Alex Rampell：最优秀的创始人能够凭空调动资本、客户与劳动力｜风投的未来

1. How to Do 5x on a $15BN Fund Pool?

Alex, dude, it’s been 8 years. I’m hoping that my question-asking ability has gone up in terms of quality in those 8 years. Now listen, I want to start with the $15 billion you raised today.

I was just looking at that, and I was wondering: in an age of venture today, do you have to go really big, or go very small and boutique, to win in venture today?

Yeah. I think this sounds like a bad word when I say “death,” but there is this kind of death of the middle that happens to a lot of asset classes in general. Venture capital was a tiny, tiny asset class at the beginning. Right now, it’s gotten bigger, but really, more of the end state of a lot of these companies is huge. I mean, Sequoia used to brag that, I think, 20% of the market cap of the Nasdaq was in Sequoia companies—companies like Apple and Oracle and all of these amazing names. They’re very, very big, right?

Companies go public much, much later today. So the ability to deploy more capital—more money—into venture capital has grown dramatically. Series D didn’t exist in 1992; that was an IPO. Amazon went public at something like a $600 million market cap, and that was the norm. There was no Series I, Series K, or Series W. You would just raise a Series A, raise a Series B, raise a Series C, and then go public.

Consequently, venture firms back then were very, very small, but the exits tended to be quite small as well. If a very, very good scenario was that you had a company go public at a sub-billion-dollar market cap, and you got 5 of those a year, you couldn’t raise lots of money. But now the opportunity is so much bigger. The 5 biggest companies on Earth are all technology companies. If you rewind 20 years, I think they were all banks. If you rewind 10 years before that, they were all oil companies. If you rewind 10 years before that, they were all Japanese companies during the Japanese stock market bubble.

The opportunity in technology is so much bigger, especially because you can keep investing venture capital dollars later. I think that’s one of the main things. If you look at the money that we just raised, almost $7 billion of that is for the growth fund.

David George has got a big appetite. Ali’s raising a Series T as we speak. He’s going through the whole alphabet.

But this is the point. If companies went public after the Series B back in the 1990s, and the average IPO was $50 million to $100 million of capital raised, the strategy would be a little bit different. But the world has changed dramatically, and the opportunity size is so much bigger.

Now you have technology companies that pervade everything. If you are a large company today and you don’t use software at your core, you’re going to get eaten by somebody who does use software at their core and then reverse-engineers into whatever product or service you promote.

2. What Two Groups of Funds Will Win the Next Decade in VC?

Every LP says the canonical wisdom and the theory of venture is that, as you scale, performance goes down. Do you legitimately think, then, that with the expansion of these markets, you can maintain 5x-plus net funds at scale?

Well, I think the difference is this: imagine that you’re an LP and you have $1 billion to invest. Would you rather invest $50 million and get a 5x on that, or would you rather invest the whole $1 billion and get a 3x on that? The answer is that you’d rather get a 3x on $1 billion than a 5x on $50 million.

One of my good friends is this guy, likely Micky Malka, at Ribbit. I was lucky to be an investor in his first fund personally. That was like a 55x fund on, I think, an $85 million fund. A 55x—that’s insane. But at some point, you can ask Micky this, too: you’re better off with a 5x on a very, very large fund.

The harder thing to do is just to return gross dollars, period. That’s what LPs actually want. It’s amazing to get a 100x. I’ve had 2 funds that I’ve invested in. One is Micky’s; the other one is this fund called AngelPad, which was kind of like a third-rate competitor. I don’t want to call it third-rate, but there was Y Combinator, and then there was AngelPad. It was just this small little experiment that was 120x. I got 120 times the capital back in DPI.

How big was the fund?

I think it was $8 million.

That’s impressive.

But this is the thing: that’s incredible. Your point is very valid. Can you get $120 on a $2 billion fund? Probably not. I’m willing to bet you that you can’t get 120x on that. But you can return far more dollars if you’re very, very good at this.

The question that you originally asked—and this is why I called it the death of the middle—is that my view is, in most asset classes, you either have to be a large generalist or a small specialist. The hard thing is to be a midsize generalist, because then you’re largely going to lose to either the big generalists or the small specialists.

Ribbit, as an example, really focuses on fintech. That’s how I know them well; that’s a specialty. They’re not trying to do everything. Kaszek, in Latin America, is focused on a specialty. They can be small. They’re not trying to do everything across the entire planet, because the entire job of venture capital is to find, pick, and win investments.

If they’re good investments, the winning is very, very hard. The winning therefore goes to the person that is the best. You have to sell. This is a sales job. You know this, right? You have an entrepreneur, they’re amazing, and they don’t come along very often. This is the best entrepreneur you’ve ever met, and you have to convince them to take your money.

How do you do that? You have to say, “I am the greatest person in the world to help you,” which means I have this amazing specialty and/or I have all these things that I can do for you. I’m connected to everybody on the planet, given the scope and scale of my generalist approach.

On the big side, if I’m just like, “Hey, I kind of do a little bit of everything, and I don’t really know that much about your business, and I’m not that big and can’t help you that much,” you’re going to lose. That’s why the death of the middle is what tends to happen for a lot of these asset classes.

LPs want to chase returns. It’s also sometimes hard to reach LPs, so the big generalists gobble them up, or the small specialists that generate very, very good returns will gobble them up as well.

I have so many things to say. The first thing I do want to say is that Micky Malka, whom you mentioned, helped me when I was 18 and agreed to be a mentor of mine 12 years ago, when it was completely not obvious. I had no idea why he spent time with me, and he’s been incredible to me ever since. He always taught me, “You’re never won or lost. You’re only ahead or behind. Keep playing.”

And I love that. You mentioned the scale of dollars, and actually, wouldn’t you rather do two 5x investments on $250 million than, I don’t know, 15x on $10 million or whatever it is? Yes, but there’s an opportunity cost to dollars. For an endowment fund, they’re able to put it into the smaller fund.

Do you accept, then, that you just scale out of certain LPs, and it’s no longer the best risk-adjusted place to put money?

Well, I think it’s obviously impossible to disprove an unknown future. But I would posit that if you were trying to find, pick, and win the best deals—and maybe you disagree with me—I’m of the view that it’s either the small specialist or the large generalist.

But who wins the best consensus deals? Every now and then, a non-consensus deal will pop up that everybody thinks is terrible. Nobody wants to do it. Sequoia doesn’t want to do it. We don’t want to do it. You don’t want to do it. Nobody wants to do it.

Then it ends up being a 1,000x, and somebody who is not the best-known venture firm ends up winning that deal—or being sold that deal, I should say—and it ends up with a great return. But a lot of the best deals will go to the best firms.

That's what's very different about venture capital than private equity. If you and I are trying to take a public company private—you’re KKR and I'm Blackstone, we're both trying to take over RJR Nabisco or something like that—they're just going to sell to whoever offers them the highest price per share. I mean, they have to. Whereas in venture capital, as you know, you have to win the hearts and minds of the entrepreneur and win that deal.

A lot of the best deals are somewhat obvious. It's not surprising: everybody wanted to invest in Uber, everybody wanted to invest in Facebook. It was self-evident that these were very, very interesting companies. Maybe when the price gets high enough, some doubts come into people's minds, like, "I don't know if I want to invest at $87 million pre for the Series A of Facebook," but everybody wanted to do it at $20 million pre. There are a lot of companies that people don't want to do at any price.

But the reason why I'm saying this is I don't necessarily think you could take it as a given that a small fund will outperform a large fund. Now, I think it has the capability mathematically. Again, if you're Micky and you invest in the Series A of Coinbase and you have a very, very, very small fund, of course you can generate a bigger multiple of that fund. That's just algebraically true, but the best deals in fintech—Micky gets to do them because he's at a great firm. He has a much, much bigger fund right now.

