control and can adjust" rather than fixed contractual costs.

For an Episode:

The per-episode breakdown:

• B-crew content producers (prorated): $11,538

• Dedicated post-production: $2,000

• Equipment costs: $0 (modern iPhones + existing gear)

• Additional crew: $0 (optimized around existing downtime)

• Total per episode: $13,538

For the Season:

• Wells Episode Budget: $4,290,455

• Business Casual Addition: $13,538 per episode

• Percentage Impact: 0.32% increase per episode

• Season Total Addition: $350,000 (13 episodes × $13,538)

• Overall Season Impact: 0.63% increase to $55.77M budget

Where the Money Goes:

• Production Phase: $11,538 per episode (integrated into existing crew costs)

• Post-Production Phase: $2,000 per episode (added to Wells' $244,272 post budget—just 0.82% increase)

• Equipment: $0 (leverages Wells' existing $125,946 camera/equipment allocation)

• Extended Production Time: 0% (captures during existing downtime)

In other words: Incremental Cost: For a $4.29M-per-episode TV budget, adding the BC Method raises costs by only $13,538 per episode. That’s a mere 0.32% bump—or $350,000 for an entire 13-episode season.

Spending Breakdown: The additional money hires a dedicated B-crew and post-production resource during normal downtime—no costly overtime or extra equipment needed.

This fits seamlessly within Wells' framework because it utilizes what he identifies as controllable budget areas without extending the production timeline or requiring additional equipment rental.

2. Ownership Control: Studio-Owned Social Infrastructure and Seamless Integration

The Business Casual Method operates through three streamlined in-house studio teams that work closely with network producers while maintaining studio control over digital implementation and social scheduling.

Team 1: Content Development Partnership (Studio–Network Content Dev) Embeds with productions to identify extensible moments and manage B-crew operations while ensuring strategic alignment with network promotional priorities and no disruption.

Team 2: Digital Experience Control (Studio Digital Control) Transforms raw content into platform-optimized formats and manages publishing calendars across owned channels while maintaining complete control over social media implementation.

Team 3: Network Acquisition Driver (Acquisition Alignment) Works with external network partners to ensure studio-owned TV IP social channels drive viewer acquisition for network broadcasts, creating mutual benefit.

This collaborative structure allows studios to maintain ownership and control of digital assets while ensuring network partners benefit from increased audience acquisition and engagement driven by studio-managed social channels.

3. Tactical Execution: Writers' Room to B-Crew Pipeline

The Business Casual Method begins in the writers' room during season planning, ensuring supplemental digital content is strategically woven into the production fabric rather than opportunistically captured.

Pre-Production Planning:

• Writers' room identifies character moments, relationship dynamics, and world-building opportunities that extend beyond episodes

• Content development team creates supplemental content calendar aligned with episode production schedule

• B-crew scheduling integrates with main production timeline during Wells' "discretionary" planning phase

Systematic Capture Strategy: The key is establishing a dedicated B-crew that shadows main production without interfering. As noted in the framework, "not every single actor is going to be in every single scene and not every single production person is going to be needed for every single moment."

• Pre-shoot prep: While A-crew sets up complex scenes, B-crew captures cast interviews about upcoming storylines

• Between setups: During lighting changes and camera moves, B-crew films character segments and behind-the-scenes moments

• Lunch breaks: Dedicated time slots for deeper cast discussions about character motivations and episode themes

• Wrap periods: Post-scene reactions, bloopers, and spontaneous moments when actors are still in character but relaxed

Implementation Details:

• Characters not filming seasonal episodes get scheduled for supplemental content (10-20 minute segments)

• Content gets programmed during development phase with writers in the writer's room

• Guest appearances in short-form content leverage existing talent already on location

• No overtime costs because capture happens during built-in production breaks

• Studios work with networks for post-production, integrating with existing infrastructure

• Output: ~3-5 videos per episode (~50 total per 13-episode season) spread across platforms

The Bottom Line: For less than the cost of a single traditional behind-the-scenes clip ($28,000+), studios can fund an entire episode's worth of supplemental content ($13,538) that generates revenue for decades while building cultural equity through deeper fan engagement and expanded world-building that strengthens IP longevity beyond the original series run.

Subscribe now

A Thought Experiment: What if SATC Could Build a $45M Digital Asset in 2025?

Picture this: It's 2025, and Darren Star pitches SATC to a modern studio and network with access to today's creator economy. The show follows the exact same trajectory—same writing, same cultural impact, same audience growth from 3 million to 10.6 million viewers. Much like Alex Cooper's meteoric rise from unemployment to $160 million in just 6 years by building a podcast empire that evolved from niche content to celebrity interviews, this modern SATC would transform from traditional television into a multi-platform media juggernaut.

But this time, there's a twist: every episode generates supplemental digital content that builds a YouTube empire alongside the television success.

This isn't fantasy. We know the show worked—it proved its audience appeal over six seasons and 94 episodes. The question isn't whether SATC would succeed, but what that proven success would be worth when combined with today's digital infrastructure. The answer, as we'll see, is staggering: a $45 million digital asset built for the cost of traditional marketing.

The Foundation: Why YouTube Changes Everything

To understand the opportunity, you need to grasp how dramatically the revenue landscape has shifted. In 1998, a hit show generated money through advertising, syndication, and merchandise. In 2025, those same audience relationships can be monetized through YouTube's sophisticated creator economy, where fashion and lifestyle content commands premium rates—$4-8 per thousand views compared to $1-3 for general content.

