Larry Aschebrook，G Squared 创始人兼管理合伙人：我们如何在 Uber 上亏钱，却在 Lyft 上赚了数百万美元

We made ridiculous sums of money for LPs in that period. We were the largest shareholder of Coursera. We returned $800 million to LPs on Coursera. We made a ton of money on Lyft, and we lost money on Uber.

What?

40% of my third fund went into Spotify. I went back to our LPs and said, “Listen, we screwed up. We need to pivot. We need another $300 million because I need to protect this thing.” The next 3 years were some of the worst of my life.

Larry, dude, we walked around the park and I heard your incredible story. To be fully transparent, I didn’t know the incredible story before, which is why, at the end, I was like, “Dude, we have to do a show together.” Thank you so much for doing this with me.

That’s humbling. It’s sometimes difficult to open up and tell the story, but I enjoyed our walk.

Listen, it’s the short shorts and the great legs that make you feel comfortable enough to open up. I completely understand.

I want to start with the entry point. “Dialing for dollars” is kind of how I was thinking about this. How did you make your way into venture, and what was that entry point?

For me, it wasn’t really, “Hey, I want to be a venture capitalist and manage billions of dollars.” I come from nothing. I was a fundraiser for academic institutions and their endowments. I was good at that, and most of the people I raised money from made their money investing in private companies. They were at private equity funds, venture funds, or running their own operating businesses.

It didn’t matter what it was—from Windows to financial management to private equity—they created real wealth for their families. I thought, “They’re not that much different from me. I work hard. I’m smart enough. Maybe I can do it.” So, I went back to business school late in life, and our business today was my thesis.

The smartphone became something that was running our lives in 2010, and I started saying, “This is super interesting. Why don’t I buy shares?” I followed Twitter. I liked Twitter. What about early Uber? What about early Spotify? I started buying shares from my classmates with my own money.

That sounds great, but these companies weren’t public at the time. How does one do that? How did you approach it?

I guess not knowing what you don’t know is ignorance is bliss. I didn’t realize it was something you shouldn’t or couldn’t do. I just asked people, “Do you have shares? Can I buy some of your shares?”

They would say, “I never really thought about selling them because I really can’t sell them.”

I’d ask, “How would you buy them?”

Even today, 15 years later, there’s a form that I created at the W. P. Carey School of Business at Arizona State University—a one-page form that we sent to the company to buy stock. It still floats around. I got it back from a broker not that long ago, and it was literally the form I created in 2010.

And it’s binding, though. That’s scary.

It’s binding, which has bitten me in the ass a few times.

So, we’re at business school, and we’re like, “You know what? I see the mobile revolution. I’m buying Uber. I’m buying Twitter. I’m buying Spotify.” What happens then? Also, how much were you buying?

Small amounts. I wanted to make a change. I was working in college athletics, and that was changing into what it has become today—the whole thing about players getting paid and coaches’ salaries. Pretty soon, the inmates were running the asylum, so to speak, and I thought, “This isn’t my passion. That’s not where I came from. I want to make a change as a former athlete. Maybe I can use my brain, not my brawn.”

I cashed in my retirement. I went through a divorce, got remarried, and had nothing. I had no money. I had just paid the tax on my retirement and started buying these shares.

And this was literally your last money?

Yeah. Well, half of nothing is nothing. My wife came to our marriage with $50,000. I had gone through a pretty tough financial situation. Rightfully so, my former wife kind of took what I had. I was raising my 3 young kids at the time.

My wife and I started a journey together, and she backed me with the $50,000 and my small amount of retirement. I started buying Twitter shares and Alibaba stock from Jack Ma’s family office. Then, holy shit, it worked.

When you review those today, which was the single best investment on a multiple basis?

The early ones. Twitter and Alibaba were pretty good. Those were when I realized there was something here. Making a little bit of money in your whole life, working 10 years in a profession, thinking you’ve kind of made it—coming from where I came from, growing up in an orange trailer in Utah—having $100,000 in your bank account on one random Tuesday felt pretty good. I said, “This is something interesting.”

I remember the first time I tried it, I got some advice: “You should raise some third-party capital.” That’s when I really understood the power of OPM—other people’s money. I went around to all those alums who, basically, I think gave me some money because they were tired of me asking. I had worked at 5 different large academic institutions.

Just so I get it right, what was the thesis?

It was very simple. If there were fewer institutions to help companies go public, coming off the financial crisis, funds would start deploying large sums of money. The byproduct would be that companies would stay private longer.

The average age from inception to IPO was about 3 years leading up to 2010. From 2010 to 2018, it got to 7 or 8 years. Today, our average portfolio company is 15 years old. There was something clear there to me: There was an opportunity to buy shares from people who had no liquidity, and you should be able to make more of a return.

Also, companies wouldn’t return my phone call, frankly, because I had no money. It was, “Hey, Twitter. Hey, Dorsey. My name’s Larry Aschebrook. I’m a retread athlete who wants to become a venture capitalist. Can I invest in Twitter?”

By the way, my fund size was—I didn’t have a fund. I just had $200,000 in my name. I’d like to buy $10,000 worth of Twitter stock. That conversation obviously wouldn’t go anywhere.

You had this moment when you were thinking, “We’re seeing this extension of the private markets. We’re seeing a lot more capital flow in.” So, you went out and raised for the first vehicle. How big was the vehicle?

It took 3 years, with $35 million deployed along the way.

Wait, pause. It took 3 years to fundraise?

Yeah, 3 years—from 2010 to 2013. It was $34 million or $35 million, and I deployed the money as I raised it.

So, you did multiple closes?

Oh, yeah. Multiple closes.

What was the first close?

I don’t even know how many. A couple million bucks.

What do you advise founders on closings? I know it sounds strange, but you get different advice: Close as soon as possible, 50% in one close.

Founders for companies or founders for new managers?

New managers.

First of all, if I knew then what I know today, I’m not sure I would have started the journey because I didn’t know how hard it would be. I just wanted to do something different with my life, try to create value for my family, and try to set my children up for a different future than I had.

I was lucky. My siblings grew up in the same house as me, but we don’t have the same life today. I got out of the squalor because I could run, throw, and catch. My siblings didn’t have that benefit. I didn’t want that for my children. Who knows what their outcome would be?

The whole genesis of it for me was trying to create something, seeing an opportunity, and capitalizing on it. With my mentality, still to this day, if you’re raising $1 billion- to $2 billion funds, the money’s there: You close, start deploying it, build a portfolio, and show some improvement in NAV. It’s inertia that makes it easier.

I don’t have the benefit, even today, with the returns and DPI we’ve had, to say, “Hey, I’m raising a new fund,” and have $2 billion show up in 2 months. We don’t have that benefit, and I’m okay with that because I think the way we’re forced to raise money makes us better.

In 2021, it was very different. We both sat at home. I was sitting at my farm. You sat in your underwear, probably, and raised—I don’t know—in 3 months. I don’t know what you raised: $150 million bucks?

We raised $1.4 billion and then turned down another $700 million in a very short period of time. Companies, by the way, also raised a ton of money—insane, insane, insane numbers.

We’re going to get there, but I want to tell the story before we move into that. It’s such an amazing story. When we look back at that—3 years, $35 million—what was the first big mover in that portfolio? Where was the real momentum?

The real momentum happened in 2014. That’s when I really felt the power of what I was trying to do—when Alibaba went public.

And you had that in the fund?

I had it. I bought shares from Jack Ma’s family office.

Wow. What a creator of shareholder value. How did you get in touch with Jack Ma’s family office to buy shares from them?

That’s the story of my life. It’s a bit of luck. I met an alum who knew somebody who knew somebody and introduced me to a gentleman named Barry Pcell [?], who was running some of Jack Ma’s money in Virginia, of all places, I believe.

He said, “We have some shares to sell.”

I said, “Okay.” I hired a group in India to do some due diligence because I had 2 people, and I wrote an investment committee memo.

Oh yeah. My early investment committee memos in the early vintages were outsourced to a group of really smart analysts in India that you could pay a fraction of what you could pay talent in the US. I didn't have any money to invest in talent, and that today is what Sam Altman does with Deep Research.

Alibaba was the first one for me, and it was, “There’s a business here.” It was grinding, grinding, grinding. That was the first one that was material for me. It was 2014 Alibaba.

And were you putting size into these?

Yeah. From the beginning, when you don't know what you don't know, I didn't like the idea—and my personality is kind of, you go big or go home. I didn't like the idea of trying to manage a lot of these positions because I didn't really know how.

My idea was that I wanted to put a little bit of money in, understand the businesses, and then pick a few and put all my money in. Today, we call it “land and expand.” We have all this stuff that we say in our marketing: our team is second to none, reserves are heavily modeled, but we're diversified on entry.

See, it sounds like that, but really, for me, it was, “Fewer companies.” Ultimately, how do I get liquidity if I have 50 companies? I don't. It's my money; I don't have much. I've got a few people who trusted me, and I want to get their money in and out as quickly as I can.

