The Flix of Theseus

In this issue:

• The Flix of Theseus—It's fun to review the story of Netflix as one situation after another where they're clearly winning in the short term, but there are longer-term questions about their viability—from Blockbuster, studios, international competitors, their own content budget, there's always been a terminal value bear case. And they've won by continuously pushing back the point at which they stop growing.
• The Dollar—Share shift in currencies happens very slowly, but it's also hard to reverse once it starts in earnest.
• Full-Service VCs—Build-your-own-a16z.
• Advertising—When companies find a way to cut costs, they can cut prices or just spend more on sales and advertising. And they have an obvious first choice.
• Tiny IPOs—Exchanges and regulators finally focus on small-time crooks.
• Partnerships—Big social networks can afford to play nicely with their punier suppliers.

This issue of The Diff is brought to you by our sponsor, Trata.

The Flix of Theseus

Netflix is one of those companies whose long-term stock chart prompts lots of people to imagine how much money they'd have today if they'd put $1k into the IPO. ($838k, if you're curious.) But, more so than other long-term buy-and-hold investments, they'd own something very different from what they started with. Some companies set up a model, and just iterate it—Costco is pretty much the same model it's had since it introduced the Kirkland brand and adding gas stations in 1995, though it's added locations and gotten better at the real estate to subscription revenue with everything else as a loss leader pricing setup. With Netflix, the company you invested in in 2002 was a clever logistics play—DVDs were cheaper to make and easier to ship than VHS, fulfillment centers could stock a bigger catalog than a network of physical stores (especially if they were stocking them in a space-efficient format), and the rental business was already well-established and well-loathed^[1]—any time your competition is describing their relationship with customers using the term "managed dissatisfaction," there's probably room for a competitor.

Someone holding Netflix shares today owns stock in a company that delivers media a different way, sources it from a new mix of suppliers, monetizes it through both subscriptions and ads, and operates in every country not subject to US sanctions.^[2] Of course, this business ship-of-Theseus was built according to the same specifications each time, just using the best raw materials they could access. In 1997, when Netflix launched, the typical Internet-connected home was going online at a top speed of 56kbps, and going online meant tying up a phone line. Shipping a DVD with 4.7GB of data over three days was 150kbps. So Netflix, then and now, delivers movies in a digital format using the highest-bandwidth option available. The original Netflix model certainly benefited from the fact that studios were cranking out lots of DVDs, but didn't strictly depend on it—and the fact that people could rewatch the same movies over and over again probably contributed to the shift to sequels and spinoffs over original IP, which improved studios' negotiating position relative to distributors. A pure subscription model made a lot of sense when Netflix's audience was relatively high-income, but at 300 million viewers they're pretty middlebrow by definition, and the way you monetize a middlebrow audience is mostly by interrupting their viewing with commercials. So the Netflix of today is in some sense a direct descendant of the original business, but it's gotten big enough to converge with the rest of the media industry because it's a much bigger piece of that industry.

As of last Friday, they've now participated in another big media rite of passage: participating in a bidding war for another media company that has a great content library and relatively lackluster recent performance. (And as of this morning, they’re bringing back another one, potential participation in a bidding war with someone launching a hostile takeover.) This is a sufficiently fraught task that there's a whole book on why not to do it, co-written by a TMT banker.^[3] Your hypothetical Netflix buy-and-hold investor started out owning a company that exploited certain weaknesses in Blockbuster's rental model, and now they own a full-stack media business that makes movies and shows, buys some more, and owns its distribution and monetization.

It's a pretty strange trip, but Netflix is an interesting story as a company that's always been in a fight to prove that its terminal value isn't zero. Early in the company's existence, the short bet was pretty obvious: Netflix's pricing was something Blockbuster couldn't replicate without blowing up important parts of their business model, but if Netflix worked, Blockbuster would just copy them and crush them (in 2004 when Blockbuster finally launched DVD by mail, Netflix drew down 70% on this fear). Blockbuster did eventually copy Netflix, but it's just very hard corporate politics to champion an idea that will both lose money on its own and kill the pricing power of the company's existing cash cow business, so they were late, and timid about cannibalizing their business, and Netflix was more agile.

And Netflix was, of course, Netflix because the ultimate plan was to do streaming; DVDs were a convenient hack to get cash flow and brand recognition. This was the explicit plan so far back that the early articles about it feel archaic: Reed Hastings was telling Wired in 2002 that in five to ten years, they'd offer "Downloadables." (In an early instance of Netflix's habit of sometimes offering investors guidance that's both wildly bullish at the time and retrospectively conservative, he overestimated that timeline, and Netflix's streaming service would actually launch in a little over four years.)