That's the thing that's hard to know. Again, I agree with you algebraically, but I would put my own personal money—and I do, right? I invest in our funds. I would put my own personal money in funds that have the small specialist or the big generalist, because I think that's where the best returns will be.

Can I ask you, when you think about the best returns, what is the multiple of your best return, give or take?

For a single deal, there's a seed deal that I did that's probably marked up at 200x right now.

You said something about consensus deals, and I immediately thought of an Andreessen deal, ElevenLabs, which was the most non-consensus deal ever at seed. You were competing with OpenAI, you're in London, it's a pre-seed—it was very non-consensus. When you look back at your best deals, have they been consensus or non-consensus?

Well, I think—but if you look at ElevenLabs, the entrepreneur was pretty consensus. It's like, all right, likely Mati's super; that whole team is incredibly talented.

Sure, but a lot of people turned down the pre-seed and the seed.

Yeah, but I think our job—and I say "our," right? Your job is to tell me if you agree with me—is to find the smartest people in the world who have very high agency. There's been this thing going around about agency. I love this term. How do you define it? People aren't going to be told what to do; they just take matters into their own hands. This is a very rare trait, right?

You obviously had this trait when you could have just done the normal thing for a 19-year-old to do—or however old you were when you started. You were younger than that: 17?

17. Yeah.

Yeah. It's like, what you did is not normal, right? You had agency and said, "I am going to not do the normal thing. I'm going to email every famous VC to death and get them to talk to me." It's pretty incredible what you've done. That's a very rare trait.

You find people like that who are hopefully experts in their domain, and I think this is why the specialty thing that I mentioned is very, very important. I believe that there is a certain level of consensus around who has agency and who is an expert in the domain. If you talk to an amazing entrepreneur, it's like, "Wow, this person knows everything about this. They've studied it for decades. They've read every book about it. They've talked to every entrepreneur who's tried this before."

You have to give them money. That's our job: to find these people and give them money. It won't always work, for sure, but I actually don't agree that ElevenLabs was a non-consensus deal. If it was a high enough price, if it was not a seed, if it was like, "Okay, it's a Series B, they have $500,000 in revenue and it's shrinking every month," yeah, of course it's not going to be consensus. But I'm an entrepreneur.

That was going to be my question: at what stage does that no longer hold true? The Series A partner who leads our Series A fund is like, "The seed guys have it easy. Amazing founder, great, let's roll the dice." For us, it's not quite enough. There comes that Series A.

No, I agree. At some point, reality converges with reality. My kids and I have been watching Silicon Valley, the show, and there's that famous scene where the Mark Cuban character is on the phone and hears revenues: "No, no, you can't do revenue. You have to be pre-revenue, because then you're a pure play," right?

There is this element. The way to explain this financially is that we buy out-of-the-money call options. You know what a call option is, right? We are buying out-of-the-money call options, and we hope they expire in the money. This is how I explain to people why a Series A that has $1 million in revenue and is losing $10 million a year is worth $100 million.

Of course, it isn't worth $100 million. What you're doing is you're buying 15% or 20% of the company and hoping that eventually your call option expires in the money. That's the thing that you're doing. Once eventually that value converges on the equity value, it's like, "Oh, what's the discounted cash flow?" Once it gets closer to that—and it's not a binary thing, right?—at the seed it's like, okay, out-of-the-money call option. This guy or gal is very, very smart. I want to buy 20% of whatever they're doing and hopefully it expires in the money, and they're the smartest person I've ever met.

We do those deals 100 times a day. We will do them 100% of the time, right? Consensus, non-consensus—there isn't really anything to be consensus or non-consensus on, right? It's just like, this is a very, very smart person. It only becomes non-consensus, to your point, when the price goes up high enough, because I think most people have the same viewpoint of, "This is a very, very high-agency person who has studied history."

3. What Three Things Are the Best Founders Able to Do?

There's a memo that I wrote internally for our firm about how to invest in people, and I think you want to invest in people that can materialize labor, capital, and customers, especially today, where people get paid a fortune to stay at OpenAI, Anthropic, Meta, or any of these companies. If you quit your job to start a company and you can snap your fingers and 5 people follow you tomorrow for a 50% pay cut, that's pretty magical. That doesn't happen every day. So that's materializing labor.

You also want to make sure—this kind of goes into the consensus/non-consensus part—is this person really good at fundraising? Are they telling a good story? Can they convince people like me to give them money? Oh, wow, they really can. That means hopefully the N+1, N+2, and N+3 rounds will be a little bit easier. They will converge on reality in terms of numbers, for sure, but they have the thing around raising money.

And then this is more of an enterprise-focused thing, but can they get their first 5 customers, which is as hard, if not harder, than getting their first 5 employees? Imagine this company, Toast—you know Toast, the restaurant POS company.

Yeah, I love them, dude. I'm a vertical SaaS nut.

4. The Best Companies Have Hostages, Not Customers

Oh, I know. I love vertical SaaS, right? So imagine that you're Chris at Toast. You start this company and you go to a restaurant and say, "Hey, I want you to use my product." The restaurant asks some very good questions, like, "Okay, well, how much cash do you have left?" It's like, "I have a week." "Okay, interesting. How many other customers do you have?" "Zero." That's impossible. How can you pull that off?

If you are this rare breed of person that can materialize labor, capital, and customers, then I have kind of 2 sub-appendages after that. I really, really like people that have studied the history of the space, and I say this because the best entrepreneurs that I've met have learned everything about the space.

To show what a great investor I am, when I was running my company, TrialPay, I met with Patrick Collison. I know a lot about payments. I've been doing payment stuff since 1997 on the internet, which is kind of the early stage of online acceptance of credit cards. I met Patrick and, obviously, I passed on doing the seed round of Stripe because I'm a genius. It was called /dev/payments at the time. I was not in Andreessen Horowitz, so don't hold it against me. It didn't hurt our DPI, and luckily the firm invested in them.

But 2 things really impressed me. Number 1, I asked Patrick, "Where are your customers going to come from?" Because everybody uses Chase Paymentech. It's like, "Oh, my customers don't exist yet." It's the stupidest answer I've ever heard, but obviously it was genius. Number 2, what really did impress me is that he knew everything about the history of the payment systems.

I think he actually went out to meet Dee Hock, the founder of Visa. John Collison gave me a book—one of those Springer yellow academic textbooks—on the origins of the payment system.

They had studied history so much. Same thing for Vlad Tenev at Robinhood. He studied history so much. Same thing for Apoorva Mehta at Instacart, who went out to meet the founders of Webvan. This is a very, very classic trait.

On the other side, I will meet people who will start a company almost exactly like TrialPay or almost exactly like Affirm. I know a lot about these 2 companies because I started them, right? And they’re like, “What was TrialPay?” or, “I had never heard of this.” It’s like, “Come on, man. You’re going to spend 10 years of your life building this thing, and you really should study history.”

Brian Chesky at Airbnb studied everything about the bed-and-breakfast and hotel industry in the 1800s. This is a very, very classic trait. Let me just finish with this: labor, capital, customers—study history.

My favorite book of all time is The Count of Monte Cristo because it’s a story of revenge. The reason why this is so important, if you know the book—it’s by Alexandre Dumas—is that Edmond Dantès is wronged. He’s sent to prison for bogus reasons, supposedly for being a supporter of Napoleon, for 17 or 18 years. Eventually, he gets out and becomes the richest person in the world, but he doesn’t give a shit, if I can use that language. He just does not care; he wants revenge. He wants to destroy his enemies and conquer the world, or just really destroy his enemies.

You need that kind of motivation because, going back to fund size, if somebody offers you $100 million and you’re an 18-year-old kid, that is transformative. You’d have to be an idiot to turn that down, or you have to want revenge or redemption. Revenge and redemption are kind of the same thing.