SATC's audience represents the holy grail for digital advertisers: affluent women aged 25-45 with disposable income and strong brand affinity. YouTube's algorithm would recognize this immediately, serving luxury fashion, beauty, and lifestyle advertisements at premium rates. The show's production quality would further elevate its content above typical user-generated videos, commanding the highest tier of advertiser spending.

But AdSense revenue, while substantial, represents just the foundation.

The Production Strategy: Maximum Impact, Minimal Investment

The beauty of this approach lies in its efficiency. Creating supplemental content doesn't require reinventing television production—it requires smart utilization of existing resources. During the natural downtime of any television shoot, a small B-crew captures behind-the-scenes moments, conducts cast interviews, and documents the fashion choices that define each episode.

The numbers are remarkable: for just $13,538 per episode—less than one-third of one percent of a typical Wells Productions budget—a show can generate 3-4 pieces of digital content. Across SATC's 94 episodes, this represents a total investment of $1.27 million and yields approximately 376 videos. No additional equipment needed, no disruption to primary production, no compromise to the show's quality- far outstripping the value of traditional ‘disposable’ marketing campaigns

See Appendix A for detailed per-episode cost breakdown and content production methodology.

The Audience Transformation: From Viewers to Community

The magic happens in how television audiences convert to digital communities. SATC's proven viewership provides the foundation, but the conversion mechanism is what creates lasting value. Harvard Business Review reports the value of engaged audiences: the 90-9-1 rule suggests that 10% of viewers engage beyond passive watching, and well-managed communities have the potential to achieve 50% lurkers, 23% contributors, and 27% creators—creating opportunities for media companies to transform passive viewers into engaged fans who generate higher average revenue.

This targeted approach focuses on quality over quantity. With SATC's ~3 million Season 1 viewers, capturing just 10% of engaged viewers translates to 300,000 highly committed fans who actively seek supplemental content, participate in discussions, and become part of a lasting digital community.

This growth compounds throughout the show’s run. While there’s no official Nielsen data season by season, we can conservatively model ~4.8 million television viewers by the end of Season 3, which at a 10% engagement rate equals ~480,000 potential community members. By the series finale, SATC’s actual audience of 10.6 million implies ~1.06 million deeply invested fans—exactly the kind of community that transforms completed content into living cultural artifacts.

These aren't just numbers—they represent a community that traditional television could never capture. Where broadcast ended at the credits, digital engagement continues indefinitely, creating ongoing relationships that can be monetized across multiple revenue streams. As the research shows, when viewers develop emotional connections with content, they inevitably seek additional touch-points outside the streaming platform itself, creating opportunities for media companies to direct these fans to official owned channels with minimal investment.

See Appendix B for complete audience conversion modeling and subscriber growth projections.

The Revenue Revolution: Four Streams of Lasting Value

YouTube AdSense: The Reliable Foundation

Every view generates revenue through YouTube’s ad network. With SATC’s premium audience commanding $4 RPM, the numbers scale quickly. Season 1’s 48 videos averaging 225,000 views each produce 10.8M views, generating $43,200 in the first year alone. Factor in the long tail—20% of peak views annually for Years 2–11, plus 10% of peak views annually for Years 12–20—and Season 1’s contribution reaches $168,480 over 20 years.

Scaling this across six seasons of increasing popularity, anchored by the show’s 10.6M finale audience, yields $4.12 million in total AdSense revenue. This represents just the foundation—baseline income generated simply by turning on monetization.

See Appendix C for detailed revenue calculations, assumptions, and 20-year projections for all four streams.

Sponsorships: Where Real Money Lives

The creator economy’s biggest lever is sponsored content. Fashion and beauty brands pay premium rates to reach SATC’s demographic, with integrations commanding $30,000–$55,000 depending on the season and subscriber count. By Season 6, with luxury brands competing for access to 1.06M engaged subscribers, sponsorship revenue scales to ~$2.5M annually.

Assumptions: Sponsorship density ramps from ~20% of videos in early seasons to ~50% by Season 6, with average rates scaling alongside subscriber growth. Over 20 years, this produces $9.25M in sponsorship revenue.

This isn’t theoretical—nearly 70% of creators cite brand partnerships as their primary income source. With SATC’s cultural cachet and premium audience, the show would slot directly into the high-value sponsorship tier. (See Appendix C)

Merchandise: Beyond Traditional Licensing

Digital audiences buy differently than traditional TV viewers. With direct relationships and authentic engagement, merchandise conversion rates rise dramatically. Even conservative assumptions—1% of subscribers purchasing $25 of product annually—generate significant value.

Over six seasons, gross merchandise sales reach $4.12M. After applying a 50% cost of goods sold (COGS), this leaves $2.06M in net revenue flowing into the MAGR waterfall. (See Appendix C).

Platform Partnerships: The Long-Term Play

Supplemental content also becomes valuable IP in its own right. Streaming platforms and FAST channels license behind-the-scenes content, fashion breakdowns, and interviews, creating new monetization layers that continue decades after production.

Conservative modeling places this at $5.23M over 20 years, a steady revenue stream that underscores the durability of supplemental assets. (See Appendix C).

The Financial Reality: Building Assets, Not Expenses

When you combine all revenue streams across 20 years, the numbers become extraordinary. The conservative case projects $15.65 million gross, which after YouTube's 45% platform fee and merchandise COGS leaves $8.61 million net revenue. Against a $1.27 million investment, that's an ~1,130% ROI—every dollar invested returns more than eleven dollars.

