Issue #41: The Back Catalog 💎

Issue 41
July 16th, 2026

I keep coming back to that because it's the part everyone is going to skip past. Three Netflix creator deals landed inside about six days, and if you only read the headlines you'd think the story is "Netflix wants YouTubers." That's been true for a year. The new thing is that the three deals have three completely different shapes, and the shape is where the money is. One creator got paid a fortune to leave. One is showing up on both places the same day. One handed over an archive and kept the originals running. Same buyer, same week, three different answers to the question of what your old videos are actually worth.

Which is a question most of us have never really had to answer. Your back catalog has always been sort of dead weight, right? You post it, it does its numbers, it slides down the page, it stops existing. This week it started looking a lot more like inventory.

Also this week: I'm genuinely excited to share that Trovio (the company behind this newsletter) is launching something HUGE for brands next week.

We've always been creator first, and this doesn't change that. The whole point is helping creators who normally wouldn't get a seat at the table earn a living the way the top 1% already do. So we built a brand dashboard that lets brands discover creators. Discover, not search, that's the important part. Instead of keyword filters and follower counts, we're pairing brands with creators who are already having real conversations with their audience, so a partnership feels like joining that conversation instead of dropping in a promoted post. Brands get creators who are already talking about the right things. Creators get paid for talking about what they were already going to talk about.

Wednesday's the official launch, but there's a preview up now on Product Hunt if you want an early look (and following along there helps us out when we go live). If you’re on the brand side and want more information, reply to this email and let me know – we’ll give you a special offer to get started. If you know someone who is on the brand side and think this sounds cool, please forward. 😃

In this issue:

Let's dig in.

Netflix isn't buying creators. It's buying shelf space, and the price depends entirely on whether you're allowed to keep your keys.

Start with the one that actually broke this week. On July 14, Netflix and Mythical announced that Good Mythical Morning, Mythical Kitchen, and Last Meals are coming to Netflix starting September 7. Here's the part that made me sit up: the episodes go up on Netflix day-and-date with YouTube. Same day. Not a window, not a delay, not "Netflix first." GMM posts every weekday, and starting in September it'll post to both places at once. Netflix is also taking the library of past episodes.

Think about what that means. Netflix is not asking Rhett and Link to choose. It's not even asking for a head start. It's paying to have the show appear on Netflix at the exact moment it appears for free on YouTube. The Hollywood Reporter framed why that's new: Ms. Rachel, Mark Rober, and Jordan and Salish Matter all repurposed content they'd already made and packaged it for Netflix. The Stokes Twins are handing over an archive. Mythical is the first one running daily episodes on both at the same time.

Now put it next to the Stokes deal. On July 9, Tubefilter reported that Alan and Alex Stokes, who have somewhere north of 140 million YouTube subscribers, are bringing an archive of their videos to Netflix starting July 18. The deal is non-exclusive. The videos stay on YouTube. Nobody is taking anything down. They're also developing a long-form show for Netflix set for 2027, but the archive piece is the interesting one, because it's the same library existing in two places and getting paid twice.

And then there's the one that goes the other way. Jay Shetty's On Purpose finally moved its video episodes off YouTube on July 13, over to Netflix and Spotify. Worth being precise here, because I've seen this get muddled: the deal itself isn't news, it was announced back in May and reported by Bloomberg and Variety as worth up to $100 million. This week is just when the videos actually walked. Older episodes and promo clips still live on YouTube, but new video episodes don't. That one's exclusive, and Variety reported three other companies were bidding in the nine-figure range for it.

So line them up. Exclusive, and a reported nine figures to make it exclusive. Day-and-date, non-exclusive, no window at all. Archive licensed, non-exclusive, originals untouched. One buyer, six days, three prices for three different amounts of control.

The thing I can't stop thinking about is that only one of those three deals required a creator to give anything up. Netflix clearly wants a creator library on its homepage. It's also licensing short-form from Penske, BuzzFeed Studios, Condé Nast, Hearst, and People Inc. starting August 3. But wanting the library and needing to own it are apparently two very different budgets, and Netflix is only paying the ownership price when it specifically wants you gone from somewhere else. The rest of the time it seems happy to just rent placement next to YouTube.