As you're trying to raise that fund—and you know what it's like—what do people want? Track record. They want pedigree. They want, “What's your TVPI going to be? What's your MOIC going to be?” I had people who were running their own businesses give me a little bit of money and say, “I want the money back quickly. I want optionality.”

From the early days, it was about how I could build a strategy that I could actually make a good living at, with velocity of the capital coming back and optionality. Today, our business has become that.

The early generation of it was to start off with little checks because I didn't have much money. As I raised more money, the challenge of raising the first vintages was that you didn't have a lot to deploy. So, I was deploying it as I had it. Inherently, I was doing this land and expand without even knowing it.

Then you look back at the returns and say, “Holy shit, it works.” You start off with a couple of small checks, and then all of a sudden, you keep getting the data. I was sending it to my awesome group in India to give me the data back. I was sending it over there. It's amazing—you work across 24/7 time zones.

Anyway, what comes back is, “Hey, these 5 companies are better.” So then what I did was put a lot of concentration in, and I still do to this day, which some people are not comfortable with: 80%, hopefully 90%, of our risk is in 10 companies.

So when we go back to that first 35, how many companies was that, give or take?

Like 7.

7 companies.

But most of it is Alibaba.

This is funny. Alibaba. This is just dumb luck. Come on. You're not that smart.

Alibaba, Spotify, Palantir, and Twitter.

Okay. No, you're not that smart. You're either the best stock picker ever, or—

I'm not that smart.

But you can't get that lucky, and so you have to actually be—

Because if you just did one, I would have—

What do all those have in common?

Right. It's the—now you look back and say, “Well, how do you do that?” There were some dogs in there. Blame some cleantech shit that Kleiner Perkins sold me that was just a flaming bag of turds on your front porch.

“Look at this guy. He's energetic. Let's give him some garbage,” you know?

That's what the old-line Silicon Valley firms did in the beginning. “Oh yeah, we'll let you in the syndicate. Sure.”

“Oh, well, how much are you putting in?”

“Oh, we're putting in very little, but our LPs are co-investing a lot.”

That's an alarm bell. Now that you know, it's ringing as loudly as a 4-alarm fire. When Doug Leone calls you and goes, “I've got something just for you. You're going to love it. It's really going to help your first-time manager.”

So what do those 4 have in common, then, when you look back on them and reflect?

The timing of that period, no liquidity, large private valuations, and really differentiated, unchallenged business models. None of them had anyone doing anything like them.

The liquidity window was still early from the financial crisis, because when you look at the number of IPOs from 2010 to 2014, there weren't many. There was a pricing premium applied to them.

What I mean by that is, if you think about that applied to today, your SpaceXes of the world or your—

Yeah.

—which we had in our 3rd vintage. We can get to that, but there's a pricing premium on it where you're paying an exorbitant price for that defensibility and for that moat.

Yeah. Today, probably, there are no options for liquidity. It wasn't that—it was that you actually could get really good value.

One of the interesting things about the cycle we've been in over the last 15 years is that now we look back at the funds and dissect them and say, “What kind of value did you get by being a secondary direct buyer?” In the early vintages, we got about 35 cents on the dollar by being a secondary direct buyer over primary buyers, because there were no other secondary buyers.

So, for a lot of reasons, I give myself a little credit: hard work, drive, running through a wall regardless of how thick it is, and just keeping at it until it falls. That's my mentality, which is good and bad.

You're really leaving the brawn, not brain, department, aren't you?

Yeah, right. I'm trying. I'm trying. I'm just running through a wall. It's there, but I'm going to break it.

No, but I think that in the early vintages, up until 2020—all of those vintages—you were getting a lot of value by being a secondary direct buyer. Now, that's a key differentiator.

What I wanted, and still to this day what I want, is to touch and feel the founders, the companies, and the data. I was amazed by what I started. The paradigm that opened for me mentally was something that I had never tapped into in my entire life. This was really interesting—what these companies were doing.

Do you think your business model still applies today when you're so detached from the data and the founder at the level that you're going in now to a lot of these secondary businesses?

That is the misconception of what we do, and I love it, because when a good LP has the light bulb go off, they realize that it's the opposite of that. We get primary-level data. We touch and feel the founders because, actually, the value of the discount is back, and they value the work that you do today because there's still a massive need for liquidity, but they want a trusted partner.

That's part of the journey of how we got here today: by doing something different.

But sorry, the value of the discount is back. Are you kidding me?

No, it's amazing.

No, no, no. But for the premium assets—

Yeah, you're coming in at cost.

No, not always. The market on direct secondary buying is really fragmented. You have the people—

First of all, in the US, you have to be regulated in a bit of a different way than most fund managers to do it at scale, which is through an RIA.

Yeah, a fully registered, kind of like a hedge fund.

You're seeing a lot of people opt into that. Why? Because they want to pair primaries and secondaries together, and they want to do some one-off secondaries with big checks.

One of the things from the early vintages that teased out, and that we still do today, was the frequency of transactions and touch points. Really microtransactions—sub-$2 million, let's say today on a $2 billion fund—give you a lot of interim data that you would not normally get at that touch point as a primary investor.

It starts to provide you that Trojan horse moment to know when to triple down on Wiz, which we did in our 2022 vintage fund, and have the outcome that we're going to have. Those founders are amazing. But to make 3x in 18 months is because of what they created. The opportunity to do that is because you offer something different than others.

If you were going to start raising a fund today, what would you do?

Raising money? Don't take no for an answer. Open every door. All those things. Deploy as you go, but also create a firm that's different.

The world doesn't need just another early-stage seed manager, growth manager, or crossover fund. There are so many.

Dude, I agree completely. That's why we have media companies. You differentiate your business by sitting here and doing this and being in the know. How many awesome people do you get to talk to, to find and get the references to the next great thing?

You have figured out a niche. Most don't. Most say, “I worked at XYZ old-line firm for 10 years, and I know how to operate. I'm going to start a VC firm, and I'm going to go to 10 LPs that were in that old one that I got to know and ask them to give me some money.”

And they fund them. They really do. Right before this, we were saying, “We love the spinouts. You spin out of Accel—or name your big firm—and we fund you. Great.” That also provides the ecosystem for us to operate in.

Totally get that.

I want to go back to the 4 names you mentioned. How did you get Spotify? What's the story there? I love Daniel. He's one of my oldest friends. Shak as well. They're all special. How did you get Spotify in 2014?

One of the first employees I hired was a young guy out of Berkeley. He's now my co-PM, Spencer Mlot. You've got to meet this guy. He's a living caricature of what Silicon Valley is. He was working as an analyst with me.

This was 2014. I think he was 22 years old at the time. I was telling him the story about how my mind exploded when I was in college and I could download Metallica on Napster. He said, “You know, you’ve heard of Spotify, right?” I said, “Yeah.”

He said, “Well, maybe we should buy some shares in Spotify.” That’s literally how it started. We started searching around and found a celebrity who was going through an unfortunate change-of-life scenario and a divorce. He had about $4 million of stock, and we had a $300 million fund at that point. I said, “Yeah, let’s buy it.”

Spotify had this interesting process for approving shareholders. It was onerous and difficult, and my mentality was, “They’re not responding, so let’s go see them.” Spencer and I flew to Stockholm without an appointment. My wife came, and we spent a week there in the terrible Nordic November.

We had a new baby and were staying in an Airbnb that we rented, which became a great investment for us later on. Anyway, we started trying to get a meeting. Finally, this young lawyer, Peter Gandelius, took pity on us in the waiting room and took a meeting.

We pitched how we could add value with a small fund and a lot of co-investment. Spotify is what it’s become, but they said, “Yeah, we could use the help.” They were very humble and very nice people. They introduced us to a gentleman named Johan Bjurquist, who, at the end, I think, was the treasurer and then went on to Bolt, where we also became an investor in the EU ride-hailing business.

Before we left the meeting, they asked, “Do you think you could buy $150 million worth of stock?” I said, “Sure.” I didn’t have the money because we were in the process of raising the fund, and I think we had closed on about $125 million.

Had you done the diligence? Had you done the diligence?

Yes. By that time, we had done some due diligence. We had written a memo, and there’s diligence and there’s diligence. There’s diligence for a $20 million investment. This was mosaic theory.

Most of the early work had to be mosaic theory because, until Spotify, information wasn’t really available. The early wins came from gathering data, searching through public sources, and trying to gather information through contacts. We put as many feelers out as possible and then sent all that information to what became our own research team to develop a thesis around it.

The Trojan horse became the check to get the information. In the land-and-expand strategy we’ve developed, that first check is often the Trojan horse because you like the company, you’ve done some mosaic-theory work, and you think it works, but you don’t really know.

We go in, and they give us a deep dive on the business. It was amazing. It was like nothing I ever could have imagined, like nothing I’d seen before.

