Mark Pincus

1. Key Themes

Proven Better New: The Product Framework That Became a Religion at Zynga

Mark Pincus developed a three-part framework for product development that separates instinct from idea, and structured innovation from guesswork. The core insight is that human instincts are directionally right, but specific ideas are usually wrong — making systematic testing essential.

"Your instincts are right 95% of the time. Your ideas are wrong 75% of the time, or at best right 25% of the time. The framework of Proven Better New takes that philosophy and says, okay, what do we do with that? Let's isolate your innovation zone. Let's isolate that thing you have in your gut. And let's just test many, many ideas around that. And let's fail for the right reason, not the wrong reason." 00:00:06

The framework demands precision: "proven" only counts if it's proven on this platform, for this audience, for this experience — not something that worked in the 90s or on a different device.

"We have to be precise. We have to think at the pixel level of this experience. Proven is on this platform for this audience, for this experience. And you can look to what's been proven before as great sources of new ideas, but if it's not on this platform, it doesn't count as proven." 00:13:51

The Moral Arbitrage of Copying

There is a psychological and cultural resistance to copying that creates a real market opportunity for founders willing to get over it. Pincus frames this explicitly as an arbitrage: your competitors' egos prevent them from doing the optimal thing.

"In the Peter Thiel sense, it's almost a moral arbitrage because there's something in our gut as a product maker. You became a founder, an entrepreneur because you wanted to go be an innovator. And so it can feel like a beatdown that your path to innovation starts with copying other people's work... But that also makes that opportunity in some ways more available for people who have less ego involved." 00:15:27

The key reframe: define your ambition through the eyes of your consumer, not your peers.

"If you're truly ambitious, burn your resume. And if you define your ambition in the eyes of your consumer, not your peers, you're not trying to win awards and respect from your peers. You're trying to win the hearts and minds of nurses in Indiana, like for Farmville." 00:00:00

Radical Humility as the Path to Radical Ambition

The paradox Pincus returns to repeatedly: the more ambitious your ultimate goal, the smaller and more embarrassingly humble your starting point should be. Over-ambition at the outset is one of the most common failure modes for experienced founders.

"The paradox is the more ambitious you are, the more humble you should be and the smaller place you should be willing to start." 00:28:14

He illustrates with his own experience: after early success with Freeloader and Support.com, he over-reached with Tribe. Then, humbled by failure, he launched something that seemed beneath him — a poker game on Facebook — and it became Zynga.

"By the time I got to Zynga, I did something that was embarrassingly small. I mean, to be 41, multi-time successful founder that could go do something important in the world. And what did I do? I made a Facebook app, like a poker game. Not even a Facebook app, I mean a poker game. It was people thought I had no dignity." 00:26:47

Kill Hope Before Hope Kills You: MVP vs. Maximum Launchable Product

Pincus distinguishes sharply between "hope" and "belief" — and argues that most founders and teams are running on hope without realizing it. The word "viable" in MVP is a trap.

"There's a difference between belief and hope. Hope is confidence without basis... I think that too many founders and teams keep going with the hope that this next release is going to do something magical." 00:33:37

AI makes this worse because it dramatically lowers the cost to reach "viable," giving founders a dangerous drug that enables them to build the wrong thing faster.

"AI is a dangerous tool today. It's so powerful that it means we can get to a viable product in much less time and money than it took — maybe three months instead of three years. And that's a dangerous drug because it means that what I thought would happen where we'd be today is that people would be using AI to build these incredible testing machines, failure machines that are testing more ideas in a week than your industry tests in a year." 00:35:17

The A vs. B+ Test: True Product-Market Fit Has No Ambiguity

When you have real signal, you don't need to ask. The very act of questioning is the answer.