I want to be honest about the ceiling here, because I don't want anyone screenshotting this and walking into a bad meeting. These are enormous creators. 140 million subscribers. A show with a fourteen-year run. A podcast with five million YouTube subscribers and Michelle Obama on the guest list. Netflix is not licensing your 40,000-subscriber cooking channel next month, and nothing about this week says otherwise. The access is still going to proven, legible, easy-to-buy IP.

The bottom line: What changed isn't your odds of a Netflix deal. It's the default assumption underneath one. For a decade the mental model was that a real deal means signing away the thing, and the archive was just whatever was left behind on the channel. This week a very large buyer demonstrated, three separate times, that it will pay for a library it doesn't own, sitting next to the same library available free, sometimes on the very same day. Exclusivity is now a line item with a price tag, not the entry fee. If your work is your IP, the useful move is to stop thinking of your catalog as a channel and start thinking of it as rights you can sell more than once. Which means knowing what you actually hold. More on that at the bottom.

Three data points that tell a bigger story.

Tubefilter reported on July 14 on an analysis from TikTok, WPP Media, and System1 covering 1,217 paid TikTok ads across eight markets, which found creator-led ads carried a 3.35 long-term sales multiplier and were 23% more effective than brand ads for awareness and ad recall. The same report bundle cited Public First data putting TikTok behind roughly 153,000 UK jobs and 300,000 small businesses on TikTok Shop, with 84% of featured businesses reporting a sales boost. Here's why I put it in an issue about back catalogs: the interesting word is long-term. A multiplier that shows up after the campaign flight is a measurement of stuff that keeps working once nobody's paying for it anymore, which is the actual definition of an asset. The honest caveat is loud, though. This is TikTok and its agency partners measuring whether TikTok creator media works, and they have an obvious interest in the answer. It's evidence that creator-led formats have durable value, not evidence that any particular creator gets paid for that durability.

Sourcing Journal reported on July 2 on Statusphere's 2026 micro-influencer report, which tracked more than 68,000 creator posts and found brands increasingly treating that content less like short-lived social chatter and more like a durable discovery asset, feeding social SEO, AI answer engines, and paid amplification. That's the same shift as the Netflix story, just at the other end of the market and without the press release. A post used to have a half-life measured in hours. If brands are now valuing it as something that gets indexed, cited, and re-amplified for months, then the thing being bought is the shelf life, not the impressions. This one's vendor-framed and brand-side, so treat it as market direction rather than proof anyone's cutting bigger checks. But it does sharpen the thing worth arguing about in your next contract: if the post keeps working for a year, whose year is it? Usage rights and reuse windows stop being boilerplate the moment content is an asset instead of an ad.

Digiday reported on July 13 that creator-led livestreams from stadiums and FIFA fan zones have turned into a parallel World Cup broadcast layer, citing Twitch's Football Fest category and a Twitch survey where 44% wanted sports coverage that wasn't available on traditional media, and 29% said their fandom went up versus 2022. We covered the credentialed version of this back in Issue #35, when the NBA handed Kenny Beecham the footage and TikTok credentialed 30 World Cup correspondents. This is the uncredentialed version, and honestly it's the more interesting one. Nobody gave these streamers rights. They built an audience for the parts of the event the broadcast doesn't cover, and 44% of viewers turned up specifically because the official product didn't have it. Related, and worth stacking next to it: Digiday also noted on July 6 that livestreaming's growth is now personality-led rather than game-led, with Twitch's own CEO describing the platform as entertainers who may game rather than gamers first. Both point the same way. Live is turning into programming, and programming is a library.

What the platforms shipped this week and why you should care.

Tubefilter reported on July 13 that YouTube renamed and clarified its monetization guidance around "Generic or Repetitive Content," spelling out generic templates, unsatisfying or off-putting content, and AI personas used around sensitive topics. YouTube told Tubefilter this is a clarification of existing guidance, not a new policy.