I find a lot of the best entrepreneurs have that going. They want to prove they’re better than everybody else. They had some childhood chip on their shoulder, or they were wronged at their last company. Dave Duffield had this hostile takeover of PeopleSoft. Of course, he starts Workday and he’s like, “Fuck you, Larry Ellison.” There’s always that kind of energy.

The Count of Monte Cristo thing—I don’t know how to describe it—but the motivation has to be beyond just, “I want to make $50 million.” If that’s the motivation, that’s great motivation. I’m a capitalist; we live in a capitalist society. But if that’s the only motivation, it’s not going to work for our fund size.

I love seeing that fire. A lot of the most successful companies I’ve seen always have that. Renaud Laplanche starts LendingClub, gets fired from his own company, and has made tons of money, but he doesn’t give a shit. He starts a competitor called Upgrade. No accident that the company is called Upgrade—it’s like an upgrade over you motherfuckers, right? He starts an upstart, starts Upgrade. Upgrade has a multiple of the market cap. It’s probably worth 10 times more than LendingClub now. That’s a very, very classic commonality.

My question to you—I want to stay with the questions there because there was so much to unpack. You said how you love them studying history, and you said about passing on Stripe. That was my concern: there is a level where you can know too much.

I think I know quite a bit about lending now—a beginner’s beginner compared to you—but I know quite a bit about lending. It’s quite easy for me today to see a lending business and go, “Fucking horrible. It’s a hard market. I don’t want to be there. Look at LendingClub. Look at the market cap there.” I’m very dismissive, as many were with Stripe when they knew payments. How do you prevent yourself from knowing too much, such that it becomes a negative?

I think this is a great question, and this is the number 1 thing: I do a couple of things. Number 1, if it’s an ad tech company, I know a lot about ad tech. I know a lot about payments. I will force somebody else to join me for the pitch who has a beginner’s mindset. I think that’s 1 thing: have a sparring partner internally who has that “What if it works?” mindset. You always have to be like, “What if it works?”

Number 2, I like to ask the entrepreneurs, “What is different?” The thing that’s different—the reason why Patrick and John made Stripe work, partially—is that they just believed that a great number of new companies would be created, that those companies would pick the best product, and that they were going to have the best product.

This informs a big part of my investment thesis now. I call it greenfield, but there’s a saying that I use a lot, which is, “The best companies have hostages, not customers.” You’ll appreciate this if you’re an enterprise SaaS guy. The best companies have hostages, not customers.

If there’s a company that has something marginally better than Workday, they’re not going to say, “Workday has hostages. They don’t have customers.” They’re not going to be able to sell to GE and say, “Oh, wow, I love you, 2 YC kids. I’m totally switching my HRIS from shitty Workday to amazing AI-whatever YC Silicon Valley HRIS.” It’s never going to happen.

But if the rate of new-company creation is high enough, those new companies will pick the best product. They’re like, “Oh, wow, I could use Workday, but I’m not a hostage, so I’m free. I’m going to pick this other thing.”

I was the first investor in Mercury, the SMB bank. Until SVB failed, I don’t think they ever stole a customer from SVB. But if you’re a brand, as long as the rate of new-company creation is high enough, you can play this game that I call greenfield bingo. You pick every software category, build a better version of that, and then you’ve got a shot. That’s what Stripe was, right? That’s why it worked.

If the rate of company creation is very low—for example, if I build a better EHR, an electronic health records company—it’s just not going to work because the rate of new-hospital creation is too slow. You can’t just sell to the new companies. But you can do that for payment processing. You can do that for ERP. You can do it for a bunch of other categories.

You look for greenfield bingo markets where the rate of net-new companies being created will supplant the slow, slow sales cycles of the larger enterprise customers who will eventually switch.

Or maybe they don’t, right? Who cares if they switch or not? They’ll hopefully die because they’re using shitty software. The fact that they won’t switch is actually indicative of their mantra on everything: they want to use old technology, or they’re hostage to old technology. Let’s just sell into the future.

Betting on the future is more fun. One of the things that’s very, very challenging is that you go start a company, recruit 10 hotshot people from Meta, Google, or whatever, and then they’re bored to death. Why are they bored to death? Because they can’t do anything.

They were used to making little tweaks that 1 billion people experienced every minute, every hour. Now they’re at a startup, and the startup has been going for 1.5 years and they’ve made 1 sale. That’s quite tough. Then you end up losing your talent because it’s boring—you can’t actually do anything.

It’s nice to have these markets that can ramp quite quickly. You want the market to be a tailwind for you. It doesn’t mean there’s not value in creating big companies that sell big software products to big companies. You can do that. It’s just a much, much harder thing culturally for Silicon Valley, I think.

Shows are a bit like venture: the majority that you do are actually not very good. Then you get the once-in-a-while episodes like this, which remind you why you love what you do so much. You know what I mean? When you meet that special founder, it’s so great when you have a show like this.

My question to you is, you said, “Hostages, not customers.” How should I think about that in a world of Cursor, or any of the foundation models—Anthropic or OpenAI—where they are customers, not hostages? They can switch very easily. The promiscuity of customers has never been higher. How should we think about that?

It’s a really good question. I think this is where, behind every technology revolution—and to go back to Silicon, then the personal computer, then the internet, then Web 2.0, where you could write to the internet, things like Facebook and YouTube, then mobile, then cloud—there’s always been an infrastructure layer and an application layer.

You go back: the infrastructure layer for PCs was Microsoft and Apple, the operating-system players. The infrastructure players for the internet were Cisco and Akamai. The infrastructure players for everything AI are all of these back-end model providers. Then there’s the application layer on top.

If we do what we were talking about with AskLio, that’s an application-layer company. If I were Vlad, I would love to be promiscuous with all the back-end models, because I should be. The infrastructure players are like, “Fuck, all of our customers are being promiscuous. Let’s figure out how we specialize in a particular area.” I imagine that’s why Anthropic has gotten very good at coding.

The application layer tends to be a little bit stickier. But the problem is, you might have 9,000 competing companies at the application layer, in which case you’d rather be the infrastructure layer. The infrastructure layer is pretty hotly competed as well right now.

So, I don't know. I think the more relevant question for me is that, in 2025, the ability to go create a software product is so easy. I published this chart with the help of my friend ChatGPT showing how long it took VisiCalc, which was the first spreadsheet that came out in 1979, to lose to Lotus 1-2-3, and then how long it took Lotus 1-2-3 to lose to Microsoft. It took about 5 years for VisiCalc to go from 100% market share—they were at 100% because they were the only one and the first—to 50% market share.

It took about 15 years after that for Lotus, which had 70% market share in 1986 or something, to go to almost zero. This would normally take a long time. In 2025, this can take weeks, which is bonkers, right? All of these layers of past innovation have almost like a Russian nesting doll, kind of concentrically grown against each other.

Because you have cloud and because you have mobile, everybody in the world has a smartphone in their pocket. All of those smartphones are connected to, like, infinite computing in the cloud, or near-infinite computing with, like, a dollar of energy in the cloud. If I build something marginally better, I can get it into the hands of a billion people overnight. That’s just so different.

But I think, on the hostages point, if you build a system of record, right, it’s just so hard to switch. That has not changed. But now I can compete: I could build a software product in 2 weeks that would have taken me 2 years. So that’s going to massively increase the pressure on the application layer.

The best thing that you can do if you’re an application-layer company is hopefully to have something that—I hate to say it—but you want to have hostages. You want to have all of the data in your company. You want to have all of your customer’s data in your product.

This is, I think, the thing that we talked about the last time I was on your show. The battle of every startup versus incumbent is whether the startup gets the distribution before the incumbent gets the innovation, right? So what do you do? You go boring. You build the most boring thing possible. Nobody really cares about it; nobody’s that interested.

This is why I love Vlad and AskLio: who cares about procurement? It seems kind of stupid. It’s not attracting 9,000 competitors. But hopefully you get all of the data in there, and then you can build these interesting things on top. You’re not going to attract that much competition, and even once you do, it’s just kind of hard to switch. So, I don’t know if that answers your question.