Once they started streaming, the Blockbuster critique applied again: if streaming doesn't turn out to be a big deal, they've wasted a lot of money on a dud. If it does turn out to be big, the big IP owners will all just launch streaming services of their own, and Netflix will be left with the dregs. Which is one reason Netflix invested in originals so early—they knew they needed a backup plan to deal with the movie industry well before the movie industry really understood what it had to do to compete with Netflix. Part of what Netflix is trying to do with originals is attract an audience that's specifically interested in Netflix-owned IP, but another reason is to fill in gaps in whatever bundle of media consumption includes some third-party content they're licensing. As they scale, there are more of these complementary opportunities, where they simultaneously build and buy a content library to appeal to a new audience.

For a while, one of the bull/bear debates on Netflix was over international markets. Would they win everywhere? Would they be a niche product in some countries? And would paid streaming even work as a business in jurisdictions that didn't care very much about piracy? They've had some success in turning regional hits into global ones, but in many places their biggest advantage is pure scale. If they have the biggest audience, they're the default highest bidder for content.

Now, there's a different debate, over how much share short-form video will eventually take. That ends up being an AI-adjacent question, both because generative video will take share and because one of the strengths of short-form video apps is that they collect more user feedback per minute of watch time, and can thus customize feeds much more than a service like Netflix can. And that argument is absolutely correct. If Netflix sees someone watch a two-hour movie all the way to the end, they can assume that viewer liked that movie, but they won't know much about why. If someone watches videos on TikTok that average thirty seconds of runtime, and promptly skips the boring ones,TikTok gets 360 datapoints on what that user likes and can serve them a lot of it. In other words, interaction frequency matters.

But at that point, Netflix gets to present itself as a defense against brainrot: it's an endless supply of video content, but at least you can be confident that there was lots of human input along the way. Netflix is the rare public company whose bull case is the way that it's not as AI-first as its competitors. Netflix is certainly using plenty of GPU-hours to provide all those recommendations, but it's still recommending content that has a human in the loop. Buying an existing studio (particularly one focused on prestige TV), or at least attempting it, is a bet on more of that. And it will end up being a bet on some parts of the movie industry Netflix has played a big role in killing: they'll annoy lots of talent if theatrical releases aren't the default, but that will eventually mean that they can actually price-discriminate a bit; if getting a big theatrical release is attractive to the talent, Netflix can put a dollar value on it, and send everything else straight to streaming.

If there is a case for a media merger to work out, it's usually one where a company operating at a more fragmented layer of the supply chain gets bought by someone in a more concentrated layer. Subsidizing the complement to a product with monopoly pricing tends to produce good returns and good PR. Having more share of more layers of the media stack puts Netflix in a good position to ride out whatever the next media trend is. And it's yet another case where Netflix pushes out the date at which it's plausible to say that they'll saturate their market.

A Word From Our Sponsors

Imagine being a fly on the wall while two industry leading portfolio managers candidly debate your largest position. That's the Trata experience.

We're building the world's largest repository of buy-side knowledge. By connecting 125+ funds with over $150bn in combined AUM and counting, Trata captures the nuance that expert call transcripts and sell-side reports miss. Our platform hosts and transcribes anonymous 1:1 debates between the sharpest bulls and skeptics, giving you a front-row seat to a real-time evolution of the institutional thought process around a name.

Featured by Matt Levine and backed by YC and others ($3M+ raised), Trata is rapidly becoming the standard for checking your blind spots. Stop relying on commoditizing expert networks and the same podcasts everyone else listens to, start listening to the raw, unfiltered dialogue of the market’s highest-signal participants.

Elsewhere

The Dollar

The dollar will be the world's reserve currency for a long time, but not necessarily forever. One of the forces that maintains that network effect is that cross-border lending tends to happen in whatever currency both sides can use, and that this creates dollar-denominated obligations in other parts of the world. So, if you're tracking the dollar's long-term dominance, one thing to look for is borrowers and lenders shifting to something else. So it's notable that Asian companies are pricing more of their borrowings in Euro. Bond sales are a slightly lagging indicator of what companies' priorities are, so it's entirely possible for this to mean-revert, but it'll be worth paying attention to which global currency borrowers think is the default one.

Full-Service VCs

Money is by necessity the most commoditized product around, which means that it's tough for companies to differentiate themselves when they're in the business of slinging money around. This almost tautological observation helps explain a surprising number of phenomena, from bank branch networks to the fact that some of the most heavily-marketed fictional characters exist to sell insurance (the GEICO gecko, Mayhem, Progressive Flo). For VC, it explains why venture firms are increasingly full-service, offering a package of cash plus things like recruiting, business development, marketing help, GPUs, and the like.