Why? What was it about it that was amazing?

At that time, the penetration of the product in Sweden was about 25% of the population. I mean, that’s amazing. Twenty-five percent of the population used the product, and you’re like, “It’s already cool because you’re using it.”

Of course, there was Apple Music, and I think Beats, Tidal, SoundCloud, and all these competitors. But what they divulged to us in that one-on-one meeting, after saying no to us for 6 straight days, was the fact that the record labels were also investors.

To me, that was like, “Okay, this is Napster on steroids,” and it worked. Obviously, they went on to create amazing shareholder value. It’s one of the best businesses we ever invested in, and we had $150 million to go find.

So what do you do then? They’re like, “Hey, we’ve got $150 million.” Are you shitting yourself?

Yeah.

Because is it $150 million or nothing?

Yeah. The way the transfer process worked on that business was pretty interesting. It actually gave you time. You signed the documents today, and it took about 60 days to clear.

How long did you have?

We had 60 days. We had, let’s say, $25 million ready to go. It was an awesome, concentrated position in a $300 million fund. We were going to start with that.

We had $4 million they said yes to. We said, “We’ll take another $25 million.” So we went out on a world tour with our thesis. Spencer and I went to different countries.

How long did you have?

We had 60 days to raise $125 million. We got to day 59, and we had sent $141 million to one of the earliest investors, a very prominent Norwegian fund. I’ll leave their name out of it. The founder of that fund was an iconic heavy-metal star in his own right—a very serious guy, a very hard guy.

We had sent him $141 million, and the offer ended. If we didn’t send him the $9 million, the deal would blow up, and I didn’t have $9 million to my name.

You borrowed the $9 million from him?

I borrowed it from him. I called him up and said, “I’m so sorry. I don’t have the $9 million, but if you lend it to me, I’ll close, and I’ll pay you back.” He said, “Sure, I’ll lend you the money.”

It wasn’t exactly that pleasant, Harry, but it worked. God bless my wife. She said, “Do you love this business?” I said, “It’s the most amazing business I’ve ever seen.” She said, “Well, how do we figure out how to own the $9 million?”

We figured it out, and that $9 million turned into a very different number. That was the life-changing event for me. That was Spotify buying into our business model.

We went on to become a top-10 global shareholder of Spotify. We went as far as putting up signs in their break room saying, “We’ll buy your shares,” buying all these odd-lot transactions along the way. We built a huge position, and it ended up being about a billion-dollar outcome for our LPs.

Wow. Forty percent of your third fund went into Spotify?

Yes.

How much was that of the third fund?

The third fund was $380 million, so it was a lot of money.

Wow. Very concentrated risk. What price did you get in at, just give or take?

In many ways, we did it at the perfect time because Adele announced she was leaving the platform. Apple Music was really taking market share, and Taylor Swift said, “I want my library off.”

Daniel Ek came out and did, I think, one of the smartest things he’s ever done. He agreed with the artists. He said, “I have your back. I understand why you pulled from our product. You’re not the issue.” The record labels had an option to buy more shares, and they executed that option.

To me, that was an aha moment where we could triple down. Nobody else knew this information. You buy from all the people who were scared to death, do the counterintuitive thing, and take the bet. We did, and it worked.

We bought at about a 50% discount to the current financing round, and we just kept buying there for the next 2 years.

After the $150 million, you kept buying?

Yes, we kept buying. We pulled a bunch of money together. Some of it we were able to put into our fund, and others we had very prominent logos come into our SPVs to buy the stock because we had locked in the access.

To me, that became this journey. You start out just hustling, you get to $35 million, and the second fund was $36 million, raised in a year.

A $1 million increase.

A $1 million increase, yes. Fewer investors, thank God. Then you get to a $380 million pool of capital.

In that fund, you take a lot of concentration in Spotify, but on top of the concentration we had, we did another $700 million in co-investment in Spotify.

In Spotify? $700 million?

Yes, huge. But that fund also had Lyft, Uber, SpaceX, Instacart, Impossible Foods—the list goes on and on. Spotify’s acceptance of our model opened the door.

Johan Bjurquist and Peter Gandelius are the 2 people who have probably made the most direct impact on my life because they trusted that we could do it. They put their names on the line to Daniel Ek that we could actually achieve it.

Then Spencer and I went and somehow made it happen. Poor Spencer literally got shingles. He was about 23 or 24 years old and got shingles.

Daniel is the guy I just send rude memes to most mornings. I mean, it’s an amazing business. It’s truly amazing. He’s absolutely incredible when you go through those names. What about Uber? How did that come about?

Uber came about because, in our thesis, we’ll often back 2 businesses trying to attack the same thing. Lyft in that fund became the value play. We sold it.

Did you make money?

We made a ton of money on Lyft, and we lost money on Uber.

What? How does that work?

When Uber chose to go public, I think it was a tough time. Today, Uber is valued off gross profit, which I think is a good metric for them. It’s a massive business, and I don’t think that when it went public, it was rewarded properly.

In our business model, we rarely hold post-public. Oftentimes, we sell before companies are public. As Uber went public at the price it did, on a converted basis, we were underwater. I think we lost about $0.20 on the dollar. On Lyft, we made about 3×.

How much did you lose on Uber?

I don’t know—probably $50 million.

And you made 3× on Lyft?

Yes.

Because you sold before the IPO?

We sold most of it before the IPO. We won’t talk about the names that buy, but you’d be surprised. It’s one of the issues with Silicon Valley. One of the benefits of being in Chicago is that we isolate ourselves a bit from the herd mentality.

And sometimes it's like sheep. They just jump one after the other, and they buy more stock in things that you see you should sell. It just takes 1 brand name to jump on board, and 10 others jump on board, and we just kept selling our Lyft stock to others. I get why they did it; it was a value play compared to Uber, and it went on to work for them, but we made our multiple, went home, and distributed the cash.

Do you think it's very clear when rationality leaves the room?

Yeah, that's something I've spent a lot of time in the last few years trying to figure out, because in 2020 and 2021, all rationality left the room. We, like everyone else, just kept deploying capital, but it left the room in public markets, too. The multiples were off the charts.

Yeah.

No, everybody left the room. I think, you know, you want to play the game on the field, as Bill Gurley says.

Yeah. Yeah.

I think a few things about that period—that was rationality leaving the room—that you want to protect yourself from. I think guardrails are important. A good friend of mine, and I know you had him on the show, is Mitchell Green. I'm a huge fan of what they've done. The best things in life are copied; in some ways, he has his Lead Edge 8, and we have our G Squared 8. Sorry, Mitch, we kind of borrowed it from you.

Is it the same?

It's not the same. They're similar because we have a similar style. They're similar, but I think we had to really adjust ours after 2021 because I think it had holes in it that were exacerbated.

What were the holes?

Too much qualitative analysis, too much “my gut, my feel,” too much of who else is in the room—just the soft stuff that venture capitalists and growth managers like to pride themselves on.

Our strategy has to be about the numbers. It has to be cut-and-dry, cutthroat: does it work financially or not? At our stage, unlike yours, it matters. From your seat, in a lot of ways, I envy your seat, because if you pay $50 million pre or $100 million pre, if it's a good business, you're still going to make a 5x. $10 million pre or $50 million pre, does it really matter for you? Maybe. For me, is it $3 billion or $2.5 billion? You walk that forward, and I can't make my 2.5x net in a 5-year period if I miss that. Now, I get an okay return, but I'm not rehired.

Does your mindset change around that when you see the elasticity of outcome sizes that we have today?

I completely agree with you in a normal world of the last 5 years, but when you have companies that are hitting $1 trillion, $1.5 trillion, $2 trillion, who gives a shit if it's $10 billion or $15 billion?

Yeah, I think that's in a shorter thesis. You're able to play that game, and that's one of the benefits to our thesis. Yes, as long as you don't get caught at the end of that cycle, which happened in 2021. Our earlier vintages were selling into it like it was happy days, right?

Did you just offload?

Oh, just massive amounts. I mean, we made ridiculous sums of money for LPs in that period, which set our business up to be able to survive a bad vintage. We're working our tails off—our asses off—on our 2020 vintage, but it's hard when you're shipping a ton of money back.

LPs happy.

Yeah, I think LPs are happy. It's also their expectation. Looking for a pat on the back because you did your job is something I talk to our team a lot about. It's a big win, but, hey, we also lost $400 million of their money over here. We made them money, but we look at these losses like, how can we lose that kind of money? I think keeping our team focused on that and reducing the amount of mistakes is important, because in a short fund life, liquidity is already hard. It's really hard as a fund manager, as you know.

You're a 5-to-7-year fund life, aren't you?

Yeah. With some levers, we can get it a bit longer, but for the first time, we've had to pull one of those levers, and it really guts me.

Is that a mistake on your behalf, or is that a changing landscape?