But the upside case reveals the true potential. With optimized execution—higher RPMs, premium sponsorship density, enhanced merchandise strategies, and strategic platform partnerships—total revenue reaches $38–45 million, producing ROI of 2,886–3,436%. This represents one of the most efficient uses of production capital imaginable.

This isn't marketing spend that disappears after broadcast. This is asset creation that compounds in value, generating passive income for decades while continuously promoting the original show.

See Appendix D for complete financial modeling, sensitivity analysis, and upside scenario calculations.

Digital Residuals: The Future of How Everyone Gets Paid- Fixing Streaming’s Broken Compensation

Perhaps most importantly, YouTube revenue can be structured to mirror television's traditional residual system, creating ongoing income streams that solve the fundamental problem plaguing the streaming era. The crisis is stark: where writers once relied on 40-week network seasons and substantial rerun residuals, streaming has replaced that system with shorter seasons, long hiatuses, and minimal one-time payments.

The MAGR Framework Adapted for Digital

The beauty of this approach lies in its familiarity. YouTube revenue can flow through the same Modified Adjusted Gross Receipts (MAGR) waterfall that already governs traditional television backend payments. Gross receipts flow into YouTube, which takes its 45% platform fee. The remaining 55% is treated as net revenue. From there, production deducts network shares, overhead, crew allocations, and then distributes backend participations.

The critical difference is transparency and immediacy. Where streaming platforms guard their data behind opaque algorithms, YouTube provides clear metrics that enable accurate residual calculations. Where traditional syndication takes years to generate meaningful returns, YouTube residuals begin with the first upload.

Real Financial Impact for SATC Talent

Traditional streaming residuals today often amount to a one-time payment of $50–$200 — barely enough to cover a dinner out. By contrast, the Business Casual (BC) Method applies a MAGR waterfall to supplemental digital revenue, ensuring everyone from writers to crew participates in long-term success.

Backend Pool (20-Year Net):

- $18.81M MAGR Net × 40% = $7.52M allocated for participation

Talent Compensation (Option A — crew included in backend):

- Showrunner / Executive Producer (15 points): $1.13M over 20 years

- Series Regulars (0.5% each): $37.6K per actor over 20 years

- Writers’ Pool (5%): $376K distributed among the writers’ room

- Technical Crew (10%): $752K shared across eligible crew

Residuals Comparison:

- Streaming residuals today: $50–$200 one-time

- SATC YouTube model: $38K–$100K+ per series regular over 20 years

- Improvement: ~200–500× higher

Key Takeaway: Instead of disappearing after broadcast, SATC’s supplemental ecosystem becomes a living digital asset. Backend revenue flows for decades, aligning incentives across creative and technical teams while solving the compensation crisis of the streaming era.

See Appendix E for a full breakdown of backend pool alternatives, payout structures, and sensitivity scenarios.

Solving Streaming's Residual Crisis

This framework addresses every major issue that led to the 2023 writers' and actors' strikes:

• Transparency replaces opacity. YouTube's clear data eliminates the guesswork that plagued streaming negotiations.

• Performance-based pay replaces flat fees. Revenue scales with success, just as guilds demanded.

• Long-term income replaces one-time payments. Compensation continues for decades.

• Established frameworks eliminate reinvention. MAGR is already understood by guilds and studios.

The Precedent: Television's Proven Success

Throughout the decades, TV residuals have served as a clear example of how backend participation (profit-sharing, royalties) functions effectively when scaled across an industry. A successful show with $500M in gross receipts can generate $10.5M for a showrunner with 15 points. SATC's digital model operates on a smaller base but uses identical mechanics—with the crucial advantage that income begins immediately and compounds over time.

The strategic insight is profound: SATC's YouTube channel would generate backend payments comparable to moderately successful syndicated television, but with complete transparency and immediate cash flow.

Strategic Advantage: Why This Matters Now

Studios have the core audiences, resources, and infrastructure to win the creator economy, but lose out to digital-native creators on supplemental revenue.

Early adopters gain compounding returns and can lock in top creative and technical talent by offering superior, clear, long-term incentives.

This isn’t theory; Alex Cooper’s journey and TV’s persistent audience data prove it works.

Conclusion: Redefining the Future of Television Investment

The SATC thought experiment proves one point: what worked once can work even bigger today when paired with the creator economy’s infrastructure. A single show could generate tens of millions in supplemental revenue while solving the residuals crisis and deepening audience loyalty.

But the bigger story is this: what happens if studios apply the Business Casual Method not to one flagship, but to every show they greenlight?

Creators like Alex Cooper have already proven the model. With nothing but a podcast and a YouTube strategy, she scaled from unemployment checks to a $160M empire in six years. If one individual can generate that level of value, what’s stopping legacy studios—with vast IP libraries, built-in marketing muscle, and global distribution—from multiplying that impact dozens of times over?

The opportunity is staggering:

• Every prestige drama could become its own mini YouTube empire.

• Every sitcom or unscripted hit could monetize its community long after the finale.

• Every kids’ show could mirror the Disney playbook, compounding into decades-long franchises.

This isn’t marketing—it’s infrastructure. And it scales. The same 0.32% budget allocation that works for one show works across the entire portfolio, creating a library of permanent, monetizable assets.