Why this matters for creators: Do not read this as a crackdown, because it isn't one, and the "YouTube is demonetizing AI" takes going around are wrong. What it actually does is make the boundary legible, which is useful right when your archive is starting to have resale value. A library of templated, faceless, mass-produced uploads is worth roughly nothing to a licensing buyer and now carries clearer monetization risk on YouTube itself. If you're using AI in your workflow, the specific thing to add is the layer YouTube keeps describing as the difference: your own commentary, narrative, or point of view on top of the generated material. Go read the updated language before you scale an output pipeline, not after.

Meta disabled an Instagram feature that let users fold other people's content into AI-generated images, roughly two days after launching it, saying it "missed the mark." Separately, TechCrunch published a guide on July 9 to Meta's Muse Image feature, which can generate or edit images using photos from public Instagram accounts when the account is tagged. Private and under-18 accounts are excluded, and there's an opt-out under Instagram's sharing and reuse controls.

Why this matters for creators: If this feels familiar, it should. We wrote up almost exactly this sequence in Issue #29, when TikTok quietly switched on AI remixing across every video creators had ever posted and then paused it after the backlash. Same script, different platform, and the rollback came faster this time. That's the part worth updating your priors on: this isn't an incident anymore, it's a pattern, and the pattern is that reuse defaults ship first and get walked back only if enough people notice within 48 hours. So the practical move is boring and annoying. Go into Instagram's sharing and reuse settings and check your Muse opt-out today, then put a recurring reminder in your calendar to re-check your reuse settings after any big platform AI announcement. Defaults are where this gets decided, and defaults change quietly.

Tubefilter reported on July 14 that Agentio struck a Meta deal bringing its AI-powered creator advertising platform to Facebook and Instagram, letting brands run insights, creator matching, and live partnership ads across Meta and combine them with its YouTube tools. Agentio claims beta Meta partnership ads delivered 81% higher ROAS and 89% higher CTR than the same brands' non-Agentio creator partnership ads.

Why this matters for creators: Those numbers are Agentio's own, about Agentio, so discount accordingly. The signal that isn't marketing is structural: more of the matching between you and a brand budget is being done by a system that never sees your DMs or your pitch deck. Systems match on what they can parse, which means a clear niche, consistent output in that niche, and performance proof attached to your profile rather than living in a PDF you email people. This is the same pressure as Issue #39, just automated. Being good is necessary. Being classifiable is what gets you into the candidate set in the first place.

Go inventory your back catalog, because you almost certainly can't answer the first question a licensing conversation opens with.

The Netflix deals this week all started from the same place: somebody could describe, precisely, what existed and who was allowed to have it. Most creators cannot do that. I'd bet if a buyer emailed you tomorrow asking what's in your library and what you're free to license, you'd need a week and a lawyer to answer, and the honest answer for a lot of people would be "I'm not totally sure."

So spend an hour this week and build the boring document. Three columns.

First, what exists. Every series, format, or recurring bit you've made, with a rough episode count and date range. Not every video, the formats. "Fourteen years of a weekday show" is a sellable sentence. "About 900 videos" is not. If you've got a thing you've done 40 times, that's a format with a library behind it, even if you never called it that.

Second, what you actually control. This is the column nobody has. For each format: did you write it, shoot it, and own the footage? Was any of it a brand deal, and if so what did the usage clause say and has that window expired? Is there licensed music in it, which is the single most common thing that makes an otherwise valuable archive unlicensable? Any on-camera guests, and did they sign anything? Go pull two or three of your old brand contracts and actually read the rights section. I've watched people discover that content they thought was theirs is technically encumbered years later, and it's always at the worst possible moment.

Third, what's already non-exclusive. This is the mindset shift from the Stokes deal. Your default assumption should be that any place your work isn't contractually locked to is a place it can be licensed in addition, not instead. Netflix just paid for a library that stays on YouTube. If the biggest buyer in the world is comfortable with non-exclusive, you should be extremely suspicious of any smaller party who insists exclusivity is the only structure available. Exclusivity has a price, and this week put a rough number on what that price looks like when someone actually wants it.

None of this gets you a Netflix deal. That's not the point. The point is that this is the work you can't do in the two weeks after an opportunity shows up, and it's completely doable in an hour before one does.

That's all for this week. Go look at what's sitting in your archive, because it stopped being old content and started being inventory sometime while you weren't paying attention. And if someone forwarded this to you, sign up to get your own issue every Thursday.