It totally answers my question, but it leads to several more questions, which is the theme of this discussion. The speed with which it takes to compete with the incumbent has reduced, and you are able to take customers or market share quicker than ever before.

With the extension of private markets, do we not have a liquidity problem, then? I don’t want to pick on anyone, but I will say it: a company like Snyk in the cybersecurity market is now getting eaten away by new entrants before it’s had the chance to return shareholder money and liquidate. Do we not have a fundamental challenge here where companies that have not gone public yet, or not provided returns to investors, are already getting eaten away because that compression of time is shorter?

Yeah, I think this is a big challenge. If you look at all of the unicorns and how many conform to the Rule of 40, it’s pretty small. Many of them are shrinking. Of the unicorn class, I would bet that maybe 5% will ever be able to go public, which is kind of shocking, right?

Because so much money has gotten into venture capital, you have this problem. I will say on the record: I hate massive secondaries, because it kind of turns you from the Count of Monte Cristo to—whatever the opposite of that would be—“I’m now going to go vacation in the Côte d’Azur,” or something. You don’t want that, because that’s going to create a fundamental disconnect from your employees and your investors. You’re rich and they aren’t. That’s not a good setup. You kind of want everybody to be in the same boat.

The reason why I mention that is that you have some companies where it’s like, “The founder is taking a $50 million secondary.” That’s fine if they just turned down a $10 billion acquisition from Google, they’re the Count of Monte Cristo, and they’re going for it. Okay, that can make sense to me. And if you offer that to all employees and all investors and everything else, that’s fine.

I don’t love the idea of people looking at this as spreadsheets. I won’t mention names, but there was a fund in 2021 that did a massive secondary into one of my companies. I was really against it, which made me super popular with the founder, as you can imagine. They were like, “We own 4% of the company. We want to own 8% of the company because 8% is more than 4%.” I’m like, “Dude, I totally agree with you. 8% is more than 4%. But you have now introduced moral hazard into the equation.”

If you give somebody generational wealth, you can hope that they’re going to maybe swing for the fences and go for it. The upside would be that they say, “Otherwise, I would be happy selling for $1 billion. Now it’s like, ‘Screw it, I’m going to go for $100 billion.’” Okay, that’s great. Now we’re all aligned.

5. The Two Types of Deals You Want To Do In VC

But the other option is that they don’t care about getting liquidity for investors. They don’t care about getting liquidity for employees. They’re quite comfortable. You don’t want to have that setup.

I don’t think that’s actually the problem. I mean this with the greatest of respect. I think we assume the next strategic steps will be the same with that money as they would be without that money. What we’ve both seen is the foie-gras-ing of startups, and then they do 10 things instead of 2 things. None of them work, the team is disincentivized, they break up, and the culture sucks.

Moral hazard. That’s the economic framing, right? It’s moral hazard on both primary and secondary, to your point. Necessity is the mother of invention. So if you have $100 billion in the bank when you really should only have $10 million in the bank, you’re like, “I’ll do 50 things. I’ll have multiple layers of people that I don’t need.”

I find that a lot of people, when I think about the difference between conservatives and liberals, or people who believe in big government and small government, a lot of it comes down to the disconnect between “more input is better output.” A lot of people just believe this.

Take the IRS, the Internal Revenue Service. People say, “There’s a lot of tax fraud. We need to hire more people, and if we have more people, we’re going to do a better job of catching tax fraud.” Or they say, “The military—we should have more people in the military, because that way we’re going to do a better job.”

Whereas, as you know, sometimes there’s addition by subtraction. If I have a smaller team, there’s less communication necessary. You’re going to come up with more creative ways of solving the problem. You’re going to solve it with technology.

If you just say, “I’m going to solve it on the input layer. I’m going to address my constituents by saying, ‘I’m just going to allocate more money to this thing,’” you’re going to get a worse outcome versus allocating less money with great people. This is the key. You can’t just say, “I’m going to allocate less money and give you the worst people on Earth.”

Take tax fraud. I would rather have 2 people at the IRS than 80,000 people, but have those 2 people be likely Noam Shazeer and some other super-genius. If Jeff Dean and likely Noam Shazeer are running the IRS, oh my God, that would be so much more efficient. The input cost would be 1/100th as much, and there’s always that disconnect.

I’m actually in trouble with my team because I just tweeted today: “Series A is the worst place to be investing. Company progression is minimal, price is 4 to 5× the seed price, and we’re paying 15 to 20× ARR with little signs of product-market fit.” Do you agree with me? Is Series A the worst place to be investing?

Well, I think the problem is that there’s the nomenclature, which varies from company to company. When I started TrialPay, our Series A was $3.1 million on a $9.5 million pre-money valuation, and that was expensive. I remember arguing with the partner at Battery, saying, “This is the most expensive deal we’ve done.” This was in 2006.

At SiteAdvisor, I think we raised $2.7 million on a $2.7 million pre-money valuation, so even lower. Hence, he was right. So now you have a pre-seed, a seed, a seed extension, a seed extension 2—what is a Series A? Normally, a Series A would be the first institutional round of money.

Now there’s so much variance. There’s the Series A where it’s 5 superstars out of OpenAI and they need tons of money for compute. There’s no moral hazard on that. They’re not going to spend money on people; they’re going to spend money on GPUs. That’s one form of Series A.

Another form of Series A is that I just did a Series A where the company had about $10 million of ARR when I invested in it. So it’s just all over the place. I think it’s hard to cast a generality.

There are certainly ones where I used to call this the Series B trap, but again, I think the nomenclature has shifted. I would have agreed with your team if you called it the Series B because, at that time, there was a seed and there was a Series A, and the only difference between Series A and Series B was that you increased your burn and built infrastructure and scaffolding.

It was like, “I have a company, I have customers, I have signs of product-market fit. I know now I should hire an HR team and a marketing team and all this other kind of shit that doesn’t actually have any impact on the metrics of the company.” That was the Series B. Then it’s like, why would I invest in a Series B when I get half as much ownership and nothing has changed vis-à-vis the Series A? So yes, there’s a class of Series A rounds that look like that. But I would say that, of the Series A rounds that I personally did in the last year, most of them have been like, “Holy shit, revenue is really scaling and these numbers are insane.” Those were Series A rounds, and I get very excited about them. Your mileage varies because the nomenclature is all over the place.

Do you worry about the quick-succession rounds when you look at companies like Rillet or Tacto? There’s just, like, a week later, another term sheet for a Series B with literally no change at all, and it’s buying the call option. Do you worry about those rounds?

6. The Importance of Founder/Capital Fit

Well, I did one of them, right? I’m on the board of Rillet. I did the Series B, and it was 60 days after the Series A. That’s unfortunate. I would have rather done the Series A, or rather done the seed, of course. But if you find the winner, it’s also very expensive not to do that deal.

That’s really interesting. I’m so pleased because I totally forgot that you did the Rillet B. But you’ve got to pay up for that, going to the point that you’ve got to assume the next strategic steps will be the same and be as focused, even though you have just funded the company. Sorry.

Well, this is where I think the motivation of the founder is very, very important. Going back to Nick, who’s the CEO of Rillet, I think he does have a bit of the Count of Monte Cristo in him. He doesn’t want to take this money and spend it on extravagant things.

I think you have to make sure that there’s founder-capital fit. Nobody ever talks about that. It’s like, “Okay, if I give you $1 billion, what will you do with it?” Ninety-nine times out of 100, the answer is going to be bad news—not even bad news around waste, but just bad news in terms of mindset.

It’s another form of moral hazard where I’m never forced into making hard decisions because I have infinite capital. You kind of want to force people into making hard decisions. I lived this—I’ve tweeted about some of these things during my painful existence at TrialPay, where I think we had to lay off 70% of the company, and then we eventually turned it around and sold it to Visa. There were all sorts of tough times in there.