"If you're asking whether or not your product is an A, it's not an A. And you're full of hope, you hope it's an A... When you have lightning in a bottle, when you have true signal, everything works. It's anecdotally, you love your product. You're addicted to it. You show it to friends and they love it. Your metrics show that it works. Did we have to ask if GPT was it?" 00:00:17

D365 Retention as the North Star Metric Nobody Tracks

Zynga was likely the only consumer company in the world tracking Day 365 retention, and Pincus believes it is the single best predictor of whether a company will become valuable. No one else does it today.

"I believe that the most valuable companies in the world, well, statistically, this is true. They have the highest day 365 retention. Can you build against day 365 retention? Yes, we can't wait till day 365. We can find early indicators of a positive day 365 and a negative day 365 retention." 00:44:16

He also surfaces an important nuance: high D30 with near-zero D365 is the norm, and the dangerous false positive.

"You can have very high day 30 and a zero day 365, which is most products. And if you don't think about why someone would use this in a year, if you don't have that mentality, it turns out when someone tries our product, they think that way too." 00:44:43

The Cocktail Party: Social Has Massive Latent Demand, Zero Good Supply

Social media has lost its adrenaline. People are proud to have quit Instagram. But the demand for genuine social experience hasn't gone away — it's just unmet. This is the largest unexplored opportunity on the internet.

"I believe it's lost the adrenaline. I think a lot about adrenaline. Like, is there a heat in it? Like, are you excited to go get on Instagram? Or do you feel a little bit like you're eating potato chips?... People are proud to tell you they're not on Instagram, right? They're not missing the party. They're like, whoa, I got off it." 00:51:55

"Today, we're all hanging out on our cloud, on our GPT, but there's no cocktail party. So my Easter egg to people is, it's a quiet, lonely cocktail party like the web was before social networking." 00:00:55

Micromanagement and Staying Close to the Metal

Against conventional management wisdom, Pincus argues that micromanagement is a feature, not a bug — especially for founders who are the best product makers in the room. The job is to be in the room as long as possible, and only delegate when physically impossible.

"Micromanagement is beautiful. You should micromanage as long as you can... The first principle is be in the room if you can be, as much as you can be. And then only delegate when you get to the point that you can't possibly be in all these rooms at the same time." 01:21:34 / 01:22:42

The #1 Job of a CEO Is to Be Right

Execution, inspiration, and people management all matter — but being right about what to build and where to play matters more than any of them.

"If I get to pick one thing that you do as CEO, I'm going to pick that you're right. So even if you don't operate the ship or the factory that well, I'd rather that you picked the right product, the right strategy... Because being in the right body of water matters more than the right boat. And a great boat in a dead lake bed isn't going anywhere." 01:25:13 (from the 01:25:13 block)

2. Contrarian Perspectives

Being Less Ambitious Produces More Ambitious Outcomes

The conventional startup narrative glorifies big visions. Pincus argues the opposite: excessive ambition at the start is what kills most startups, and the biggest companies often started from embarrassingly small premises.

"The more ambitious you are, the more humble you should be and the smaller place you should be willing to start. So many of these massive hit products or franchises started with such humble, non-ambitious places. Facebook was just an app to check out girls and guys at Harvard." 00:28:14 / 00:24:40

His own Zynga example is the clearest evidence: at 41, with multiple successful exits, he launched a poker game on Facebook. His peers thought he had lost his mind. The result was one of the fastest-growing consumer companies in history.

AI Is Not Yet a Platform — and Treating It Like One Is a Strategic Error

The dominant narrative is that AI is the next great computing platform. Pincus says it is not — yet — and that builders who treat it as one are making a category error.

"Is AI a new platform? I would argue that it is not yet a new platform. It is an important technology... A platform in the traditional sense — it was a hardware platform. It's definitely not a hardware platform at the consumer side yet. And it became at least an interface platform, whether it was a Windows interface or a browser interface or a social network that opened up. We're not at that point yet." [00:01:02:00] / [00:01:02:24]

The implication: consumer founders cannot rely on AI-native distribution the way early Facebook or mobile app developers could. The app install rate is already near zero per user per month, and the new platform hasn't arrived yet.