The strategic question for executives isn’t whether the Business Casual Method works. It’s: how much longer can you afford to leave this value untapped, while independent creators like Alex Cooper outpace billion-dollar studios?

For legacy media companies, the next show greenlit isn’t just television—it’s the seed of a billion-dollar digital asset. The future isn’t prestige vs. profit, it’s integrated media empires that compound cultural and financial value for decades.

APPENDICES

Appendix A: Production Cost Analysis

Description:
This appendix outlines the incremental production investment required to implement the Business Casual (BC) Method. The strategy leverages existing production downtime and low-cost equipment to generate supplemental content without disrupting the primary shoot.

Per-Episode Incremental Investment:

• B-crew producers (prorated): $11,538

• Dedicated post-production: $2,000

• Equipment costs: $0 (modern smartphones + existing gear)

• Total per-episode addition: $13,538

Six-Season Investment Breakdown:

• Season 1 (12 episodes): $162,456

• Season 2 (18 episodes): $243,684

• Season 3 (18 episodes): $243,684

• Season 4 (18 episodes): $243,684

• Season 5 (8 episodes): $108,304

• Season 6 (20 episodes): $270,760

• Total Investment: $1,272,572
  

Content Output Per Episode:

• Behind-the-scenes during setup/breakdown

• Cast interviews during meal breaks

• Fashion breakdowns and styling segments

• Character analysis and relationship deep-dives

• Blooper reels from daily shoots
  Average: 4 videos per episode = 376 total videos

Appendix B: Audience Conversion Funnel

Description: This appendix models how the television audience translates into a digital subscriber base through the discovery funnel. The assumption is that a fixed percentage of TV viewers engage with supplemental content, driving scalable and predictable digital community growth.

Conversion Assumptions:

• 60% of the TV audience discovers supplemental content

• 50% of that exposed audience clicks through to the YouTube channel

• 33% of channel visitors convert into subscribers

• Effective conversion rate: ~10% of total TV audience

Funnel Flow (per season):

1. TV Audience (Top of Funnel): Viewers start with the core show

2. Discovery (60%): A majority engage with clips, reaction videos, or promos

3. Channel Visits (50%): Half of those exposed actively visit the YouTube channel

4. Subscribers (33%): One-third of visitors lock in as recurring subscribers

Seasonal Conversion Outcomes (S1 = 3M, S6 finale = 10.6M):

Post-Show Stabilization:
~1.0–1.1M subscribers are maintained long-term through legacy content engagement, representing the durable community created by the finale-level audience.

Appendix C: Revenue Stream Calculations

Description: This appendix calculates the 20-year monetization potential of SATC’s supplemental content across four revenue streams. The model uses industry-standard assumptions for YouTube RPM, sponsorship density, merchandise conversion, and licensing opportunities.

AdSense Season-1 Example (S1 baseline = 3M):

• 48 supplemental videos averaging 225K views each = 10.8M views in Year 1

• Year 1 revenue: 10.8M × $4 RPM = $43,200

• Years 2–11: 2.16M views/year × 10 = 21.6M views → $86,400

• Years 12–20: 1.08M views/year × 9 = 9.72M views → $38,880

• 20-year AdSense total per Season-1 batch: $168,480

Scaled 20-Year Totals (All Seasons, All Streams):

• AdSense (gross): $4.12M

• Sponsorships: $9.25M

• Merchandise (gross): $4.12M

• Licensing/FAST: $5.23M

Total Gross: $22.73M

After Deductions (MAGR Net):

• AdSense net (55% of $4.12M): $2.27M

• Sponsorships: $9.25M

• Merch net (50% of $4.12M): $2.06M

• Licensing/FAST: $5.23M

• MAGR Net: $18.81M

Appendix D: ROI Calculations

Description: This appendix compares the return on investment (ROI) of supplemental content production to traditional marketing spend, both at the base case and an upside scenario reflecting stronger engagement.

Base Case (Net, MAGR lens):

• Revenue (net): $18.81M

• Investment: $1.2726M

• ROI: (18.81 − 1.2726) ÷ 1.2726 ≈ ~1,378%

Gross ROI (Marketing-Budget Lens):

• Revenue (gross): $22.73M

• ROI: (22.73 − 1.2726) ÷ 1.2726 ≈ ~1,686%

Upside Case (anchored to 10.6M finale):

• Gross: $55–65M

• Net: $45–52M

• Gross ROI: ~4,200–5,000%

• Net ROI: ~3,400–4,000%

Appendix E: Digital Residuals Compensation Calculations

Description: This appendix demonstrates how supplemental content revenue can be distributed through a Modified Adjusted Gross Receipts (MAGR) waterfall, mirroring television backend participation. It shows how the BC Method creates sustainable, performance-based payouts for talent and crew, solving the residuals crisis of the streaming era.

Step 1: Gross Receipts

• $22.73M (AdSense + Sponsorships + Merch + Licensing)

Step 2: Deductions

• YouTube platform fee (already reflected in AdSense net RPM)

• 50% merchandise COGS

• MAGR Net: $18.81M

Step 3: Operating & Ownership Deductions

• Network/studio share: 30%

• Production overhead: 15%

• Digital crew & channel ops: 10%

• Reserves/other: 5%

• Leaves 40% for backend participations

Backend Pool:

• $18.81M × 40% = $7.52M

Talent Compensation (Option A — crew participates inside backend):

• Showrunner/EP (15 pts): $1.13M

• Series Regulars (0.5% each): $37.6K per actor

• Writers’ Pool (5%): $376K

• Technical Crew (10%): $752K

Alternative (Option B — crew as top-line share):

• If crew is paid 10% of gross ($22.73M = $2.27M) before backend, the backend pool drops to $6.48M, and other payouts scale proportionally.