Yeah, I think it's—you are what you become, so to speak. Are there businesses that I would love to take a 10-year horizon on because I think they're amazing companies? Yes, 100%. But that's not what our LPs have hired us to do. They have hired us specifically for this North Star DPI statistic that I've pitched them on, and our team has pitched them on, for 15 straight years across 7 funds.

They hire you to make the most money for them, and you can go back to them and say, “Listen, the job of a manager is to change with moving markets, and I believe that, actually, we will make more money for you with a longer hold period because of X, Y, and Z reasons.”

Yeah, I think—and you now have the data. I'm so sorry to interrupt you. You have the data to go look at Palantir. Look at Palantir. Oh my God, why? So, Larry, even if I had held Palantir for 3 more years, who knew?

How much money would you have made if you held it?

God, we would be having this podcast in my bubble in outer space.

But is it like 2x more?

Oh my God. No. I had an LP, a longtime LP from our first vintage, who has been back every time. I was talking to him the other day about the difficulties of our 2020 vintage, the lessons learned, how I'm more energized today than ever before, and how we're going to make it up in spades. He said to me, “You know what, Larry? You get a free pass on that one.”

Because I distributed shares to some of our early funds. We distributed shares; we no longer do that. We just send cash. He's like, “I held that Palantir stock, and I've been investing in every one of your vintages from the proceeds of the first $50,000 I gave you.” So it just gives you a perspective. I sold it at, I think, a 3x, at about $9 a share.

What's it now?

Oh, I don't know. I try not to look at it, because it's one of those things where you don't want to inflict the self-inflicted wound of the pain. I think it's $80 or $90 a share.

My question to you is, have your guardrails now become a constraint—a negative constraint?

Look, I think if the answer to that is that we're doing what we're hired to do, and that is to generate the velocity of the capital for our subset of LPs, that's not for everybody. Our LPs are looking for an ability to play the fastest-growing, most dynamic technology companies in the world, and they want to get in and out with optionality of their capital to give more to firms like yours in every 5-year period. If they back-to-back funds of ours, they make a 4x cash-on-cash return in 10 years, but they have the optionality of it. If you have confidence in the mousetrap you've built, a 4x in 10 years cash-on-cash return is pretty hard to beat.

That's phenomenal. Oh my God, dude. We all count these big numbers. Look at the real numbers. I've seen them. You've seen them. They're very different in reality.

Yeah. I mean, when your North Star is a DPI figure, there's no hiding.

Listen, let's go to 2020. You said there, “Hey, you get a hall pass.” I look back on mine. I said to you after our war, it's really special to actually have the conversations that we do, and I really meant it. The candidness is important. When you look back now, what do you think your sins were?

I've talked a lot about this openly to our LPs, because I take pride in going around the world—we have LPs in 60 countries and 6 continents. It was smart, in my opinion, when I started the firm. It was fun. It's difficult now.

The main mistake we made was believing our own shit. Coming up to that point, we had some of the best returns in our industry. We were raising money. We just had to put the flag out there. It was hard. We'd go around, raise money, co-investment, raise money. Life changed dramatically for my partners, myself, and our team.

Just to ship back a 2018 vintage fund by 2020, we had the DPI to almost 1. Airbnb goes public, we made a 3x. Coursera goes public, we made a 3x. We sold our SpaceX stock privately and made a big return. Impossible Foods—we made a killing.

How did you make money on Impossible Foods?

Because of the herd mentality. I think it's all a product of the time you're operating in. You go back to that period, and Impossible Foods was a big deal. People made a ton of money on that. Beyond Meat, and what was the other one, the egg company? One of my best friends made like $10 million on a Beyond Meat SPV.

And I'm like, look at the market cap of the company today.

Yeah. No, I think that's the thing. I look at the big wins where the market cap goes in a bad way. Look, we were the largest shareholder of Coursera through our strategy. We owned 16% of Coursera when it went public. We sold the stock at $36 a share, and it's at $8. That's $800 million back to LPs on Coursera.

$800 million back to LPs on Coursera.

So that's part of the strategy, where you get the concentration and the law of large numbers actually works, and it's easier to manage a fund on liquidity. But getting to 2020, the mistakes we made: you had all these exits. Toast was our largest position in that fund. It's trading at $76 a share. Mitchell Green and I did Toast together, man.

You're popping champagne around the office. You're like, “We're so smart. We're the smartest people in the room.” And then I'm on a ski lift in Montana, and I'm looking at Yahoo Finance. It was before Perplexity. I'm like, “Wait a second. Why is Toast $76 a share? What the hell? Wait a second.”

“How is it $76 a share?” It’s a great business, but $76 a share. I mean, whoa. At the time—now, Toast is an amazing company, right? It revolutionized the restaurant space.

And, by the way, it saved tens of thousands, if not hundreds of thousands, of jobs in the U.S. because of the way they pivoted during COVID to help restaurateurs. An amazing business, great founders. All that being said, it wasn’t worth it as a public company at that price.

So, to me, that was kind of the canary in the coal mine. And I started freaking out because we had just deployed $900 million between the beginning of COVID and 2021, to your point, jumping on board with all these primaries and secondaries. “We’re super smart. We can do some primaries, too.”

So, what was the realization when you saw Toast?

I have a problem in this vintage. We’ve overpaid for all of it. All of it. I went to the team, and a lot of them aren’t with us anymore because we really went through an identity crisis and said, “Listen, I had gone through a period where I wanted to believe that our firm was more than just the co-PMs, Spencer and myself, and that we could do a traditional model.”

Because we don’t do that. We deploy the capital between the 2 of us, and we have a research team that supports us. It’s more like a hedge fund: quick decisions.

At that period of time, you’re like, “Hey, we’ve made all this money. We’re growing our business. LPs are coming aboard with big checks. They want to chase MOIC. They want to chase TVPI, right? We have to have a thesis that’s a bit longer—7 years, not 5. So, let’s divide the capital up amongst a bunch of people. Let’s build the traditional fund model because we’re so smart.”

And you know what? We’re still conservative. So, we’re going to pay 12x last-12-month ARR to enterprise value on SaaS. Public markets are trading at 25. Pat yourself on the back, guys. We’re super smart. Oh, the floor can’t be lower than 10. That’s a historic multiple. Wake up in 2025. The multiple is 4.

So, when that happened—when a really awesome business that we made a ton of money on by selling at the right time and harvesting, Toast, went on to create a tremendous amount of shareholder value as a public company—don’t get me wrong, it was trading at this crazy multiple, I went back to our LPs and said, “Listen, we messed up. We need to pivot, and we need another $300 million because I need to protect this thing.”

And so, we went in and did structured equity deal after structured equity deal. As that market’s falling and the house is on fire, we’re running in the front door with cash, and we’re going and doing minimum IRR deals. We’re partnering with the super-savvy investors—the Lightspeeds, the Dragoneers, and the DSTs of the world. And we’re doing deals that nobody knows about: Third Point Co. And we’re putting all kinds of structure in where these founders believed their own shit, and they said, “I’ll take the high price top line, but I’ll embed structure in the equity.”

That vintage, of all of our vintages, has 70% primary, and 40% of that has structure. For people that don’t know, what does that mean? It means you embed IRR hurdles and multiples in the paper.

So, regardless if you say the business is worth $8 billion on paper, the company has to generate you a 25% IRR or a 2.5x, whichever is greater. So, every day the return’s ticking, and you walk it out. Now, 5 years later, the entire pref stack is totally fucked because the last money in has this ratchet that’s just eating up all the value.

And that pivot—and the willingness, I think, of our team, not just me, the willingness for Spencer and me to get together and say, “What were we thinking? We have to try to save this thing.” Our LPs trusted us with $1 billion at that point. $1.5 billion.

But you’re going on the offensive with cash in the door when everyone’s running out. In terms of the actual saving, what are you doing with the overly inflated asset prices that you have?

Selling. Selling. Cutting your losses. Selling, taking a different perspective on multiples and saying, “Okay, you have to get the price to 10x.” So, you’re going back into assets you overpaid for, and you’re using the secondary market as the market’s falling to lower your cost basis.

So, you did a lot of primaries?

Yes, because we didn’t like the secondary multiple because it traded at a premium during that period. So, if you paid 25x in a public company to enterprise value to last-12-month ARR in a SaaS company, privately they traded at 50.

And so, you’re like, “Shit, the secondary market is overly—” The secondary market doesn’t work. My business model’s broken. And so, you go to primary. You go to primary. You go earlier. You do small-cap.

In hindsight, the mistake I made—besides thinking we’re smarter than we are—was not having the levers to pull structurally in our business. Because leading up to then, it was like whatever we did worked, and you could get in and out in 5 years and make a 2x net. You’re a hero.

You just keep raising money, and it keeps getting bigger, and your wealth generation keeps getting bigger. So, we’re unstoppable. We’re Tiger Woods in 2001. You start to create this persona that everything works, and pivoting—it was a lot of soul-searching.

We had a large meeting, and—

Was it difficult for you as a leader?