Viral-Based Companies Are Sinking Speedboats

The entire virality-as-strategy approach is flawed at the structural level. Pincus watched this play out repeatedly at Zynga and in the broader market — the growth channel masks a leaking retention vessel.

"Viral-based companies, and there's been many — BeReal, right, we've seen these — they're sinking speedboats. They're trying to drive faster than they're sinking, right? They're trying to get more people in. And you can either go faster, bail water out faster... or plug the hole in the boat. Or build a better boat that doesn't have a hole." 00:45:13

Zynga's differentiation was not virality — it was retention. They tracked Day 365 when no one else was even tracking Day 30 with rigor.

The "Better" in Your Proven-Better-New Framework Is Almost Always Too Ambitious

Founders systematically overestimate what "better" means. What feels like an improvement to the builder is actually just "new" to the user — and users resist change even when it's genuinely better.

"Better is something that 10 out of 10 of the existing users of that product would say, 'fuck yeah.' 10 out of 10, not you. What you think is better is called new. What we think is better is new. We usually are too ambitious on the new." 00:08:41

Craig Newmark spent two years adding photos to Craigslist listings — not because it was technically hard, but because he was obsessing over whether users would experience it as better rather than just different.

Consumer and Social Are the Hardest Categories to Fund Right Now — and Therefore the Best Opportunity

The market has all but given up on consumer social as investable. Pincus says this is precisely when the opportunity is largest.

"It's not an ideal moment to be — you can't launch consumer right now with high confidence. And that's just something that we have to acknowledge. And it makes consumer and consumer categories like social and games almost not investable." [00:01:04:16]

But: the latent demand for genuine social experience is enormous, the existing supply is deeply unsatisfying, and the infrastructure (AI agents, free tokens) is converging in ways that could unlock a new form of social productivity.

3. Companies Identified

Zynga

Social gaming company founded by Mark Pincus. Mentioned throughout as the primary case study — achieved 8 massive hits out of 10 large game launches, tracked Day 365 retention, pioneered the ASN (Active Social Network) metric, and grew to games with 14–30 million DAUs. Differentiated through retention focus, not virality, and the core mission of "connecting the world through games."

"We had a very high hit rate. It wasn't that we were good at virality. It was that we were focused on two things that I think we did better than anybody else... First, we had this core mission and focus around connecting the world through games." 00:41:31 / 00:42:02

Words with Friends

Zynga title used as the canonical example of Proven Better New done right — a near-perfect copy of Scrabble, with mobile polish as the "better" and Facebook social graph integration as the "new."

"In Words with Friends, it was Scrabble. But if it was just Scrabble, why was it a massive hit with 14 million DAUs and Scrabble itself wasn't, right? There must have been something better and novel about it... It was attached to the Facebook graph and your friends are just already there to play with you, which was a new idea." 00:09:10

Farmville

Zynga's largest hit by installs and engagement. Cited as evidence that the "better" metric is defined by the consumer, not the builder — middle-aged women loved it because of co-op features, gifting mechanics, and social expression, not because it was graphically impressive.

"The reason why people, so many, especially middle-aged women, love Farmville and these games was because they had this kind of hobby, but they weren't doing it alone. And they were doing it with their good friends, making new friends." 00:42:56

Bolt.new

AI-powered browser-based development environment. Mentioned as an exemplary case of unambitious origins leading to breakthrough product — toiled in obscurity, nearly ran out of money, then had the insight to combine their virtual web machine with an AI coding co-pilot to create something differentiated.

"They toiled in obscurity on something that they were into. And then they open sourced it. I think they were barely able to keep going with commercial development. And then they had this realization one day of, wait a second, if we take these web stacks, this virtual machine that we've been making work on the web, but we actually add that to an AI coding co-pilot, we now have something that's better than what anybody else has." 00:29:05

Slack

Cited as a near-perfect execution of Proven Better New — possibly just "proven and better" with no new at all, which Pincus says is even harder to achieve and more durable. Also cited as an example of a founder (Stewart Butterfield) being humbled from a failed game company into building something unexpectedly bigger.