You run into these very, very challenging scenarios where Option A is bad and Option B is bad. You have 2 choices; you’re at a fork in the road. There’s a funny expression by Yogi Berra, this famous baseball player in the US: “When you come to a fork in the road, take it.” It’s like, what does that mean? He said all these things that make no sense.

What a lot of entrepreneurs don’t realize is that the worst option is that you think you have 2 options, but there’s a third option, which is making no choice at all. That’s the worst option. You’re better off choosing something. Both of them are bad, right? It’s like, “This option is very bad. This option is very bad. Therefore, I don’t want to make any choice at all.” But you’re better off making a choice and committing to something.

If you have infinite capital, you could just continue this: “I’m not going to make any choices. I’m just going to sit here. All right, well, I have more money. My ARR is more driven by the interest on my giant $100 million cash reserve.”

Sorry for rambling on this, but this kind of goes to founder-capital fit. There’s a certain type of person where it’s like, “I give you a lot of money, and I know you’re still going to make decisions very, very quickly. I know it isn’t going to distract you, and really, it’s just benefiting me.” I hate to say it selfishly, but it benefits me in that now I’m on the cap table, I own part of this amazing company, and it’s not going to fuck up the company.

But the standard is that the moral hazard is the number-one thing. Now it’s going to fuck up the company, either with too much primary, or it’s like, “Oh, I know I won’t mess with the primary. I’ll just buy a secondary.” That also has existential risk, as I mentioned, for a certain class of person.

There are other CEOs where one of my CEOs did a very, very big secondary in 2021. He and the company have hit some tough times, but he has stuck it out, and he’s doing a phenomenal job.

7. Multiple Successive Rounds Are Dangerous… Here is Why?

How do you get comfortable about growing into that price that you have well overpaid for? Again, we’re super candid, and this is where I love where I’m at in my stage of life now versus where I was 8 years ago, because it wasn’t quite the same.

I lost to likely Seema on your team for Ask Leo. She’s amazing. You guys are amazing. Hugely well-deserved. You guys did not pay more than me. I hate this VC thing where it’s like, “Oh, they overpaid.” No, it was the same. You just beat me fair and square. Well done.

I reflect on that and I’m like, “You idiot. You should have paid $300 million and doubled them.” When I map out 18 months’ time, I looked at their revenue projections, and in 18 months, when they need to raise, their revenues would have been so much that I could still see a 3x on that $300 million. Idiot. That’s how I get comfortable with paying up for something. How do you get comfortable preemptively paying up so much?

I think it’s the same math, but it’s dangerous on both sides. I always have this speech that works maybe 1 time out of 100 that I give it, which is kind of the Spider-Man speech of, “With great capital comes great responsibility.” If you raise at too high a price, you’re fucked, because I lived this. Let me tell you my story.

I raised at this price for my Series C. Then I had Google wanting to buy me, but it was at the same price, so therefore it tanked the thing. Then, in my next round, everybody asked me what the price of my last round was, and nobody wanted to invest. I tell this story, and I can introduce the founder to 10 other founders who have lived the exact same thing. It’s like, “I wish I hadn’t raised my round at such a high price.”

But who starts a company? Let’s think about this for a second. The people who start a company are irrationally exuberant. If they thought the company was going to fail, if they thought they had a 0% chance of raising a Series B, they wouldn’t start the fucking company, right? That’s why the speech doesn’t work.

I always tell people, “The reason why you shouldn’t raise your Series A—” There was a deal that I guess we should have done, candidly, because this company just raised at a billion-dollar-plus valuation, but we turned it down. The company had less than $1 million in revenue, and they wanted, like, a $200 million, whatever, post-money Series A. It was just so crazy.

I was like, “Look, you guys haven’t started a company before. I have.” Not to pull the old bald-guy card—but more old than bald. The old part is relevant; the bald part isn’t. Your Series B, even if you have $20 million in revenue, you’re fucked. You have to be able to walk into a room, and the number-one question you’re going to get is, “What was your last round price?”

People should be wanting to compete to pay 3 times that price. They should be like, “Oh my God, what will it take to do this deal?” If you say, “Hey, my last Series A was raised at a billion, and I have $1 million in revenue,” you have ended the conversation. Nobody wants to invest. The psychology of that round is all wrong.

I give this speech, and it just doesn’t work, unfortunately. But I think the smart entrepreneurs have this risk-balancing thing. They’re irrationally exuberant—that’s why they quit their job and started the company—but they realize, “Oh, wow, there actually is a good point around my whole team now saying we have $100 million in the bank. They’re going to be wasteful. That culture is something that I don’t want.”

I guess I would want the option of maybe selling the company for $1 billion and having Salesforce come in and say, “What would it take to buy the company? What was your last round price?” I will tell you, 100% of the time, in every M&A conversation and every fundraising conversation, the number-one question—the first question—is, “What was your last round price?”

8. The Importance of Ownership in Deals

If it’s insane, they’re like, “Ooh, that’s not good.” Then, as an entrepreneur, you’re like, “Oh, no, no, but I would take a discount because my company sucks.” You can’t say that. It just destroys the entire conversation. It’s game over.

We mentioned the Rillet element in the successive rounds.

I hope it’s not too forward, and you can say, “Dude, I don’t want this in there.” By all means, I’m not a journalist. But you see the success of B because you lose the A. When you sit down—and we’re sitting down as a team—how do we reflect on that?

When you reflect on, like, an IC review, what was the takeaway from that when you sat down?

Well, there are a lot of deals that we lose because we’re not willing to go the distance on price. That is a common thing where it’s like, “Did we really lose it?” This has happened to us a number of times. It’s like, all right, we want to do the deal.

This is, again, consensus and non-consensus. A lot of times, the difference is just on price or ownership, right? If we had shown up and said, “Hey, we’ll do 10% of this company for an A round,” we could win every deal. It’s actually, I think, one of the competing elements that has shown up. I’m interested to watch how Standard Capital does. This is kind of the YC offshoot.

It’s like, “I’m going to take 10%.” That’s very, very bad for big funds because, in order to make the math work for a big fund, you have to have high ownership. You know that your ownership will get depleted, or will get diluted over time as option-pool expansions happen, even if you take your pro rata in every single successive round.

We can win all these deals, but a lot of times I am much more preoccupied with ownership at the A because we’re buying an out-of-the-money call option. The reason why I tell this story is because there’s something that I’ve used as a benchmark, which is: If you’re hiring people and 100% of the people say yes to your job offer, what can you infer from that?

Number 1, you could infer that you’re the greatest hiring manager of all time. But number 2, you might be overpaying, right? Would you agree with that? If you only get 50% or 20%, how do you know to test this hypothesis?

And if you win 100% of the deals, that’s a very, very good sign. You should try to win 100% of the deals that you want to do.

But if you’re winning them with very low ownership, you’re probably not testing this kind of efficient frontier of how far you can go. You want to have more ownership, right? That’s our objective. The founder wants less dilution; the investor wants more ownership. The 2 are perfect complements of each other.

Eventually, you realize, “I don’t want to be a fucking idiot.” This is the answer to your question, right? I wanted 20% in an A round for a company that doesn’t have that much traction because I’m at Andreessen Horowitz and I’ve got this big fund and everything else. But then it’s like, “No, no, they’re going to do a 15% round or whatever.” It’s like, “Oh, fuck that. I don’t want to do that deal.”

And then it’s like, “Holy fuck, they’ve run away with the market. This is the market leader. I’m not going to be stupid,” right? I’m not going to just say—this is why, actually, I love talking to investors, because investors, like most humans, do not have the capability to admit that they were wrong. They just want to say, “I’m right, I’m right, I’m right.”

If you’re an investor, you’re just going to lose money all the time. The most valuable insight that you can have as an investor is the self-reflection to say, “I’m an idiot.” If I’m a hedge-fund guy, I get to sell. It’s like, “Oh, I thought I was a genius buying Herbalife. Wow, this company’s not good. I’m going to sell everything,” versus, “No, I want to prove to the world that I’m right.” Well, I’m going to lose all my money.