• This option strengthens crew participation but reduces the writers’ and actors’ backend shares.

Residuals Comparison:

• Streaming residuals today: $50–$200 one-time

• SATC YouTube model (3M → 10.6M): $38K–$100K+ per series regular over 20 years

• Improvement: 200–500× higher

Key Insight: The BC Method transforms $1.27M of incremental investment into a self-sustaining ecosystem where backend residuals flow predictably for 20+ years. Unlike streaming’s current “flat-fee” structure, this model ensures that every stakeholder participates in long-term success.

When Toy Story revolutionized animated cinema in 1995, it sparked a cultural phenomenon that I experienced firsthand. Within years, I found myself dressing as Woody for Halloween, proudly carrying a Buzz Lightyear lunch box to school, and eagerly persuading my parents to take me to the Buzz Lightyear ride at Disneyland. Remarkably, nearly three decades later, today's children can still have virtually the same experience, a testament to Disney's mastery of leveraging intellectual property through a comprehensive flywheel strategy.

In the streaming era, platforms face a critical challenge: while conventional wisdom prioritizes watch time and engagement metrics, I previously argued that prestige content isn't always compatible with this approach. The true value lies not in hours watched but in emotional connection—something Disney understood long before streaming existed. Media companies must now optimize their digital presence across owned, shared, and earned channels to monetize audiences effectively.

From Content Libraries to Experience Ecosystems

The future of streaming requires investment in world-building that increases customer lifetime value. Rather than necessarily building physical parks (though imagine a White Lotus hotel experience or a chaotic Tony Soprano Cadillac ride), platforms should leverage digital ecosystems to create immersive experiences. Stranger Things has already demonstrated this potential by expanding into gaming and themed playlists. By reverse-engineering the viewer experience and designing content as part of an experience economy, creators can transform shows into living ecosystems that remain culturally relevant for decades.

Consider how Patreon enables creators to monetize audiences through tiered membership models and exclusive content—their entire model focuses on monetizing fandom around creators, not the content itself. Similarly, Spotify now provides artists with ecosystem tools like direct Ticketmaster integration and merchandise links, transforming passive listeners into engaged, long-term fans. There's proof that nostalgia sells - today, you can buy a Grateful Dead Panther t-shirt on Ebay for $2400.

The ROI of "Emotional Connection"

For media companies, the true value of their most invaluable content lies in the engagement and fandom surrounding it. SpongeBob SquarePants exemplifies this approach, generating over $13 billion since 1999 and becoming Paramount's most profitable property in their Consumer Products division since 2007. This beloved character has transcended television to become a cultural touchstone through theme parks, video games, merchandise, and even Broadway.

Disney's Q1 2025 financials support this approach, showing Experiences revenue up 3% year-over-year while maintaining strong profitability, alongside streaming services achieving profitability with substantial subscriber growth. Netflix is now following suit with Mastercard partnerships for live immersive experiences that transform viewers into participants, such as "Bridgerton" balls and "Squid Game" competitions.

A New Framework for Streaming

The streaming industry stands at a crossroads where traditional success metrics no longer capture true content value. Media companies can transform single shows into thriving ecosystems through three interconnected strategic shifts:

1. Redefining success beyond hours watched - Moving from engagement metrics to measuring emotional connection and long-term value

2. Developing stories with universal foundations - Creating content with timeless themes that can resonate across generations

3. Aligning production with platform infrastructure - Designing content with its entire ecosystem in mind from inception

This holistic approach recognizes that a series' journey from discovery through nostalgia creates compounding value over decades, not just quarters. The future belongs to platforms that understand content isn't just what viewers watch—it's what they inhabit, share, emote and revisit throughout their lives.

The Content is the Product: A New Business Model for Television in the Streaming Era

While linear TV networks maximized viewership with advertising subsidizing inevitable churn, the subscription-driven streaming world demands a different approach.

To optimize returns on emotional connection, we must view content as a complete product offering. In restaurant terms, the show is the signature dish drawing customers in, but the profit engine is the ecosystem around it. Just as restaurants earn their highest margins on drinks and appetizers—not main courses—streaming platforms should create high-margin "complementary offerings" surrounding their core content.

Supplemental content like official podcasts, merchandise, and virtual events aren't just enhancements—they're financial assets that subsidize production costs and drive profitability long after a show concludes. This transforms series from cost centers into discrete business units with dedicated P&Ls.

Rather than solely chasing watch time, platforms should develop integrated experiences monetizable throughout the customer lifecycle. By optimizing journeys across paid, owned, and earned channels, platforms can simultaneously reduce churn, increase average revenue per user, and build sustainable business models.

The Streaming Content Flywheel for TV Series

I've created a flywheel model to visualize how we should view every customer experience when they consume television. This flywheel comprises four key components beyond the core viewing experience:

1. Discovery: The initial touchpoint ecosystem that converts cultural awareness into viewer intent. This includes repurposed content snippets, memes, viral trends, and merchandise that create multiple low-comprehension yet intriguing impressions. Consider the "mob wife" aesthetic trend that swept across TikTok, introducing many to The Sopranos without them watching a single episode.