"Slack was a great example because I think Slack might have just been proven and better and no new. And that's even better. If you can have a successful product... people don't like change. So if you take a behavior they like, but you make it much more accessible or something much more fun about it, like in the case of Slack, people love that even more." 00:14:34

Craigslist

Cited as a world-class example of product discipline — Craig Newmark spent two years adding photos to listings because he was obsessing over whether users would experience it as better vs. disruptive.

"I think that was just a world-class product maker because what he gets is that there's a sense of ownership of products that we rely on every day. And we're angry when they change, even if they change for the better." 00:29:50

OMG Pop (Draw Something)

Cited as a textbook example of Proven Better New executed under desperation — took the turn-based system from Words with Friends and applied it to something novel, became the #1 app in the App Store for 60 consecutive days in 2012.

"OMG Pop created the hit game Draw Something and Zynga bought them. They were the number one game and app in the App Store for 60 days in 2012. But OMG Pop was on their last dollar... And now they did Proven Better New perfectly." 00:22:53

Rovio (Angry Birds)

Counter-example to Proven Better New — cited as the "wildcat drilling" approach. Made 45 completely different games with no apparent learning from prior failures. Hit #45 was Angry Birds — innovative and successful, but representing terrible expected-value odds.

"That is kind of Rovio getting to Angry Birds. They made 45 games. Every single game was totally different, not as far as I could tell learning from the previous failures or the market. And their 45th shot on goal was Angry Birds. And it was a hit. And it was innovative. But the odds — that's wildcat drilling." 00:22:00

Freeloader

Mark Pincus's first company. Cited as an example of finding a buried feature (offline browsing in Netscape/IE) and building an entire product around it — the same instinct that Nikita Bier would later apply with TBH.

"What I found was in the Netscape browser and in the Internet Explorer browser, they had buried in there a feature of offline browsing. And my entire product was offline browsing. But they had it as this buried power user feature. I saw the genius in that for that moment." 00:20:52

Tribe

Pincus's social networking company, which had Reddit-like "tribes" functionality and other working use cases inside it, but failed because he was too ambitious and didn't isolate the working elements.

"There was a point in Tribe where I knew my metrics weren't working. Our D30 retention was terrible. We were a sinking speedboat... If I had had this attitude of I'm going to try a lot more ideas, I would have massively changed my odds of success." 00:11:17

Niantic / Pokémon Go

Briefly mentioned as the company where Pincus's daughters had internships, building a first-time user experience.

"They had internships at Niantic working on Pokémon Go. They built a first-time user experience." 01:32:54 (approx. 00:32:54)

WorkOS

Sponsor. Enterprise auth/compliance platform powering OpenAI, Anthropic, Cursor, Vercel, Replit, Sierra, Clay, and hundreds of others.

"Literally every startup that I'm an investor in that starts to expand upmarket ends up working with WorkOS... It's essentially Stripe for enterprise features." 00:06:44

Vanta

Sponsor. Trust and compliance automation platform used by Cursor, Ramp, Duolingo, Snowflake, and Atlassian. Automates SOC 2, ISO 27001, HIPAA, and 35+ frameworks.

OpenAI / ChatGPT

Referenced as the clearest modern example of undeniable product-market fit — nobody had to ask if it was an A. Also discussed as a potential future distribution platform for consumer AI apps and agents.

"Did we have to ask if GPT was it? Did we say, oh, is GPT it? I'm not sure. You know, we're like, no, I'm fucking living on that." 00:58:45

Discord

Founders cited for correctly diagnosing and fixing a structural product management flaw — they were delegating the most critical UX decisions to the least experienced people, and inverted that.