It was. Yes, it was difficult because I couldn’t believe that I talked myself into it. I had a Silicon Valley coach here. You’ve got me, a person who grew up in a trailer in Myton, Utah, an athlete, and I have to have a coach to tell me how great I am.

I bought into the whole hook, line, and sinker bullshit of the entire problem, in my opinion, of Silicon Valley, and so did our firm.

What is that problem?

I think it’s losing focus on what you’re hired to do and paying more attention to the lifestyle that comes along with being a money manager, the circles that you roll in, the deals that you’re doing, and the money that you’re managing versus, at the end of the day, the actual value you’re creating for the people who have trusted you with their capital and making a difference for your underlying companies.

That’s what’s important. Not going to a dinner party and saying, “I just invested in Wiz or Anthropic or OpenAI or Airbnb.” That’s the culture, in my opinion, in Silicon Valley, and it’s people deploying capital that don’t have the responsibility to sit in front of the LPs to explain the problems.

It’s a herd mentality that rushes in, and LPs follow. By the way, LPs follow at the wrong time. It’s very easy to raise money in a crazy market like you and I both did, and many other managers did, when it’s the wrong time, and it’s very hard. Our 2022 vintage was very difficult to raise.

How big was that fund?

$1.2 billion.

$1.2 billion. Was that the first billion-dollar fund?

No, 2020 was $1.4 billion.

So, you went down in fund size?

Well, the 2020 vintage, we topped that up. Remember, I was saying we went back and said, “Hey, we’ve made all these mistakes. We have to protect this. We need more money.” And it went from $1.2 billion to $1.5 billion.

The protection there was because pay-to-play was coming in. It was because we saw that we had overpaid on the first tranche of capital, so I wanted more to combat that with secondaries and also structure.

So, you hadn’t lost faith in the underlying assets?

Some of them have gone on to monumental failures that have been well written about. But most are really fundamentally good underlying companies.

So, when you review that period of 2020, what do you wish that you had done? Or if you could replay the tape, go back to the first half or quarter, or whatever you Americans like to call your timings in sports games—

Innings. That’s cricket, my friend.

No, that’s baseball.

Oh, right. Sorry.

It’s okay. Both, I guess.

Both. There we go. Both.

What would you have done now if you could replay that tape?

If I could go and talk to myself then, I would say, “Sit on your hands. You did not need to be part of that ridiculous $700 million priced round for a $10 million ARR SaaS company.”

I think that’s the easy answer: wait. I think if I could do it all over again, I would keep control tighter to Spencer and myself on deployment. And that way, as we went forward, the only excuse we could have is that he and I made the decisions, and we wouldn’t have any excuses.

It would be clear that we made the mistakes versus, “Hey, we tried to do the Silicon Valley mentality and hand out the capital and build teams underneath.” And it’s, yeah, it’s the G Squared logos that we invested in, but the attribution actually goes to X, Y, and Z.

You don’t like attribution?

No.

Why don’t you like attribution?

I think that it causes unintended consequences.

In what way?

Because people want to pat themselves on the back for the wins and deflate the losses and deflect the losses. Excuse me. I think if you want to build a firm that stands the test of time, the logo makes the investment.

It withstands any transition period and any leadership role in the firm because the logo made it, not the individual. And that’s what we’re trying to build at G Squared.

When I look out 20 years from now and say, “Where do I want it to be?” I want, hopefully, some grandkids to think about the logo that was built and say, “My grandpop built that, and it’s still there.” It’s like the Jim Simons 40% annualized IRR for 20 years.

Dude, it’s what I always say. I always say that about my grandchildren. I don’t have children yet, so it’s really quite a step.

Can you tell—we mentioned lessons on a company basis. If you look back at one of yours that was a big loss, what was your subsequent lesson from that?

Well, we’ve had a lot in that concentrated equity portfolio.

You’re going to have some monumental misses. Hundreds of millions of dollars you’re going to light on fire if you have the fortitude to stick to it. We’ve got a lot of those lessons to draw on.

What’s the one that’s most painful?

There are a couple, I would think. One is Theranos, and the other is 23andMe.

How did Theranos happen?

The Theranos secondary—we actually signed up to do a secondary at a really good price and met with management. Something didn’t feel right. I don’t know what it was. I can’t say that I was like, “Oh, it’s a gigantic fraud.” I didn’t know what it was with Elizabeth Holmes, but something just didn’t feel right.

We went back, and we had that form I told you about, binding us to transact. It was one page—simple. Companies love it. The only things that change on it are the name, shares, company, dollar amount, and binding transaction.

I ripped it up, and I got sued to finish the transaction. I agreed to settle, where I would pay a fraction of the transaction cost, but I did it personally so our LPs didn’t have to take the loss. I think what that taught me was that before we sign that paper, we better be certain, because it cost me a few million dollars at a time when I didn’t have a lot of money. I did it so that our LPs didn’t take the loss, because that was one of those spider-sense moments.

My wife would say, if she were sitting here, that she told me it couldn’t work because she’s an epidemiologist. They collect blood and saliva, and you can’t get that much data from a sample. She was like, “That doesn’t work.”

Long story short, the spider sense came up while we were going through negotiations to get out of the contract. Then some of the material started to come out, and we were like, “Hey, we’re not doing it.” I don’t think it was formally sued; it was threatened with a lawsuit. I agreed to settle it out of court and gave them a couple of million dollars to go away.

Obviously, it went on to be one of the more notable frauds of the time. That was painful for me. That was bad process and bad outcome.

Sometimes you can have bad process and good outcome.

Sure. But what that taught me was that you just need to have a good process. If the outcome is bad, okay, that’s fine. Which is 23andMe, which we’ll get to.

This was bad process going in. I was trying to jump on board, play the elasticity in pricing, and follow the herd.

Respectfully, the herd there was not professional.

No, it wasn’t.

Did that not worry you?

To me, I was chasing the discount that was available and the thought that, if it just traded at what the last round was, I made a lot of money.

That’s on the secondary side. That’s what you’re chasing, right? You want to get that discounted value, and you can get it in bulk.

I should have seen a lot of warning signs. That was a tough one to do due diligence on. I would say that added a lot of our checkboxes that we now have to make sure we don’t make those mistakes again. We really averted a massive issue in our business by doing that.

Massive. What was the check size going to be?

The agreement was to buy about $50 million of stock over roughly 4 months, because I was still raising the fund. Anyway, we got away. That’s one of the biggest wins. I know, personally, it sucks.

I agree, but actually, for the damage that it would have done to the brand—

Oh, it was catastrophic to us.

Yeah. Actually, a mega win. A couple of takeaways, though. One, always listen to your wife if she’s around. Hopefully she doesn’t listen to this, although I’m sure she will. Don’t say that out loud, because she’ll say, “See, you’re right.” She will, trust me. The clips go viral.

The second is: when you feel icky in your gut. My one spider sense, dude—I’ve had this, and I’m not going to name the company. It’s still going. We did a $5 million investment, and I called up the lead investor, who’s a very, very pedigreed firm. I just thought, “This is off.”

But you still did it.

I did. Now I don’t fucking do it. I’ll pull out of everything. If I feel the ick, I feel the ick.

You go through this phase of building a business where you feel like you have some allegiances to the people who helped you get there, right? You’re too far in to say no, even though you know you shouldn’t do it.

Oh, yeah. Now I’m like, “Fuck you.” Too far in is the first mistake on you. The second mistake is on me.

That I totally agree with.

The one that’s probably the largest from a financial perspective, and that hurt—there are 2, really: 23andMe and Getir.

23andMe not selling—that was my mistake.

When did you come in?

My first investment in 23andMe was in, I think, 2017. I really bought into the business model on the consumer side. I loved it, really enjoyed the interaction with Anne, and really bought into what she was trying to build. She embraced our model of being helpful.

I was really taken aback by what she was doing and believed in it. We built a large position in our 2018 vintage fund. I think it was the second-largest position in that fund. Including co-investment, we may have had $50 million in it, and LP capital maybe $30 million or $40 million in it.

During the SPAC craze, you pair up with Sir Richard Branson—what an iconic pair. You’ve got Anne Wojcicki, who’s a fantastic founder, in my opinion. Maybe not a great public CEO, but an awesome founder and visionary of what she wanted to build, with the ability to go do it. Amazing. Richard Branson on top of it. Awesome. Let’s go. Put gasoline on it.

It trades at $10 and goes up. We can start selling at seven, which would have been about a 2x. I think I sold the last share at 70s. That’s me chasing multiple, and that’s a problem in our strategy.

You need guardrails to protect against Larry chasing multiple, which is what a layered approach to selling does.

Yes. Dollar-cost average out, just like you dollar-cost average in. Start getting liquidity when you can privately and continue it through when they list. Liquidity is hard.

How big was your position in total?

Probably $100 million.

$100 million. And you said you could do a 2x.

Yeah.

And instead lost $70 million. Were the LPs cool about that?