2. Active Engagement: The real-time participation layer where passive consumption transforms into active contribution. This is where viewing becomes a social experience rather than a solitary activity, with audience theories and interpretations becoming part of the entertainment value itself—like a subreddit about the series or trending discussions on social platforms.

3. Deep Investment: The emotional and financial commitment phase where viewers transition from passive observers to fans or super-fans. This represents the monetization of deeper engagement through supplemental content, from film location tours to brand collaborations, where significant revenue opportunities emerge beyond subscription fees.

4. Long-term Value: The nostalgia engine that transforms completed content into living cultural artifacts. This is where content transcends its original medium to become a reference point in viewers' personal timelines and shared cultural language, creating sustained value long after a series concludes.

Historically, television shows evolved into franchises only after demonstrating critical success through two or more seasons. However, media companies should now consider franchise potential during the initial development process, designing content ecosystems that anticipate and nurture passionate fan engagement from the outset.

Case Study: The Enduring Value of The Sopranos

Today's reality reveals a significant oversight: streaming platforms neglect hundreds of thousands of fans who engage with their original content outside the streaming apps themselves. From character analyses to episode recaps, we see an increasing number of viewers having meaningful experiences with long-form content either to extend engagement beyond the platform or to remain connected to cultural touch points without committing to watching the actual shows.

The Sopranos provides a compelling example of this phenomenon. Twenty-six years after its premiere, the show maintains remarkable cultural relevance and fan engagement:

• YouTube channels featuring Sopranos season recaps attract ongoing viewership

• Merchandise sales continue to continue to sell decades after the show's conclusion

• Podcast discussions of individual episodes attract hundreds of thousands of listeners

• Reddit communities maintain active daily discussions about character motivations and plot interpretations

• TikTok clips featuring memorable scenes regularly go viral among younger audiences who never watched the original broadcast

The missed opportunity lies in media companies' inability to capture this recurring fan engagement and incorporate it into the customer journey. When viewers develop emotional connections with content, they inevitably seek additional touch-points outside the streaming platform itself—on YouTube, TikTok, Reddit, and other platforms. Media companies could easily direct these fans to official owned channels with minimal investment.

For example, if a Sopranos fan wants to watch an in-depth analysis about Dr. Melfi's therapeutic approach, they should be able to find this content on an official Sopranos YouTube channel, with revenue flowing back to the content creators and HBO.

From Passive Viewers to Revenue Generators: Monetize the Existing Micro Audience

Media companies focus too heavily on total subscriber numbers rather than maximizing return from engaged fans. Research supports the value of engaged audiences: the 90-9-1 rule suggests 10% of viewers engage beyond passive watching, while Harvard Business Review reports well-managed communities have the potential to achieve 50% lurkers, 23% contributors, and 27% creators.

By monetizing fan-created content around original IP, streamers can transform passive viewers into engaged fans who generate higher average revenue and offset acquisition costs for broader audiences.

From a product management perspective, losing our most engaged fans to external platforms signals the need for a fundamental strategy shift. We must move beyond traditional success metrics and reimagine content production from the ground up. While streaming platforms shouldn't necessarily become social networks, designing sustainable experiences and leveraging external ecosystems around television properties from inception can develop revenue streams that persist for decades rather than seasons, improving both platform recurring revenue and creating new opportunities for the cast and crew.

Reinventing the Writers' Room: TV Development for the Digital Age

In Q4 of 2024, Warner Bros. Discovery restructured to separate their linear networks from production studios, reflecting a broader industry shift as prestigious events like the Oscars migrate from traditional broadcast to streaming platforms. This transition marks a critical inflection point for television production.

A Vox video essay examined how the economic foundation of television has transformed. In the pre-streaming era, TV writers enjoyed stability through 40-week network seasons and substantial rerun residuals. Streaming disrupted this model by introducing shorter seasons (typically 6-8 episodes), extended production hiatuses, and minimal one-time payments. While initially offering creative freedom, this shift has ultimately created economic instability, transforming writing from a reliable middle-class profession into precarious gig work—a key factor driving the historic 2023 writers' strike.

If streaming platforms begin viewing television series as holistic products that appreciate in value through supplemental experiences, the development process could evolve in beneficial ways. Involving production teams in creating content around shows—from consulting on merchandise to developing additional material—extends employment timelines and provides more consistent income for cast and crew members.

Beyond Star Power: The Perils of Celebrity-Driven Development

This approach creates space for emerging talent rather than solely focusing on established stars. Current media executives often prioritize celebrities with substantial social media followings to drive brand awareness, considering loyal audiences among a show's most valuable assets. However, this strategy proves counterproductive when a show centers on celebrity without substantive storytelling or character development.

The Idol exemplifies this pitfall, though it paradoxically demonstrates how even flawed content can generate short-term business value. Despite backing from the creators of Euphoria, the show was primarily developed by Abel Tesfaye (The Weeknd). High staff turnover and inconsistent creative direction signaled deeper issues, but the fundamental weakness was insufficient story foundation.

The underlying logic seemed plausible—perhaps executives identified an overlap between The Weeknd's fanbase and Euphoria viewers. Yet while Euphoria explores intense themes through a distinctive visual style, its success stems from character-driven storytelling performed by a cast perfectly suited to embody complex character arcs. Euphoria fostered opportunities for unknown actors who achieved fame after the show, not before. The series was designed to elevate its performers into household names rather than relying on pre-existing celebrities.