It’s the same thing here, but for upside, we can’t sell. But we can say, “This is the winner. I want to be in the B at a lower ownership because this is the fucking winner.”

But if I was your partner, I would be pushing you with all my might to take the 10% at the A and have a higher win rate, specifically with your profile of fund. I get it in other funds where you don’t have the ability to follow on and lead the B, the C, the D. You may not even be able to do the pro rata, in which case I get that thinking. But when you can, why are we not having a higher win rate and doing 10%?

This is actually one of the things that we looked at because I kind of feel like my job here is kind of quasi-portfolio manager. I run our apps fund, right? We have 7 different funds, and my job is to make sure that that fund is as successful as possible. We’re winning the right deals.

If we just say, “Hey, everybody, win every single deal. Just win every deal. It doesn’t matter. That’s all I’m going to optimize for,” and we end up with 5% checks in every Series A, that’s not going to work, right? We can win every deal that way. What is the frontier? How far on this curve can you go?

It’s, again, the exact inverse conversation that an entrepreneur is having, where it’s like, “I want a tier-1 investor. I want an amazing specialist, or I want whatever. I want this person that I want on my board. What is the least amount that I can give up to get an amazing person?”

They would love to get 5%-a-round deals done, but they’re like, “Oh, wait a minute. That’s not going to work.” That’s the tension between the 2. I agree with you, but I think—where do you draw the line? It’s like Zeno’s paradox. You know what that is, right? You will never get to the destination if you go halfway each time.

Is it 9%? Why not just do it at 9%? Why not do it at 8%? Where do you draw the line on that? I would do the simple math of where do I think—and this is a very dangerous and bad answer to your question, because the biggest mistakes in venture have been when you underestimate market size and you don’t see what it can be—but I’d sit down with you and I’ll go, “Okay, 10% entry, 5% on exit, assuming 50% dilution. Do we think this can reasonably be a $15 billion company? If so, that is a number that returns the fund with comfort.”

I know, but the problem is it’s kind of garbage in, garbage out. You can always say that for something, because otherwise you’re like, “Oh, wow, I underestimated the size of the black-car market. Hence, I’m not going to do it.” It’s hard.

The way that I do it, just to be pithy about it, is that we either want to buy any percent—any percent, right?—of something that could work, that is absolutely working, or high ownership of something that could work. If you really draw a line, you have to bifurcate the market.

It’s like Facebook. If you look at that round, I think Greylock put $25 million into Facebook at—actually, I think the round was maybe $25 million at $500 million. I think that was the B round for Facebook, split between Meritech and Greylock. But that was absolutely working, right?

Are they getting 10%? No. Are they getting 5%? No. But it’s the market winner, and things can go wrong, but holy fuck, it’s absolutely working. I don’t see that many things that look like that, but when you do, you throw away all the rules.

Or it’s like, this is not working, but this person looks like a super genius. They have high agency. They can materialize labor, capital, and customers, but it’s not working yet, right? I have to have high ownership in order to correspond with that level of risk. Those are the 2 types of deals to do.

The danger is, you could say, “Oh, well, this has $1 million of ARR, and they’re ahead of the number-2 player that has $900,000 of ARR. Therefore, it’s absolutely working.” No. You have to have a high bar on the “absolutely working.” This is crushing. This is the fastest-growing company we’ve ever seen. It probably comes around once every decade. Throw away the entire rule book, and you should be fine owning 5% of that company because it’s an absolute winner.

I’m so pleased that you said “the fastest-growing company that we’ve seen.” We’ve never seen growth rates like we have today, and I’m a little bit stuck, if I’m honest, so I’d love your advice. When we look at companies going from 1 to 20 to 30 to 40, there are actually quite a few that do that today. Before, that was completely unheard of.

How much weight should we place on revenue growth today versus not? Is there a world where these companies that are going from 1 to 3 or 4—where 3 or 4 actually used to be good—are left behind?

If you want to know the 3 investment theses that I have for our fund—this is exactly what I told LPs—and we’ll answer your question in a second, I think we have 3. We have 1, which I call Greenfield Bingo.

Most of the greenfield companies—these are existing software companies, but selling to new companies, as opposed to selling to the hostages that will never leave. They tend to be systems of record, right? Like the ERP or vertical operating system.

The reason why I love Rillet so much is that it’s never going to grow from 0 to 100 in a month, but it is very, very sticky revenue. Once you’re on, NetSuite has hostages, not customers—they’re not going to leave. If Rillet can sell into every new company, they’re going to do great.

The revenue growth will be slower, but it will be so sticky, and they have infinite option value on adding, “Hey, do you want to have a collections AI agent that runs on top of overdue invoices?” That’s optionality on top of your sticky system of record.

So, number 1 is greenfield systems of record. Number 2—and this goes to the fastest-growing companies in the world that you're talking about—is software that does the job of labor. These are new. I'll give you an example: We have a company called Eve. They sell into plaintiff attorneys. What is the dominant software product for plaintiff attorneys? It's called Microsoft Office, right? There isn't one. There are so many categories—what's the dominant software for manicures? There is none. You could pick all these areas where there's no greenfield being built; there's just nothing.

But because what you're selling is effectively in lieu of labor, the way that Eve works is, if you're a plaintiff attorney and you get paid on contingency, you're not charging by the hour. You have a case where you will, with 100% certainty, win $1,000. Will you take that case? The answer is absolutely not, because it's not worth your time. So you turn down all the small-ticket cases because you want the big-ticket cases. But now you have a software product that can do all the work and help you win all the small-ticket cases. You're absolutely going to do that.

These are the things that scale like crazy, right? Instead of hiring somebody for $80,000 a year that I cannot hire, I can now hire this software product for $20,000 a year. Before, I was paying $0 a year for software. Those are all the things that are hyperscaling.

But to your point, if they don't eventually back into a system of record—if it's something that just does outbound phone calls with an AI agent, a thin wrapper on OpenAI plus 11 Labs plus something else—it will attract so much competition. It won't be sticky. So, the conversation that I have with every entrepreneur that has one of these companies is: How are you going to make this sticky? How are you going to, pardon my language, get the hostages? How do you hold these customers and make sure that if—

I'll give you an example. I'm an investor in a company called Saliant, which is probably the market leader in outbound loan servicing for autos. This is the conversation I had with Ari: What if Taliant shows up—the competitor of Sallient, the make-believe competitor to Sallient—and says, “Hey, we're going to do it for 50% cheaper”?

I loved his answer, which was, “This is my wedge, right? I recognize that this is not super sticky if we're just making outbound phone calls and combining these different layers of the stack, because we're not the infrastructure layer. But we are going to back into a software product.”

I love that answer, and it's true. That's what they've done. So, that's my answer to your question: You have to back into this mega-revenue growth that is largely predicated on doing the job that people would do before. That's why you can grow so quickly into a sticky software product that is not that dissimilar from software products of yesteryear.

If you want to know the third thing that we're doing, number 1 is greenfield systems of record, number 2 is software that does the job of labor, and number 3—I wrote a post about this—I called “the walled garden.” There are two examples of this.

There's a company in Europe called vLex. vLex was started by an entrepreneur who basically bought up every legal record in Spain—physical legal records at the courthouse—put them into digital form, and then started selling them to law firms. I think he got this to something like $20-something million of ARR after 25 years, but then added AI and it grew something like 5×. I mean, something crazy. Why? Because let's just say OpenAI is purely a sentient being: AGI is here. OpenAI has done it. Tomorrow, GPT-5.5 is here. If you say, “Hey, help me draft a response to this Spanish court case,” they don't have the data, right? They can't do that.