No, the pool of the capital is awesome.

What was the number-one position in that fund?

Toast. Awesome business. It was about a 3x. We started in 2018. During COVID, Mitchell Green and I were talking about it, saying, “These people don’t get it. Let’s buy a bunch of secondary.”

Yeah, awesome. I’m on. Asana, too. Mitch and I teamed up on Asana. It was fantastic.

It was just like you were conquering the world, you and your buddy—buying stock, people running out the front door because their house was on fire during COVID, and us running in with capital.

It was fun.

What about Getir?

Getir is one that, emotionally, is really hard for me. When I think about that, it was good process and bad outcome.

You had good process.

We had good process on the first money. Gorillas, the investment in Gorillas, we had good process. We had made a lot of money in the food revolution. We were in Instacart, Postmates, and Meituan.

Just to be clear, because people will hear these names: you made money in Postmates.

Oh, we made a 3x. That’s a Founders Fund special, man. We did a lot of work with Founders Fund early. Fantastic group of investors there. I mean, I think Peter Thiel has more money than God now, but they were the largest investor in Postmates.

Bastian Lehmann was an awesome guy. He spoke at our 2018 LP day. We made a 3x on Postmates.

In how long?

18 months.

Wow.

Spencer Mlod, who was my co-PM, led that deal. He was like, “You’ve got to meet this founder. He loves what we’re doing, and he wants to help all his employees. Let’s do it.”

I was like, “Yeah, the data looks good, man. Let’s go.”

And you made money on Instacart?

Yeah, we made a lot of money on Instacart. We sold Instacart privately during COVID.

You’re one of the guys who sold in, like, the $20s.

Yeah. I was like, “Come on, Sequoia, up the price,” because Sequoia bought.

Yeah, Sequoia made money in the long run on that. They also bought along the way.

Sequoia’s business model is fantastic. We have a lot of investments in common, and their partners are awesome. There are few evergreen structures I’d ever invest in. Theirs is one.

I agree with that. So why did you sell at $25? Did you just want out?

Funny story. Personally, you have experiences with companies, and I had a very frustrating experience with Instacart during COVID. It went from Whole Foods-only pickers and grocery shoppers, which was an amazing service, to having your Uber driver shop for your groceries and try to get a ripe avocado and COVID from your Uber driver. It’s not a great experience.

I remember a point when I was hunkered down at my farm. The world was falling apart, and I just couldn’t get the order. It just wasn’t right. They didn’t take it back; you just kept getting stuff delivered. I was like, “It’s time. I just have to get out.”

I’m sorry. Wait a minute. So, the fact that your Uber driver can’t get you a fucking ripe avocado, you’re like, “You know what? I’m not the only one.”

Larry, come on. I’m out. You know what? We’re going to sell at 25. And you do a 3x on it.

Yeah. But also for us, the decision to sell becomes somewhat easy because we’re solving for something that’s different from others, and we’re solving for this velocity of capital. It’s part of keeping in the middle of our fairway, to use a sports analogy. Don’t chase the moon.

We make a 2x net in a 5-year period: cash return, not shares. We stopped distributing shares because LPs blame you if they hold them.

Don’t founders see you as a bit of a tourist? No offense, though. If you want that short duration, you’re not exactly a long holder. You’re not a long believer in the company.

No, we’re a point-in-time problem solver. We’re like the janitor who’s cleaning up the mess. They have an employee who leaves, and they’re out there in the market making noise. They don’t want them to take their shares, so we get the phone call.

You’ve got a fund that wants to show their LPs that there’s value in the extension they’re giving them. We get the phone call to buy a little bit of their position. You don’t buy 100% of somebody’s position, typically, in a company unless they’re in year 15 of their fund and have no extensions. If it’s a large position in their holdings, in their NAV, and you buy it all, you kind of scratch your head and think, “Why are they selling it to me?”

Yeah, but dude, I’ve had this before. I’ve got a great company and I want to buy some, and then ROFRs happen. How do you get around a ROFR?

That’s where the relationships come in. Companies buying into our business model, as well as your peers, have to see value in what you can bring to the company, because what we’re after is large, concentrated positions.

We’re not indexing secondaries, buying with no information, playing the arbitrage in pricing, and putting 200 secondary positions in a fund. That’s not what we do. We’re going to have 10 companies make up 90% of our risk, and we’re going to become very large shareholders of those companies by doing our business model.

It requires your peers to say, “Okay, they do something unique. They’re going to provide value to us by consolidating the cap table. They’re going to help us with employee-structured tenders and shareholder tenders. We have an employee, former shareholder, or current shareholder who needs to leave; they’re going to buy it.”

It’s really the barbell of the small microtransactions that nobody cares about. Some of our companies we’ve built through 50 transactions to get to $75 million, and some we’ve done through 4 transactions to get to $200 million. It’s just a different muscle memory.

You have to sell that. You’re going back to raising money as a new manager: be different. Be different, because different is a pain in the ass to do 50 transactions. The muscle memory of that is not replicable for others.

And the back office requires investment. It’s not just about paying yourself and hiring big teams to run it. It requires you to know that the actual outcome is the benefit of the carry, not the management fee.

No, I totally get you there. Going back to Getir here—yeah, that one took years off my life. Good process. Look, how did you come to Getir?

We made those investments and made a lot of money for LPs and got introduced to Getir. We have an office in Zurich and a team of about 10 in Europe. We’ve been here since 2017.

We said, “Hey, you’ve got to meet this company in Berlin.” When I was there visiting, we had a bunch of portfolio companies in Berlin at the time. We met the founders, and the business was growing like a weed.

This was when it was in Turkey only?

No, this was Gorillas in Berlin. Gorillas was acquired by Getir.

Yeah, that wasn’t a good deal, was it?

For a group of shareholders that we were part of helping structure alongside 5 others, it actually was a good deal. We got good value for our money. Four or 5 shareholders of Gorillas received the majority of the preferred equity that we received from Getir.

Did you get cash out there?

We did get a little bit of cash. We got most of the preferred value in the transaction.

So, you were in a good place then, with Getir valued at $10 billion?

The next 3 years were some of the worst of my life.

Why?

Seeing that we needed to try to protect that money with more money, and then get involved at the board level and do the heavy lift of restructuring and battling with the founders. We’ll leave a lot of it out because it’s still ongoing, but it was really hard.

The mistake I made was the second tranche of capital.

How big was the first? How big was the second?

The total first was $50 million in total, between secondary and primary. We got that plus some in value from Getir, and then the next check was $100 million to restructure that equity and pull it forward.

So, you got $200 million at risk in total, with a lot of LPs in it through co-investment. Of course, everybody during that period signed up for it. You just put it out there.

Dude, the co-investment craze was nuts. It was nuts in 2021 and created a lot of unintended consequences that we can chat about. It was different from our normal co-investment model, which we went back to after that period.

Anyway, the long story short is that I think the major mistake I made there was not being willing to just walk away after the first check.

When you grow up in severe poverty, you’re fighting with your siblings for everything. You’re fighting at school. You’re fighting for food. You’re fighting, fighting, fighting. As you grow into business, that doesn’t leave you. It subsides a little bit, but there are many people in our industry who have backgrounds like mine, and you don’t stop fighting.

It becomes a blessing or a curse. It starts to become a curse if you can’t manage it, because in that scenario, I thought I could fight and will the outcome for little old G Squared amongst giants. It was like you were a dead man walking without knowing it. You were already dead; you just kept fighting and fighting.

Making sure we don’t make those mistakes again is important. That’s where guardrails and other people with voices in your firm come in.

Was it obvious that it was going wrong? It seemed like an enormous amount of cash going into a business before it was ready, with premature scaling. You could see that the economics didn’t work. It seemed very obvious to me, respectfully.

Hindsight makes it easy to say it was obvious. I’m sure to many people looking at me and watching from afar, it was obvious for me.

You still hold it, though?

We still hold it. It’s now a business focused in Turkey, with lots of great people working in the business. I’m still on the board, so I’ll be careful about some of the things I say because it’s a major, major part of Turkey.

Oh, yeah. By the way, that was the biggest mistake for me in the business. It should have just continued with deeper penetration.

Well, yes. Investors would say one thing, and founders would say another, about how it became a global strategy. I think the mistake the company made was that it left Turkey and didn’t focus on other markets that were similar to Turkey from an employment-cost perspective.

But look, today it’s still operating in Turkey. It’s a big business. You’ve got a lot of sophisticated people from Mubadala in there working with the business. You’ve got people inside the business who want to win. You’ve got more than 10,000 employees working hard every day, trying to create value for their families.

As an investor, I lost a lot of money. Why I’m still involved is because those people who are commuting, in some cases 3 hours a day, to work in Turkey deserve a good outcome.

That’s the piece of our investing that, as a secondary investor, you often don’t get that granularly involved with. We do it with a handful of companies, unfortunately typically when they have issues that you’re trying to solve. But it was a very hard journey.