Big venture firms operate at a scale where they can offer a comprehensive menu. All they have to do is imagine their ideal founder, make a list of all the things that founder wouldn't necessarily be good at, and then build those functions internally. But the model also works if a VC takes one service they're uniquely good at, and builds up a fund around that. So pugnacious tech PR operator Lulu Cheng Meservey is raising a fund ($, The Information). From her perspective, this is a way to get performance-based compensation: if she does a good job at guiding her portfolio companies' PR strategy, she'll monetize that skill when they sell for more than they otherwise would have. For companies, it looks increasingly possible to build an à la carte a16z by raising from smaller funds that specialize in whatever specific value-added service they care the most about.

Advertising

Earlier this year, The Diff argued that when companies save money by using AI, they'll be reluctant to cut prices, and will try to grow instead ($), meaning that one of the best proxies for AI's impact on the economy is total ad spend. WPP has raised its estimate for global ad spending growth ex-US political ads in 2025 to 8.8%, up from 6% in June ($, WSJ). Ads-as-AI-proxy will be noisy (as illustrated by the fact that these numbers exclude the lumpy spend from American political campaigns), but so far they're moving in the right direction.

Tiny IPOs

Exchanges and the SEC are finally going after the endless stream of tiny IPOs from semi-fake companies ($, Barrons). The Diff covered these odd offerings two years ago, and it's surprising that they're still around and still getting manipulated. (Just in the last few days, Treasure Global, one of the IPOs that first put this on my radar, has risen up to 10x on no real news.) These IPOs are not a big deal in terms of their total impact on the market; put them all together and you probably have less free float market cap than the magnitude of a typical Mag7 stock's intraday valuation swings. But they look bad, and they're clearly targeting unsophisticated investors who might justifiably decide to reject equity markets entirely if they get ripped off. For now, they'll still provide some small-scale shorting opportunities, but it would be nice if these offerings disappeared entirely.

Partnerships

The news business has had ups and downs over time, but it turned out to be very profitable in a situation where news was part of the mix of information on a newsfeed or video stream, monetized through targeted ads. More than 100% of the financial upside of that model accrued to the companies that owned feeds and search results pages, while the businesses actually producing that news suffered. That's part of what drove the backlash against tech—it's hard for someone to be objective about an industry that's eliminating their job security! But that same profit mismatch means that it's pretty affordable for tech to play nicely: Meta is paying some large news organizations to license their content for Meta's LLMs. They don't strictly need to do this, because big stories get aggregated and reposted on free sources eventually. But if they can make their product slightly better and give some of their critics a financial interest in their success, the ROI can be worth it.

Disclosure: Long META.

Diff Jobs

Companies in the Diff network are actively looking for talent. See a sampling of current open roles below:

• YC-backed startup automating procurement and sales processes for the chemicals industry, which currently relies on a manual blend of email, spreadsheets, legacy ERPs, etc. to find, price, buy, and sell over 20M+ discrete chemicals, is hiring full-stack engineers (React, TypeScript, etc.). Folks with exposure to both startups and bigtech, but also an interest in helping real-world America with AI preferred. (SF)
• A hyper-growth startup that’s turning the fastest growing unicorns’ sales and marketing data into revenue (driven $XXXM incremental customer revenue the last year alone) is looking for a senior/staff-level software engineer with a track record of building large, performant distributed systems and owning customer delivery at high velocity. Experience with AI agents, orchestration frameworks, and contributing to open source AI a plus. (NYC)
• Well funded, Ex-Stripe founders are building the agentic back-office automation platform that turns business processes into self-directed, self-improving workflows which know when to ask humans for input. They are initially focused on making ERP workflows (invoice management, accounting, financial close, etc.) in the enterprise more accurate/complete and are looking for FDEs and Platform Engineers. If you enjoy working with the C-suite at some of the largest enterprises to drive operational efficiency with AI and have 3+ YOE as a SWE, this is for you. (Remote)
• A leading AI transformation & PE investment firm (think private equity meets Palantir) that’s been focused on investing in and transforming businesses with AI long before ChatGPT (100+ successful portfolio company AI transformations since 2019) is hiring Associates, VPs, and Principals to lead AI transformations at portfolio companies starting from investment underwriting through AI deployment. If you’re a generalist with deal/client-facing experience in top-tier consulting, product management, PE, IB, etc. and a technical degree (e.g., CS/EE/Engineering/Math) or comparable experience this is for you. (Remote)
• A transformative company that’s bringing AI-powered, personalized education to a billion+ students is looking for elite, AI-native generalists to build and scale the operational systems that will enable 100 schools next year and a 1000 schools the year after that. If you want to design and deploy AI-first operational systems that eliminate manual effort, compress complexity, and drive scalable execution, please reach out. Experience in product, operational, or commercially-oriented roles in the software industry preferred. (Remote)

Even if you don't see an exact match for your skills and interests right now, we're happy to talk early so we can let you know if a good opportunity comes up.

If you’re at a company that's looking for talent, we should talk! Diff Jobs works with companies across fintech, hard tech, consumer software, enterprise software, and other areas—any company where finding unusually effective people is a top priority.