OpenEvidence has done this for health data. AGI is here. OpenAI has it. Amazing. I tore my Achilles. What do I do? I'd rather have GPT-3.5 plus infinite data on everything around medical science, which is a walled garden that OpenEvidence has, versus a sentient being that has no data whatsoever. So, that's also a very, very powerful way of building something sticky.

9. Is Triple, Triple, Double, Double Dead?

If you find a company that has grown like this or grown like this but just cannot be removed, either because of the data that it has that is unique to it—which is, honestly, my hope with Leo—or because it has a sticky system of record, it's just not going anywhere. Versus other ones, you might take a flyer: “Wow, this has grown from $0 to $100 million. They make outbound phone calls, and they're ElevenLabs plus this plus that. It was all built in Lovable, and it's amazing.” But that’s a harder pill to swallow.

I'm so honest these days, dude. I'm too old and ugly not to be honest. We're in this business called Aloe in Germany. It's like a Toast for Europe, but a little bit better—specialized to the European market. They've got great numbers, like 5× from $500,000 to $2.5 million. Dude, raising their Series A—like, $8 million or $10 million on a $50 million pre-money valuation—which is horrible. I was just like, “Oh my God, the triple-triple-double-double is so dead.” We're in Lovable as well. That obviously is a completely different fundraising journey.

Is the triple-triple-double-double dead for those?

No. No. I think it might be harder for a certain set of people who are maniacally focused on growth over everything else. But what really matters is growth and stickiness. If you're triple-triple-double-double with terrible retention data, that's going to be very hard. But if you actually have, again, a system of record—or, in that case, it sounds like a vertical operating system—that should not be hard. I mean, I would do it. I love those things.

I would much rather have a slower-growing, permanent system of record that will never get ripped out than the fastest-growing thing on the planet that has 9,000 competitors, all built in Lovable by 17-year-olds. I think there's no comparison. There are plenty of people who would be attracted to both, would be my answer. I'm surprised that it was as challenging as you portray it.

10. Advice on Selling Companies

We got it done, but I was surprised, too, by how challenging it was. Alex, you mentioned selling companies. I spoke to David George before the show, and he said one thing he's never talked about publicly that I think he's a phenomenal master of: advice on selling companies. Ask him about that. I know it's a bit broad and random, but I do want to touch on it because David said I had to. What's your biggest advice on selling companies, having seen so many and lived it yourself?

Yeah. I'd say a couple of things. This is a very highly choreographed dance. You can't just say, “Oh, I should raise.” If you're raising money, you're like, “Oh, I should raise money. I have the best metrics ever. I'm going to talk to 5 firms, and they're going to compete to the death over winning my deal.” That was my experience with my Series B at TrialPay.

Corporate development is like, “I'm either raising money or selling my company. It's the same thing, right?” No, it's completely different. If you're selling your company, you have to spend, in many cases, years just getting to know people at the potential acquirer. It's never the CEO, unless you're likely Jan Koum at WhatsApp.

Let's just say that you have a company and you do something amazing. Somebody at Salesforce should buy it. You would rather go public, but you're kind of seeing the writing on the wall: “I'm going to hit a wall in a year and a half.” What you should start doing then is—I call it a background process. If you know what cron is in Unix terms, right? You should have a little cron job where 5% of your time as CEO should just be getting to know people at the 3 or 4 companies that might buy you.

You never go say, “Please buy my company.” That's DOA. You don't want to spend time with the corporate development people because most people think, “Oh, corporate development buys companies.” No, they don't. They execute transactions. If Salesforce buys your company, you're not working for the head of corporate development, right? You're working for this SVP who has some hole in their personnel or needs revenue growth in order to get their bonus. There are all sorts of internal mechanics going on there.

It's just this highly choreographed dance of making sure that you get to the right people in the company, hopefully doing it years in advance, not just going to them when you need to sell your company. There are 2 independent variables here. The best time to sell, by the way, is when your company is doing great—when the rocket ship is doing 100× year-over-year growth and they want to buy, right? But rarely does that intersect. A lot of times it's like, “Oh, shoot, we started going like that.”

Now we want to sell, but nobody wants to buy this falling knife. It’s hard to perfectly choreograph this, but the main piece of advice—and I wrote a whole Twitter thread on this, I think—is to spend time with 3 or 4 companies, hopefully not under the guise of selling them your company.

Honestly, when I did this at TrialPay, I wanted Visa to be a partner of mine. I wanted PayPal to be a partner of mine. It was not wasted time. It was, “Hey, PayPal, you should put coupons that we do for this post-transactional product on your receipt page.” I was spending so much time because if I got that deal, I didn’t give a shit if they bought us or not. If I got that deal, it was worth so much money to us and so much money to them.

Unfortunately, or fortunately, depending on your point of view, they were like, “Oh, wow. This is so valuable for us. We have to buy that company.” That’s often how it works. It’s like that movie—my favorite movie is Inception. How do you incept this idea?

Again, in that movie, it happens overnight on a flight, whatever, from Australia or something. But it really needs to happen probably 1.5 to 2 years in advance. A lot of entrepreneurs make the mistake of thinking, “I have to go impress the corp dev person.” Wrong. “I have to only interact with the CEO.” Sometimes, right? We hosted a dinner for the CEO of Visa, and I sat Zack at Plaid right next to likely Alec Kelly at Visa. That worked until it didn’t, because of the Justice Department or something.

That can happen if it’s sufficiently strategic. These $5 billion acquisitions don’t happen very often, but a $500 million to $1 billion acquisition can happen without being at the CEO level. You just have to spend the time and invest the time and resources. By the way, this is the same advice that I give people on fundraising. It’s a background process.

If you’re the CEO of a company, your number-one job is: don’t let the company run out of money. That either means you become profitable, which is great; you raise more money, which is not as great, but hopefully leads to being profitable; and/or you sell your company. You probably should spend 5% to 10% of your time meeting investors in a very casual way, so that they know you, they know that you’re a very strong entrepreneur, and they can invest on the spot.

This is how I raised my Series D at TrialPay. I had spent so much time with the Greylock guys, as an example. After I met likely Reid, I pitched them like 20 times. He knew me, so he knew that he trusted me. He was investing in me, as opposed to a random dude who shows up and says, “I should raise money because I’m running out and I’m growing. Let me go pitch 5 partners.” They never would have done the deal otherwise. The background process is key.

Before we do a quick-fire round, I just have to ask: you mentioned labor displacement as 1 element in 1 of the 3 underpinnings that you have. I completely agree. My friend likely Jason Lemkin said this year will be the year where we see the demonization of technology leaders and labor displacement materially show up in labor markets. Do you think that’s true? Will we see labor displacement materially show up in labor markets this year?

I’m not sure about that. I think in certain areas, for sure. In general, I could even kick up a notch. If you think about SaaS, broadly speaking, I think there are 3 types of SaaS companies right now.

There are the ones that are almost impervious to everything that’s happening with AI, and if anything, it’s a huge tailwind because they have the distribution and they’re going to start adding features. That’s things like Workday and NetSuite, where they have the hostages. They’re never going anywhere.

On the other side, you have things like Zendesk, where it’s like, how many licenses per seat do you need of Zendesk? If now every customer-support ticket can be answered automatically, you need 0 licenses. Their revenue could go down 100%. These are very, very different.

Then you have things in the middle, like Adobe, where it’s like, “Maybe now, whenever I want a logo, I just go to ChatGPT. I don’t go to the graphics team.” Maybe you’ll need fewer graphic designers. Maybe with Zendesk, you’ll need fewer customer-support people. That probably is true. There are going to be certain areas that get hit harder than others.

But what technology has always done is people shift into other jobs, or maybe some people will be 100 times more efficient. Take the Eve example that I gave you. Now I can do 100 times as many cases, or 5 times as many cases, as I did before. I’m going to hire 3 more people. Or I can now be in business by myself because the software helps me do X, Y, and Z. I think a lot of that stuff is going to start happening.