How much money do you think you lost, though?

There are the numbers you can quantify, and then there are probably the dollars that you lost from a fundraising perspective for your next vintages from LPs because you lose their co-investments.

When you lose co-investment dollars, do you lose them in the next round?

It depends on how you structure your co-investments. In that case, from that fund, my guess is those decisions we made cost our partnership probably $500 million of capital, between capital actually lost and investors that you can’t bring back. It was a monumental miss on our part.

Now, I’ve lost money before in co-investment, and the only lesson that I’ve learned from that is to use it as a chance to build trust. In a crisis, communicate more than you ever have before. Try to overcommunicate.

I think a lot of LPs, because of that, have come back. Does it make you question your co-investment strategy? Because the thing I don’t like with co-investment is that LPs are often almost not aware that it’s a single shot.

You have a portfolio in a direct venture fund. Fine, you have a Getir. Fine, we’ve got a Toast as well. But they’re almost shocked: “I didn’t realize that was done.”

I think that, going through our co-investment process in the early vintages—let’s say Vintages 1 through 4—in some cases we were 4 times co-investment to fund investment.

Yeah, but here's why. The kind of companies we were interested in investing in, we weren't large enough to make a difference. If you want to run our business model, you need capital—a lot of capital—just to get in the door, just to be able to solve the problems that they have, to your point of being relevant to the ROFR. If all you can do is onesies and twosies and you can't write a $100 million check when they need you to do it on the secondary, they go find somebody else.

When they turn that faucet on of shares, when they buy into your strategy, when Daniel Ek says, “Hey, you guys are interesting to us. Oh, by the way, here's $150 million to put into the stock,” if I say no and it's at a good price and he picks up the phone and calls somebody else, it's gone. You're not getting that opportunity back. With our strategy, you have to have scale.

The co-investment in the early days was a significant part of our strategy. Over time, it's become more normalized. Today, it's about—I don't know—on a $1.5 billion fund, it'll be $700 million. It's still a lot.

The mistake we made in the 2021 vintage—or 2020 vintage—is that we didn't just do co-investments around our core positions. We had LPs coming to us saying, “Hey, I built my fortune in, let's say, wellness. I like health tech companies. You've got one down here in the fund. I want to buy stock directly.” At the time, with the benefit of hindsight, it was wrong. We felt that we were enabling our business model to help them achieve what they wanted by being an LP of ours.

Okay, but let me just push back on you. We're partners now. They have directly requested that asset. I share with them my concerns and how we believe the company is doing. If I do that transparently and offer them access, am I not providing them a service that they're asking for?

Yeah, until it doesn't work and then they blame you.

So you should block them from doing it?

I wouldn't. The way we run it today is the way I think people should run theirs: only in conviction, when your funds are investing alongside it. That's when you do the co-investment. You don't do it the way we did it in 2020. You do it the way we do it today.

Our 2022 vintage fund was $1.2 billion in size. The co-investments we ran are our top 10 positions. You've got a couple of LLMs—the high flyers. You have companies like Fanatics, Wiz, Databricks, Turo, the Airbnb of cars, and Monzo—companies like that, where you've already decided those are your core positions, and then they make a bespoke portfolio of additional exposure to them in equal-weighted size.

Why are you in Monzo? Is it a value play to Revolut? And Tide is fantastic.

I think Revolut is generational. We were in Revolut in our 2018 vintage fund.

I agree with you. Amazing business. Did you hold it?

Yeah. Some of the stuff we can't really talk about—whether we're holding it or not—in the live positions.

For me, you don't need to. I just look at it now and I'm like, “Fuck me. This is a $500—”

We have it in our 2018 vintage fund. It's made an awesome return for that vintage today.

How did you get Revolut in 2018, dude? Sorry, I'm about to say something so offensive. You're like a Mid-American. I love the bias, and I'm sitting in fucking London and I don't see it in 2018.

I was fucking hustling, Harry. Hustle.

I was 12 then. Fuck you. But seriously, how did you see it then?

Four mega-trends for us. Fintech is one, and we had good success coming off SoFi. The founders of SoFi and the current management really believed in our strategy, and SoFi was also in our 2018 vintage fund. That fund, man, it's just crazy what was in it.

Off the back of that win, we had early N26 before the growth restrictions.

Did you sell?

No, no, we didn't sell. They're working their way through some things and looking fairly okay right now. They're doing a really good job of managing through the stress they had in the business and are now more focused on Germany versus a continental play.

Then we had Revolut. What a great little pool of companies, right? What a great pool.

And so you were like, “But Revolut went like a hockey stick, man.”

In our strategy, you need time. You're like, “I see Revolut. I'm always going to go into Monzo.”

Well, Monzo is a later vintage. Revolut's in our 2018 vintage, and we have them in every vintage. In our current vintage, we have Chime. Awesome business, but it went so fast that you can't build it over time, dollar-cost averaging in through lots of transactions. You can't build a scalable position if the company goes from 0 to 100 in no time.

Wiz was very short for us. In our current vintage, we did it in 3 years, but that was only because it started off the back of the downturn. We caught it at the right moment, right? In a normalized scenario, we would never have been able to build this position in Wiz because it was a hockey stick from day 1. But the capital availability, even to the best assets in the world, wasn't great for 18 months.

What price did you get into Monzo?

Sub-$4 billion. For us, again, we're not solving for what you're solving for.

Tide is great. I really—

Yeah, Tide has a great operator. Great guy.

I think what you have to do is build a portfolio across the 4 mega-trends. For us, the way we look at it is, you have to diversify that risk. You have SaaS, fintech, consumer internet, and mobility. Across those, we try to build conviction positions, pretty equally weighted across those mega-trends, among 10 companies.

For every Anthropic you have, you have to balance that with a fantastic business in Fanatics, which is more of a traditional late-stage growth business. The business is fantastic, and it's at scale—massive scale, billions of dollars of revenue and hundreds of millions of dollars of EBITDA. It's a massive company, a sports-apparel merchandising business significantly owned by Michael Rubin.

You balance that risk because what we've learned over time is you can't just play the momentum, because you can get burned, and that's chasing multiple.

I totally get that. You can't just play the momentum. How the fuck do you play AI, then? Because it is—

Go to the leaders. That's easy. Go to the winners.

What's holding back the next generation of the LLMs from getting to Anthropic and OpenAI scale in 3 years? Who's going to rival them?

No one. You just answered your own question, Harry.

No, I completely agree with you. No, I literally—if, for what I'm about to say, may my LPs forgive me, I would put my whole fund into OpenAI.

$350 billion.

Yeah. You and Spencer Mlot should start a fund together.

He's amazing. He's my co-PM. He believes the same thing.

At $350 billion, do I see this being a $1.5 trillion company in 5 years' time? Easily, I do. That is a 5x with fair confidence, very large confidence, in a 5-year period.

Yeah, it's an amazing business.

Yeah, I would do that 100%. It's reached escape velocity. I speak to Kevin Scott at Microsoft, and I'm like, “But, dude, there's no defensibility on the search.” And he's like, “You can't just switch from Google to Bing.” Of course you can, but it's the brand.

Poor Bing always gets picked on.

Bing—come on. Everybody picks on Bing. Everyone, even he does. He's the CTO. You don't need to pick on Bing.

It was him, not me.

But the point was, I was like, “Wow.” Yeah, that's absolutely true. And so I agree.

But, okay, I have friends who are in Anthropic and got in at $4 billion.

You didn't need to comment.

Yeah, and it's now $60 billion.

$61 billion.

And they've got like a 4x on their money. So there's a question of, okay, value accrual.

Have them call me. I'll buy all their shares of Anthropic at $61 billion.

You would buy it all day long today and twice on Sunday. Are you serious?

100%.

But even though the dilution is so intense?

See, now you're talking like an early-stage investor, Harry. I don't give a fuck about the dilution. I care about the price I pay in dollars and the price I'm going to sell it at in dollars. I am focused on DPI, not MOIC. I don't care about the dilution.

So are you not just going around now hoovering up Anthropic?

Oh, 100%.

What's that?

A lot. A lot. And I've got appetite for more. Same with OpenAI, Databricks, and Wiz. I love those guys. They're awesome founders.

If you ever talk to them—we hosted them at our LP day—half the things they say, I don't even understand what they're talking about. It scares the fuck out of me, the world after talking to the 3 of them and seeing what's out there. But, man, did they build something. I mean, how they did that.

By the way, they also sold a business to Microsoft together and made a bunch of money, and then said, “Hey, yeah, we're well-heeled.” They're humble people. They've made a lot of money, and some of them live in a 1-bedroom apartment in Tel Aviv.

I went to see them in Tel Aviv. I went to see a bunch of companies—we'll leave the other ones out. A couple wouldn't even see me, and this is after I managed $5+ million. The Wiz guys welcomed me in, gave me a sweatshirt. “How you doing?” I'm like, “Oh, by the way, I want to invest in your company. You're raising like half a billion dollars.”

