Sam Lessin on Startups & Entrepreneurship

Essays

From the Podcast

No, no, it's not the wealth tax threshold. That wouldn't work. But there's a thing which is, like, what Parker and, like, some people are really scared of, where, like, based on your voting control, you pay the wealth tax, which will bankrupt, like, the Google co-founders. I wonder if they f- came up with an interesting legal strategy to, like, maintain control while dealing with that problem.

— Sam Lessin, More or Less Ep 172

It's an everything app. To me, I'm kind of on the fence about it. Like, I, I really think that Jess is right. Historically, in a lot of categories, it's way easier to go de novo, especially when you have, like, humans and, like, all these p- like... But in certain things, like, there are just, like, dating... I actually backed this very small dating, agenticy dating startup, like, a year ago with a tiny angel check just to... It was a cool idea, and the... Getting traction's so difficult, right? And the reality is that, like-

— Sam Lessin, More or Less Ep 171

I think what's inevitable is the death of the web. I think what's inevitable is like a bunch of like the simple shim like types. Yes, you will obviously use these things, right? And for a bunch of stuff, and they'll be good at a bunch of stuff. But it's, it-- The problem is gonna be there's gonna be like a holy war of access to services and a lot to figure out about mistakes and what you do and don't trust them with and how you operate, that I think is just gonna be really messy for the coming years. I don't-- I think it's very hard for startups to survive what's coming.

— Sam Lessin, More or Less Ep 171

I got to 'cause it's like this is straight up the middle. Look, 10 years ago, Andrew Kortina, who co-founded Venmo, and I left what we were doing and started a company called Fin. It was... We called it AAI at the time, or artificial artificial intelligence. And we basically said, "In the future, there are going to be assistants that do exactly what they're all trying to do right now, but we're not there yet. So what we're gonna do is figure out the product experience, build the entire thing." They're still not, but this... So I'll get to this. Build the entire experience, figure it all out, and then we'll slot humans in to just do the minimum that we can and chip away at it in the transaction. So requests would come in, we'd understand what they already categorize them. We had a credentials vault for your credit card. We had a graph of everything we knew as a system as well as about you personally. You'd make a request, "I need a dinner reservation with my friends here," or, "Every week, make sure I send Jessica flowers," or whatever you want. We'd categorize, we'd understand, we'd have recipes for completing them. We built the entire thing and then multi-threaded, voice-activated. It was great, and people loved it. In fact, hilariously, you talk about booking the wrong restaurant recommendation, Marc Andreessen was a user at one point and got livid because we screwed him, Fin did, the same exact way, booking effectively a work lunch at the wrong place for him and telling the wrong thing. It's the exact same story, right? And so here's what we learned doing this. I've talked to several entrepreneurs over the years about exactly the moment we're in now and the problems. One is the biggest problem with running these assistant services is that you have to do everything and to make the service valuable. You can't just do scheduling. You have to do all the tasks, and you become liable for every service provider down the chain from you, right? That you say. So if you say, "Hey, there's gonna be a car there for you, and I'm gonna book it for you," it is now your responsibility that the car is there, even if something else screws up down the line. Problem one. Problem two, which is associated, if you wanna run these efficiently and cheaply at low margin, the problem is, is that some of the mistakes become massively expensive. Like if you book someone the wrong plane ticket or send them to the wrong restaurant, that's like not a small deal. Normally, what you do is say, "Well, I'm very sorry. I'll pay for it." Like DoorDash says, "Hey, I screwed up your food order. No worries. I'll just pay for it." The problem is, and DoorDash can do that because they take a ton of margin and food isn't that expensive. If you're booking someone's thousand-dollar plane ticket or doing something really important and screw it up, you don't have the money, you don't have the margin to pay for the mistakes. And so there's a fundamental problem, like with running this service. And then unlike a human assistant, which is what we're all trying to proxy, everyone has a hairline trigger. Like if you screw me once, it's so easy to fire a digital assistant effectively. There's no emotional bond. I didn't hire the person. It didn't take me a bunch of time. Like if your human assistant screws up, which they do all the time, you don't instantly fire them. Like you're like, "I have to work with you." There's some stickiness. There's like this whole human... But people, if you turn into a service and say, "We have a service for doing this stuff," the tolerance for mistakes goes to zero. Now, with Fin, people would constantly cancel and then be like, "Oh man, I really miss that," and come back a month later, right? Once there was a mistake. But the problem is you're always gonna run an error rate, whether it's one percent or point one percent, you're gonna have that. And that's before you get into the adversarial service abstraction component. And I don't know how these services are gonna be able to do it, because that, that is the real reason. People always ask me repeatedly over the last few years, "Why don't you restart Fin? Why don't you restart Fin? You could do it now. It'll work." And I'm like, "You guys don't understand why we stopped doing Fin." It was not that we couldn't provide a great service and not that the economics didn't work. Both of those were fine. The problem was what does it mean to take on that level of responsibility as a service, and can you make the economics of mistakes work?