I respect you so much, but when you look at likely Decagon in customer support, it’s clearing out. When you look at Harvey, another business that you’re in—

I don’t disagree. I’m saying it’s not unlike—that’s why I gave the example of the 3 types of SaaS. You’re going to have some that are totally impervious, and I’m talking about SaaS, not people. If you flip that to people, the users of Zendesk are probably going to go away. Therefore, that labor market might get decimated. 100% agreed.

On the other hand, if I’m United Airlines and now I don’t need as many customer-support people because every answer kind of auto-answers itself with AI, you know what? I should probably take care of my best travelers better and give them a personal human who will be really nice to them and remember their birthday. Then they’re going to buy more first-class tickets from me.

I might reallocate some of that labor to other things because I’m making more money and I no longer have this cost. likely Tony Hsieh, who you know, sadly departed, ran likely Zappos. He had this whole thing, which I think is actually correct: most people think of customer support as a cost center, but they should think about it as a revenue center. You should love your customer and make them love you.

There’s a story that he would tell about somebody who had something really bad happen and was on the phone with a customer-support person. I think her husband died—something bad that had nothing to do with the shoe order. Zappos sent that woman flowers. Doing things like that, making your customer love you, is something you can now focus on once you take away the cost-center element of something like this.

Or if I’m a law firm, again, I agree with you. You probably don’t need people doing this tedious work, and the number of people doing the tedious work will fall off a cliff. No disagreement. But I would not be surprised to see smart companies start reallocating them.

I gave a talk to the executive team at JPMorgan about this. They asked, “What part of our business is going to be least touched by AI?” I said, “Wealth management.” What is wealth management? Hopefully, it’s getting good returns for the dollars that you have with us, but it’s really that relationship guy or gal.

The woman who was running wealth stood up in the audience like, “Yeah, yeah, yeah.” But it’s true. If you have a high EQ and you’re good at playing golf with people, you’re going to start hiring more people like that because that’s how you get more customers. You reallocate.

Sometimes there will be an opportunity. The upskilling is not, “Hey, everybody should learn how to code.” The upskilling might be, “Stop doing tedious work, like looking at a knowledge base and then typing that back, with lots of typos, into the email response in Zendesk. Actually start sending customers flowers. Get to know that customer really well. Go visit them,” at least for the high-value customers that you just couldn’t do that for before.

11. Quick-Fire Round

Alex, I could speak to you all day. I know you do actually have to work as well. I want to do a quick-fire round with you. I’m just going to give you a couple of quick statements. What have you changed your mind on most in the last 12 months?

I’ve probably changed my mind on—well, as I mentioned, you have to be able to change. It’s more about companies where we didn’t do the early round. We often talk about, “I’d rather be rich than right.” We’ve probably done a couple of deals where we passed on round N minus 1 and ended up doing round N.

But I don’t think I’ve changed my mind on that much. Maybe I would say this idea of private-equitizing venture capital. I wrote a piece—I was probably the first one to talk about this in 2023—about how you’re going to start buying a company and then adding AI to it. I think General Catalyst and a bunch of firms are now doing this.

I was the first person to talk about this, and I called it “Barbarians at the Gate with an AI.”

I’d probably become more bearish on that just because it feels like a founder-market mismatch. So that’s probably the thing that I’ve changed my mind on the most.

What product does a16z not have today that you would most like a16z to have? You mentioned General Catalyst having the fund that does that roll-up play. They’ve got the consumer performance marketing fund—I can’t remember what that’s called—but what product do you not have that you would most like to have?

I probably think something around credit for a lot of our companies. We have equity products, but we don’t have debt products. They have very different return profiles, obviously, but every one of our companies needs them. General Catalyst actually has one of these; they have a credit fund.

That would be either for customer acquisition or, if you’re a fintech, for doing lending. That would be interesting. But, in general, we listen—we don’t want to be at odds with our entrepreneurs. There’s a very solid reason why we don’t have that, which is, “Oh, you didn’t pay back the bill; I need to go foreclose.” As a venture capital firm, you can earn a 1,000x on a winner. You don’t really want to beat up the companies that are struggling, and that’s kind of what the credit instrument needs to do. But I think it’s a good product.

What piece of investment advice has stuck with you the most? Josh Kushner once told me, “If you’re willing to take less, don’t do the deal. If you’re willing to go from 10% to 7% and say, ‘Yeah, sure, why not?’ don’t do the deal.” What would yours be?

I think it really is: find high-agency people who know the history of the space, who can materialize labor, capital, and customers, who are the Count of Monte Cristo, and don’t second-guess anything. Just give them money, be their best partner, and go, versus questioning the market or questioning the thesis. I think I’ve just become 100% convinced that this is entirely about people. 100%.

Every round, by the way—it could be a Series D, it could be a Series E, it could be a Series A, it could be a seed round.

Can you please tell Martin Casado? He tweeted and then took the piss out of me because there’s this graph where it starts here, then goes up here, and then goes down here. It’s like, you start here, it’s all about the founder, and then you end here, it’s all about the founder, and here is when you think you’re smart: no market and product. He was like, “You’re an idiot.” But it’s not that—

It’s all about the founder. I mean, you have to—again, it converges on reality at some point in time. There’s going to be a public company. You can’t tell everybody in the order book of an undersubscribed IPO, “No, no, no, the founder is really good.” Yes, of course it has to converge on reality.

But I think it is materialized labor, capital, and customers. That’s kind of it for me, with the right motivation, which is the Count of Monte Cristo.

Penultimate one. What’s your biggest miss, and how do you reflect on it? I missed Deel’s seed round, another one of yours. How do you reflect on that?

It probably was one of the first rounds of Plaid, which I subsequently corrected myself for by doing the Series C of Plaid. I think we invested at a $2.4 billion valuation for the Series C, and I was debating a $5 million difference with Zach for the Series B. I think I wanted to do it at $130 million, he wanted $135 million, and I think Goldman was willing to pay $200 million. But he was willing to work with me because of my fintech.

It was just so stupid, right? Luckily, I was willing to admit that I was stupid and did the next round. You could see the difference. This is why it’s so important to do 2 things: to correct yourself if you’re wrong and not be proud about it, but also, if you really believe that this can be a huge company, to act on that belief.

I was burdened by what has been, to quote the likely Kamala Harris: “Oh, wow, Yodlee,” which had predated Plaid, went public, and had a terminal valuation of $600 million. So, of course, this $130 million versus $135 million, or whatever the hell we were talking about, was very material. But it was so stupid.

12. What is the Future of Venture Capital

I love that “unburdened by what has been” meme. Yeah. Final one for you, dude. What does venture look like in 5 years’ time, when we look at the $15 billion that you raised today? I mean, it is obscene to even think that would happen 5 years ago, when we go back. What does it look like 5 years out?

I think it ends up eating even more of the world. This is kind of going back to Mark’s essay, “Why Software Is Eating the World.” That has largely happened. As I mentioned, the 5 biggest companies on Earth are technology companies, which was unthinkable in 2005. Technology companies were little service providers to big companies like banks and oil companies, right?

So I think this momentum of everything becoming a software company kind of goes into this thesis, too, that I mentioned around software doing the job of labor. You’re going to have all of these areas where there’s going to be, you know, Toast. Vertical SaaS proved this, or kind of V1. It’s like, “Oh, how is Toast worth $20 billion?” You’re going to have a lot of things like this where there are brand-new markets that have grown like crazy.

AI is now allowing software and technology to do so many things that it didn’t do before. And this is before even things like robotics. If robots actually work, wow—now you’ve expanded the market another 100x. So I’m just so bullish on the ability of technology to create enduring value. My guess and my hope is that it’s going to go up and to the right.

Dude, I told you, this is like venture. You have most shows which are fine, and then you have the ones in a while which are truly special. Thank you for being my truly special show. It really is rare to have one like this.

All right. And hopefully I’ll see you in London soon.