"They came to this conclusion too that they needed to actually be an inverted pyramid. And they said, we realized that we outsource, we delegate the most important product decisions, the UX, to our least experienced people... We decided to turn that upside down and say, we as founders need to be kind of the first and last mile for the product." [00:01:19:57] / [00:01:20:24]

Amazon

Referenced multiple times — Bezos's practice of working two days a week deeply with key teams, and the Andy Jassy "tech assistant" pipeline where every C-suite member had previously been Bezos's technical assistant.

"Andy Jassy at Amazon, everyone on the C-staff used to be someone who'd been the tech assistant to Bezos. So that idea actually scales and you kind of get it for free." [00:01:24:40]

Microsoft

Referenced as cautionary example of a brutal platform that absorbed everything around it — used as a lens for thinking about whether OpenAI will eat the entire market.

"Microsoft was a brutal monopoly. They were a brutal platform. They ate everything around them." [00:01:15:06]

Facebook

Referenced throughout — as the platform Zynga built on, as the early example of starting with an embarrassingly small idea (checking out people at Harvard), and as an NPS case study where users who quit went from +35 to -35.

"Facebook was just an app to check out girls and guys at Harvard... We saw that when people quit Facebook and quit Instagram, they went from a plus 35 to a negative 35." 00:25:05 / 00:52:44

Instagram / TikTok / Snapchat

Referenced as examples of social platforms that have lost adrenaline — Instagram developed "TikTok envy" and abandoned social productivity for passive content consumption.

"Instagram had TikTok envy. So they started doing that. And so I think there is a new step function of productivity that we could be getting in that experience." 00:53:40

4. People Identified

Founder of Zynga. The guest. Built over a dozen successful consumer products, created 8 massive hits out of 10 large game launches at Zynga. Pioneered D365 retention tracking and the ASN metric. Author of Life at the Speed of Play.

"I'm trying to share my playbook and philosophies. And I'm hopeful that somebody will steal from my ideas and take it further." 00:37:17

Nikita Bier

Serial consumer app founder, created TBH (acquired by Facebook). Cited as a perfect example of finding a hidden "proven" feature buried inside an existing product and building a new product around it.

"Nikita saw his perfect product buried inside an Arabic-only version of his product. And he said, oh my God, they nailed it." 00:21:12

Craig Newmark

Founder of Craigslist. Cited as a world-class product maker for his obsessive, two-year process of adding photos to listings — understanding that users feel ownership over products and resist change even when it's better.

"I love referencing someone like Craig Newmark of Craigslist, who took two years to add photos to Craigslist listings... I think that was just a world-class product maker." 00:17:35 / 00:18:31

Bing Gordon

Partner at Kleiner Perkins, longtime games and consumer product expert, early Zynga investor/board member. Co-developed the "bold beat" philosophy with Pincus and said that internet treasures will one day be in the Smithsonian.

"Bing Gordon and I kept coming back to this idea that I love... how can we add more dimensionality? I love this idea of what I call a bold beat." 00:42:02 "My friend Bing Gordon says, one day those treasures will be in the Smithsonian. And I think he's right." 01:36:11 (approx. 00:36:39)

Stewart Butterfield

Co-founder of Slack and Glitch/Flickr. Cited as a repeating example of being humbled by an ambitious game company failure and then discovering something transformative from a small internal tool.

"Stewart keeps trying to start game companies and those turn into unexpected bigger hit companies... he started a big idea for like the mass market MMO, which I've also loved, but really difficult. And they were humbled in that enough to say, okay, wait, there's this little teeny thing that our engineers use. Let's build a product around that." 00:30:01 / 00:30:53

Brian Chesky

Co-founder and CEO of Airbnb. Cited for his practice of building things in non-scalable ways by hand, and for the quality of his product conviction — he already knows he has a hit before launching, he doesn't launch to find out.