— Sam Lessin, More or Less Ep 171

I think it's very hard for startups to survive what's coming.

— Sam Lessin, More or Less Ep 171

I'm willing. For, for all those founders out there listening, I will sue you over my pro rata rights, just to be clear.

— Sam Lessin, More or Less Ep 170

They're very important, and whenever we do deals, I'm always like, to the founders, I'm like, "Listen, I wanna be really clear now," and I'd love... Usually I would, like, ask for written documentation, but I'm always put, I'm like, "We care about these. This is not like, 'Ha ha, you have pro rata rights.' Some people, later stage funds, are gonna tell you later that your seed pro rata doesn't matter, and, like, we'll just, like, quash it and, like, not respect it because they don't need that. They already own plenty, and, like, make the cap table work by, like, taking out the seed." And I'm always like, "Listen, that's gonna happen later because it always does." It happened early in Slow's history with a deal that I did I n- I'll never forget, when a well-known large VC that we thought we were friends with, and literally is an LP of ours-

— Sam Lessin, More or Less Ep 170

I'm actually really confused why you guys... Uh, what you're even talking about. But, like, this is too important to just abandon and turn over the keys to, like, innovative entrepreneurs. I actually believe that.

— Sam Lessin, More or Less Ep 168

Uh, well, yeah, well, no, we're not, but here's my problem. I'm all for, like, higher ed reform and making different types of education more accessible and creating more pathways, and, like, amen. I am woefully uncomfortable with the technology entrepreneurs being in charge for all of this, and Alpha School's a good example. I am seeing more and more people talk about moving their kids to Alpha School.

— Sam Lessin, More or Less Ep 168

Why, hello, friends. Welcome to More or Less, and a very big welcome to our special guest today. We have Matt Mazzeo, folks, joining the pod. Now, let me tell you, right now, Matt is with us as the founder of NoHo, the hot up-and-coming king of peptides, and so we're gonna talk about that.

— Sam Lessin, More or Less Ep 168

Matt is with us as the founder of NoHo, the hot up-and-coming king of peptides.

— Sam Lessin, More or Less Ep 168

It's an interesting company. So we funded a long time ago. We were the first investment called Sublime Security. When we were talking to the founders of this, the whole point with Sublime early on was like, look, we all know the top 10 tech companies in the world. What they do is they get on a weekly call between the CISOs, right? And they're like, "This is what we're seeing," blah, blah, blah. They're, like, sharing things.

— Sam Lessin, More or Less Ep 167

But that's a great sourcing pipeline. But it's actually not cra- Like, especially kitesurfing in that moment was a pretty smart way to do it. 'Cause you think about it, it's actually a great sport for entrepreneurs. It's a little risky. It's, there's a high barrier to entry. It selects for the right type of people.

— Sam Lessin, More or Less Ep 165

Look therememelordwhichitisinterestingisacompanywebackedithalhilariousirreverentmemegeneratorthatsrelevant toit Heres IllgiveyouoneInterestingDatapointIveSeenThisWeekAbunchWhichActuallyReallyResonatedWithMe Ivemetwithtwostartupsnow ... that no one has heard of in Silicon Valley at all, right? It's not even not generous. They're kind of just LLM wrappers for middle America, right? They're like y- they're things that... Like, they're not sexy. Like, no one here is gonna care, but they're crushing it on growth and revenue with affiliate programs, deploying in small businesses with all this stuff. And you know why? It's a really interesting insight

— Sam Lessin, More or Less Ep 165

it's funny From a first principles perspective it seems obvious that Canva would have to evolve at this moment because you're making a paperless post or making an ad and it's way better to just have an Ai MAKE IT fOR YOu AND RUN IT foR YOU THAN TO move Things arouND onA screen THiS ISThe Same PRoblem WITH Figma I had A Funny meeting yesterday With Some very Very young Founders AND I Was LikE wherE's yOur figMA SOMETHIng LIke we don't--we jusT Use Paper IM likE finEBuTthe poinT IS fiGMaAndICanVaAre thesameThingTheyhavetoevolve Ontheflipside It'skindawild OurkidsreallyusecanvA

— Sam Lessin, More or Less Ep 165

Tweets