Nevertheless, The Idol's notoriety generated significant social media engagement, with individual posts accumulating over a million views. While public relations teams might have struggled to manage brand perception, such controversy creates opportunities to humanize streaming platforms through supplemental content on external channels. There remains untapped potential to monetize audience engagement with this additional content—viewers aware of cultural phenomena without committing to full series viewing. For example, streaming services could derive business value from trends like the "rat-tail hype" by converting casual cultural participants into subscribers.

Beyond Viewership: The True Valuation of TV Series

With this new approach, streaming platforms must shift focus from short-term metrics to holistic business value throughout the entire customer lifecycle. Different content serves different purposes within this ecosystem—shows like The Idol generate cultural buzz and brand awareness at the top of the funnel, while iconic series like The Sopranos should be optimized across all facets of the customer experience to maximize long-term return on investment.

As an exercise, I've begun developing a framework to properly calculate the true ROI of content like The Sopranos as standalone business entities.

While implementation challenges exist, particularly with the "Frankenstein" systems resulting from industry mergers and acquisitions, the fundamental truth remains: targeted micro-audiences represent significant untapped potential for long-tail profitability. The mathematics of this approach is compelling, as supported by MIT Sloan Management Review research.

At its core, television series should be treated as products with unique customer experiences. The ultimate value of IP isn't just the content itself—it's the engaged audience that comes with it and the emotional connections that endure long after viewing.

Looking Ahead: From Content to Experience

My upcoming article will explore 'The Content as a Product Business Model' — a comprehensive framework offering tactical and technical strategies for optimizing the TV series customer lifecycle. Through data-driven analysis and theoretical modeling, I'll demonstrate how streaming platforms can strategically leverage supplemental content beyond their native environments to create sustainable engagement ecosystems and accelerate growth.

As we move forward, the most successful streaming companies won't just be those with the largest content libraries, but those that most effectively transform passive viewers into passionate fans and create new customer journeys to consume content that was originally not meant for binging. The future of streaming isn't measured in hours watched—it's measured in lives touched, memories created, and cultural landscapes shaped.

While streaming has revolutionized content access, it's disrupted television's power to create electric moments of collective experience. The metrics driving streaming success - watch time, completion rates, binge behavior - miss something crucial: the cultural gravity that once made television series generational touchstones. These traditional metrics have instead fueled an industry-wide pattern of serial churners who subscribe, binge, and cancel, creating an unsustainable cycle that undermines both cultural impact and business growth.

This two-part exploration examines why today's tv shows, despite their sophisticated storytelling, rarely achieve lasting cultural impact, and how streaming platforms can evolve beyond conventional metrics to foster deeper audience connections and drive customer lifetime value in this new landscape.

From Water Cooler to Watch Queue: Television's Cultural Transformation

Let’s rewind the clock to 9 PM on a Sunday in 2007. Millions of Americans are settling in to watch The Sopranos, knowing tomorrow they'll gather around office water coolers to dissect every moment. This wasn't just appointment viewing; it was a cultural ritual that bound audiences together through shared anticipation and collective discovery. Today, that communal experience has largely vanished. In our streaming world, viewers watch in isolation, on their own schedules, fragmenting what was once a unifying cultural phenomenon.

Streaming platforms have mastered immediate gratification, but at a cost: they've eroded the deep emotional investment that once transformed shows like Game of Thrones into cultural phenomena. Instead of fostering anticipation for "the next episode of my show," these platforms have conditioned viewers to approach content with the mindset of "I need something to watch." It's like choosing between fast food and an omakase dinner – both satisfy hunger, but only one creates an experience worth remembering and sharing. When content becomes interchangeable entertainment rather than a cherished ritual, we lose the magic that turns watching television from a passive activity into a meaningful journey.

The Numbers Game: When Streaming Success Kills Cultural Impact

The challenge lies deeper than content quality - it's rooted in how stories reach and resonate with audiences. In 2017, when Netflix's former CEO Reed Hastings declared sleep as their main competitor, he revealed an industry-wide fixation on maximizing watch time above all else to compete in the blitz-scaling era of Silicon Valley. However, in today's saturated media landscape, where content options are endless and attention is scarce, this metric-driven approach overlooks a crucial truth: viewers don't just return for mere entertainment.

The streaming revolution began with promise. When Netflix pioneered the market, it freed creators from traditional network constraints, following HBO's groundbreaking model of allowing more experimental, authentic storytelling. Shows like House of Cards and The Crown emerged from this creative freedom, backed by seemingly infinite content budgets. Yet these series, despite their quality, haven't achieved the cultural staying power of traditional network shows like The West Wing or Downton Abbey.

In 2018, this paradox became evident when HBO was acquired by AT&T. Warner Media's leadership, determined to outperform Netflix, learned a costly lesson: prestige content isn't always compatible with the metrics-driven pursuit of mass audience engagement and increased watch time.

Beyond Binge Metrics: The Hidden Cost of Instant Gratification

When it comes to binging versus weekly programming for tv series, the key difference isn't content quality – it's a distinctively different customer experience and this idea isn’t new. In a 2015 Vox article, media critic Emily St. James and media scholar, Scott Eric Kauffman contrast the experience of watching Halt and Catch a Fire when it aired on AMC in 2014 versus binging the series on Netflix. Kauffman concluded, “Binge-watching fundamentally changes the basic unit of cinematic storytelling.”