"If you talk to Brian Chesky about his launches, he already knows that he has a hit. He's not launching it to find out if people like it." 00:34:20 "Brian Chesky believes in doing things in non-scalable ways and doing it himself with a team by hand." [00:01:20:49]

Steve Jobs

Referenced in two contexts: his obsession with the MIT touch-screen demo at TED (the single "new" insight that became the iPhone), and his insistence on picking carpeting in conference rooms as an example of staying close to the metal.

"Steve Jobs was obsessed with it... okay, there's his new idea is a touch screen. It's the only new idea." 00:12:55 / 00:13:23 "I was so impressed when I heard that he insisted on picking out the carpeting in the conference rooms." [00:01:19:57]

Sid Meier

Legendary game designer ("godfather of game design"). Used as a cautionary example — even the most revered designer in the world failed because he didn't nail the proven first-time user experience on the Facebook platform.

"Even Sid Meiers tripped over what were understood by the most junior product managers at Zynga was the best of breed approach to onboarding a new user... Because he didn't perfectly copy that, he didn't do the proven right, his innovation never got seen by anybody." 00:05:55

Sam Altman

Co-founder of OpenAI. Quoted in the book's endorsement: "Today the only bottleneck to building great products is knowing what to create."

"Sam Altman, the co-founder of OpenAI, said that today the only bottleneck to building great products is knowing what to create." 00:01:49

Reid Hoffman

Co-founder of LinkedIn. Credited by Pincus for the definition of the metaverse he finds most useful: blurring the lines between the virtual and the real.

"I give Reid Hoffman credit for this definition as blurring the lines between the virtual and the real. And that is where our lives are going anyway." [00:01:00:10] (approx. [00:01:00:10])

Andy Jassy

CEO of Amazon. Cited as the exemplar of the tech assistant pipeline — every member of Amazon's C-suite was previously Bezos's technical assistant.

"Andy Jassy at Amazon, everyone on the C-staff used to be someone who'd been the tech assistant to Bezos. So that idea actually scales and you kind of get it for free." [00:01:24:40]

Gary Tan

President and CEO of Y Combinator. Cited for his insight about "intelligence on tap" — building consumer services premised on tokens becoming essentially free.

"There is a great idea that I got to that Gary Tan had that I'll repeat here because I loved it so much this week, that if we get in the mindset of basically intelligence on tap, free tokens." [00:01:06:49]

Brian Armstrong

CEO of Coinbase. Referenced in the context of the trend toward removing middle management — everyone as an individual contributor.

"Brian Armstrong was saying everyone should be an individual contributor. Everyone should be managing a lot of other people. That's really kind of, to me, the best CEO is like the best player at the position." [00:01:17:44]

Jeff Bezos

Referenced for his practice of spending two days a week working deeply with teams on what matters most, and as the origin of the tech assistant pipeline.

"There's so many stories about Jeff Bezos and Zuck of the thing that mattered most. They would go and work two days a week deeply with the team on it." [00:01:21:15]

Elon Musk

Referenced as the rare exception to the "start humble" rule — someone who doesn't start in humble places, but compensates by having access to essentially unlimited capital. The exception that proves the rule for mortals.

"I know Elon is the exception to the rule. Part of it is I think he can raise magical, unlimited amounts of capital, which helps. But then he does not start in humble places at all." 00:25:26

Eric (founder of Bolt.new)

Referenced as a prior podcast guest and someone Pincus cold-emailed because he was so impressed with what they built.

"I loved it so much I cold mailed the founder because I was just so impressed with it." 00:29:05 "Eric was on the podcast." 00:29:59

5. Operating Insights

The ASN Metric: Measure Round-Trips, Not Just Engagement

The Active Social Network metric Pincus built at Zynga is a concrete, actionable retention predictor that nobody else has adopted. It measures bidirectional interactions — you took a turn and they took one back, you gifted and they gifted back. The numbers are specific and striking.

"We built a metric called ASN. And it stood for what we measured, what was your active social network? So we looked at how many round trips you had with a friend or another player... We found that if you went from zero ASN to one, there was an 80% chance we saw you again in the next month. And if we got you to a four, there was an 80% chance we'd see you 22 out of the next 30 days." 00:46:12

Any consumer social, gaming, or communications product should consider building this metric immediately. It is platform-agnostic and the principle applies anywhere reciprocal social behavior exists.

Turn Marketing Spend Into Product Signal — and Then Revenue

The Farmville expansion pack case study is a concrete operating playbook: before spending $10M on external advertising, test on your existing engaged base. Use the results to simultaneously optimize your product variant, validate your marketing message, and generate pre-launch revenue.

"What if you just put something on the game board? And you just see how many clicks you get... When people click on it, it says, 'coming soon, click here to be a pre-user.' It turned out everybody clicked on it... We started selling keys to people for you and a friend. And we ended up selling $19 million worth of keys to get early access to the new expansion pack." 00:37:52 / 00:39:11

The tactic: lock a teaser of the unreleased feature on the existing product UI, A/B test creative, capture intent, add scarcity, and convert that intent into pre-sales. Do this before any external paid spend.

The Teaching Hospital Model for Scaling Founder DNA

As a company scales, the founder's product instincts become increasingly diluted. The teaching hospital approach — putting as many people as possible in the room during high-signal product decisions, plus rotating a dedicated "tech assistant" who shadows the founder — is the most efficient mechanism for transferring judgment, not just process.

"The first thing is put them in the room with you... And then grab someone from the ranks and make them your tech assistant. Have them just follow you around for six months or 12 months and make them the mini me... And then hopefully you've created a mini me that you can now put in at a much bigger role somewhere else. And in fact, Andy Jassy at Amazon, everyone on the C-staff used to be someone who'd been the tech assistant to Bezos." [00:01:23:42] / [00:01:24:40]

6. Overlooked Insights

The "Better" Framing Hides a Design Research Method

Pincus's requirement that "better" must earn a "fuck yeah" from 10 out of 10 existing users — not the builder — is actually a precise, underappreciated design research protocol. He is describing a mandatory user validation gate before any feature qualifies as "better" rather than "new." The implication: most product teams are shipping things they've classified as improvements that are actually untested innovations, which means they're taking on far more risk than they believe while thinking they're being conservative. Almost no one runs this test formally. The corollary is devastating: if you can't find 10 existing users who will unanimously endorse a proposed change as strictly better, you must reclassify it as "new" — with all the failure probability that implies.

"Better is something that 10 out of 10 of the existing users of that product would say, 'fuck yeah.' 10 out of 10, not you. What you think is better is called new... We usually are too ambitious on the new. And so better could be it's now free. There's no download. There's something that you can both statistically show works, you know, is better." 00:08:41

The Free-Token Consumer Business Model Window Is Open Right Now and Will Close

Pincus briefly surfaced a specific, time-bounded investment and founding opportunity that neither he nor Lenny spent any time on — and it deserves serious attention. The insight: AI token costs are dropping toward zero while consumer demand for AI-powered services is enormous, but today's freemium apps get killed by token costs and rate limits before they can reach scale. Founders who build consumer services now — premised on tokens being essentially free within 12–24 months — are building on the right cost curve, the same way that the founders who bet on bandwidth or storage costs falling were right even when it looked like a bad business plan in the moment.

"If today we start building, reimagining consumer services based on free tokens, I think that's such an interesting innovation zone... The price of tokens will come down. And so I think that's another kind of Easter egg of an insight that I'm intrigued by... Getting around that today might unlock really interesting innovation." [00:01:07:08] / [00:01:08:06] / [00:01:08:30]

The analogy to Pets.com is instructive precisely because it clarifies why this is different: pet food costs didn't fall, but token costs will. Founders willing to run at a loss on tokens today — acquiring the right users with the right product — may be positioned identically to early AWS-native SaaS companies that looked expensive to run until cloud costs made their margins outstanding.