The shift toward binge-watching comes with hidden costs, despite satisfying our appetite for immediate gratification. This model inadvertently undermines the creator incentive for careful world-building and character development that historically created lasting cultural impact. When shows drop all at once, they often prioritize immediate engagement over sustained storytelling—the kind that builds deep viewer investment and ultimately drives long-term value. As streaming platforms increasingly measure success through hours watched, creators face a crucial dilemma: Does this environment still incentivize the creation of generational phenomena like Star Wars or Game of Thrones?

An article from Rolling Stone highlights how streaming has fundamentally altered series structure, with seasons shrinking to roughly ten episodes. While this format can yield tighter storytelling, it often results in what feels like extended movies rather than true television series. This condensed approach, while efficient, may limit a show's potential to embed itself in the cultural consciousness. Though algorithms excel at optimizing watch time, they struggle to create the environment necessary for transforming casual viewers into devoted fans.

Consider The Sopranos, with its expansive seasons of 13+ episodes. The series wasn't designed for binge-watching—it was crafted as a weekly ritual that spanned years, creating a shared social experience akin to following a sports team. Each week brought new anticipation, discussions, and communal speculation about which character might meet their end. This extended engagement fostered a deeper connection with the story and its world, building a loyal audience that remained engaged for almost a decade while it aired and invested for decades to come.

Television's "golden age" illuminates why this matters. Weekly programming required shows to maintain a compelling narrative rhythm that could sustain viewing rituals, while advertising revenue demanded consistent quality. In addition, there’s a financial incentive to keep the show running for at least five seasons for syndication residuals and cultural relevance for franchise potential. Viewers didn't merely tune in to see plot resolutions—they returned week after week. Whether it was the multi-year investment of the “will-they won’t-they” plot line between Ross and Rachel in hit sitcom Friends or the cathartic comfort of dark humor about the mundane routines of daily life on Seinfeld, these stories forged emotional bonds. These characters became an extension of a home and created a world viewers could inhabit between episodes and across years. The weekly cadence wasn't just about creating viewing habits; it was about nurturing cultural phenomena.

Reimagining the Digital Water-Cooler: Building Tomorrow's TV Culture

This model of sustained engagement offers crucial insights for today's streaming landscape. The solution lies in expanding how we define successful content engagement in a digital world. Great stories aren't just about plot; they're about creating immersive universes that viewers want to inhabit and discuss. Game of Thrones exemplifies this approach – its meticulously crafted world fostered generational fandom through authentic world-building that prioritized storytelling over immediate metrics.

While "water cooler" conversations have evolved into digital discussions and fan theories, most platforms have failed to capitalize on this transformation. The future of television demands that streaming platforms evolve from content libraries into experience architects. By leveraging cost reducing technologies to optimize content repurposing and micro-audience targeting, they can create environments where stories become cultural touchstones that sustain engagement beyond the immediate release window.

In today’s content consumption landscape, it demands we think beyond traditional episodes – it challenges us to build expansive story worlds that deepen audience investment. By crafting supplemental content that enriches rather than fragments the viewing experience, creators and platforms can leverage digital ecosystems to transform casual viewers into engaged communities. This approach doesn't just adapt to modern consumption patterns; it creates lasting narrative universes that resonate across generations.

Television's Strategic Reset: Where Classic Fandoms Meet Modern Engagement

The contrast between traditional television and today's streaming landscape reveals a striking paradox: despite an unprecedented volume of content, platforms struggle to maintain subscriber loyalty. This challenge presents two critical imperatives for the industry:

1. The Legacy IP Alchemy: Creating Evergreen Content Engines

The first challenge is transforming how we present classic series in the streaming era. Shows like The Sopranos weren't designed for binge viewing, yet their storytelling power remains undiminished. The opportunity lies not in reinventing the wheel, but in leveraging the rich ecosystem of fan-created content and discussion that already exists across digital platforms.

Classic series have inspired vast libraries of analysis videos on YouTube, scene breakdowns on TikTok, and ongoing discussions across social media platforms. Rather than building new infrastructure, streaming platforms should focus on integrating with these existing communities and content. Fan-created episode guides, cultural analysis, and historical context already thrive on these platforms, offering natural pause points and deeper engagement opportunities for new viewers. By acknowledging and elevating this existing ecosystem, platforms can enhance the viewing experience while respecting both the original episodic structure and the organic way modern audiences discover and discuss content.

2. Breaking Media Company Silos: The New Methods of TV Development

The second imperative is revolutionizing how we approach new series development. Future shows must be conceived not just as linear narratives, but as expansive worlds that invite exploration and community engagement. This requires a fundamental shift in how stories are structured, moving beyond traditional episode formats to create narrative architectures that work seamlessly across both weekly and binge viewing patterns.

In today's digital water-cooler landscape, creating lasting cultural relevance means reverse engineering the viewer experience. Streaming platforms have the advantage at monetizing micro-audiences, and with software solutions to optimize content repurposing, there's an unprecedented opportunity to design television content around an experience economy from the ground up. This approach doesn't just extend a show's lifespan—it transforms it into a living, breathing ecosystem that can thrive for decades after its initial release.

In my next article, Content as a Product: The Future of Consuming Television in a Streaming World Part II, I’ll dive into specific strategies and case studies on how these concepts could be applied to any streaming platform that serves tv series in their catalog.