“I want to give you $3 million.”

“Three million bucks?” They’re like, “You know, that’s a little small, Larry, but we like you. Will you settle on $9 million?” “Sure, I’ll give you $9 million.” That’s how we started. We then went on to build a $200 million position, but it takes that buy-in.

You go back to Anthropic and the others. We were fortunate enough, during—you mentioned FTX earlier—to be awarded a bunch of stock during the bankruptcy. We bid on it, went through the whole process, bought a big chunk, and now the company—in my opinion, Anthropic is unstoppable. That and OpenAI are just the beginning.

So how do you play the space? You go to the winners because I have a shorter thesis. How many winners are there?

I think the LLM space is challenging for new entrants because it takes time and capital. I think it’s OpenAI and Anthropic. I don’t see a ton of room for more.

Not because I don’t think so? Are you going to play the application layer too? Are you going to try to do Cursor and Abridge in medical?

No. For us, our view is to take a few concentrated bets in the foundation models and the winners, and then play a bit of the picks and shovels. That’s why we have Lambda. We also had CoreWeave.

Then, on the business models themselves, let’s play Scale AI and software to help them scale on the hyperscaler side, and focus the balance of our capital on our other megatrends, where AI has been embedded into all businesses. You have exposure to it throughout cybersecurity and SaaS. You have exposure to it in fintech. You have exposure to it in consumer companies today that are created without AI embedded in their DNA. Why would you do it if you’re not going to put it in there?

The challenge you have in our industry is that you have vintages of managers who invested basically for the last 20 years, many of them without liquidity, and most of their businesses don’t have AI embedded in the DNA of the company. They’re chasing now.

You have incumbents that are large—we call them, internally, two types of companies that are big and private but aren’t in AI. We call them vampires and zombies because they’re massive, and many of them are profitable. They now have to implement AI in their business models, and some are going to get out of it and be okay. Those are the vampires. The zombies are dead. There are hundreds of them. I think there are far more zombies than vampires.

I think it’s a much more difficult transition to implement AI. A lot of product change is actually bottoms-up, and you can’t just slap AI on it. It’s hard. Companies are going through that.

I really want to do a midlife crisis and name it PFU, which is “private fuck-you.” I just want to go down Anaplan, Coupa, you name it, and take them out one by one. You’ve got to have a couple of things: the founder’s not there, the engineering team is cut, customer service is cut, and there are price increases. Then, by the way, layer some secondary on, lower the cost basis, and you can make a 5x in 3 years.

Yeah. No, it’s a great idea. I think there are long-in-the-tooth companies from 2015 vintages.

So is my generation, and the generation of LPs funding us, morons? I mean that in the nicest way, and I’m being deliberately glib, but the liquidity profiles that we’re doing—it’s 15 years.

Yeah, I think it’s hard, Harry. Liquidity is hard. The one thing—and you asked me this question on our walk—is, what do LPs think they know that they actually don’t?

I would say how hard liquidity actually is for all of us. It’s really hard, and that’s why the DPI numbers are so low. It’s not because people don’t want to make money for people. It’s because it’s hard. It’s just as hard as getting into the best deals. It may be harder to get out, and the timing that people are doing it on makes it even more difficult.

Is that even the case in the high-demand assets we have today, like the LLMs? You want to sell your Anthropic?

Yeah, today. I think there are a handful of liquid private companies. Basically, they’re quasi-liquid. This is the hard thing, but at scale, small positions, yes, because I look at—

But try to move $1 billion of Anthropic.

Agreed completely.

But I look at my first fund, which was tiny, admittedly, and it was large compared to mine. All of my winners—I can sell tomorrow. I don’t want to sell any of them.

Yeah.

But all of my dogs, I’m desperate to sell.

Yeah, that’s the classic: sell your dogs and keep your winners. I think it’s somewhat the inverse. You have to be willing to sell your winners to have a sustainable business model and to generate the return that will actually drive your fund forward.

What has been the single best investment for you?

The single best investment that we’ve ever made on a multiples basis?

On a multiples basis.

It’s probably early Wiz, early Spotify, and Bolt, the early ride-hailing company. It’s yet to be determined with Bolt, but the first investments we made in all 3 of those are probably 10x.

Why did you do that?

Which? Bolt?

Yeah. I love Markus Villig on the show. He was fantastic.

He’s amazing.

Dude, respectfully, when you went in, this was pre-Sequoia and pre-institutional capital. I think it was like 20 bucks a share or something. It's equated at 260. And this was not the brand.

Oh, no. But we had come off, in fairness to us, a great outcome in Lyft. We didn’t know yet what Uber was going to be. I was going through all the challenges around the transition of the founder and all that.

One of the people at Spotify, Johan Bjurquist, called me and said, “What do you think of Lyft and Uber?” I gave him my opinion on both, and he said, “Well, there’s this small business out of Estonia that I’m going to go be the CFO of.”

Johan has been, to date, our single best investment we’ve ever made. We were in Spotify early—that investment that he trusted our business model on was, let’s say, 8 to 10x. Bolt is yet to be determined, but let’s just assume it’s somewhere between 50% and the face value of what Sequoia paid. It’s 6 to 7x, and now it’s a big business. It’s a really big business, and Markus is phenomenal—a young kid building that business.

The thesis was, “Well, I’ll go where Uber’s not and do it profitably because I don’t have any money.” He bootstrapped it.

Being blunt, as I said, you mentioned your humble beginnings earlier. You’ve done very well.

It’s all relative.

Well, yeah. But, yeah, go ahead. Does money make you happy?

Money doesn’t make me happy. Money makes my life easier.

Money makes your life easier.

It’s complex in some ways.

What did you think about money that now you have it, you see differently?

When I was growing up, I thought, “If I could just make $5,000 a month, I would be set.” If I could make $5,000 a month, own my own farm, and farm it, that’s all I wanted to do. Then I spent 20 years trying to get away from the farm, and now all I want to do is go back.

Money doesn’t make you happy. I think it makes your life complex. How you handle that money is important.

And you learned about handling money when it comes to you?

Yeah. Personally, for me, don’t increase my spending with my increased wealth.

Yeah, I think that’s a good goal, Harry, and you should try to maintain that. It’s very hard.

For me personally, my journey has been about not having that fear of going back to powdered eggs, powdered milk, and government cheese—that fear that it could happen again—and setting my family up so that my children won’t ever have to go through that, and generations won’t have to go through that. That’s been my desire.

Now that we’re partly there and creating that, I would say I’m probably in the middle innings of doing that for where I want my life to be. Reminding yourself constantly of where you came from is important, because it’s easy to be around people in our profession and, just in general in life, as your financial status increases. The people around you change a bit, and that pull to be more like everyone else is there.

Maintaining the thought in your mind that this hasn’t always been what your life is, and being thankful that it’s here without losing the edge, is important. I think the most difficult thing to do as you start to make real money, regardless of how you define that, is keeping the edge that got you there. You see that across sports, across business, and with high-performing CEOs.

When you think about keeping that edge, is it running away from the government cheese and abject poverty, or is it running toward the grandchild memories of G Squared?

I think keeping the edge for me isn’t about either of those things. Keeping the edge for me is the chase of the next win. No matter what the dollar amount is, the win is what gives me satisfaction, and the loss is what hurts.

I spend a lot of time talking to you about the losses because I can remember almost every moment of those. As an athlete, I remember all the failures.

Should you ever build immunity to losses?

Yeah, immunity to losses would be nice. You should do that, because maybe you’d be happier, plus the money.

Maybe you're high on life.

Yeah. I think some people can. For me, it's ingrained in my DNA that I don't want to leave. I don't want to know who that Larry is. I want to fight for everything. I enjoy the fight, the journey to win, and everything in life. I just want to win.

Final one before we do a quick fire. What does no one see about firm-building that, having built a firm, you think is so cool?

Building something that you don't realize people outside of your small sphere recognize as positive, and then being randomly faced with that—like this conversation for me—is humbling. I'm a huge fan. It's like sports radio: longtime listener, first-time caller. You're a young guy, but you're super successful, and you have some of the most amazing people in our industry on the planet sitting here and talking to you. To think that I'm sitting here with you, telling my story, is very humbling—the fact that you're even interested.

I think that piece of building a firm and staying focused on the logo, not me individually, has led to this moment. That tells me that, for all of our flaws, which we have many, there's something in the water at G Squared. We're going to keep grinding and doing it.

Dude, I'm going to do a quick fire with you. Otherwise, I could keep you all day.

Yeah.

Tell me, whose life do you secretly admire and why?

Because of my background in sports, who do I admire? I think about the longevity of Cal Ripken Jr. in baseball, who basically, for 16 straight seasons, didn't miss a Major League Baseball game. Major League Baseball plays 160 games a year. For 16 years, from the time the guy was 22 to the time he was 38, he didn't miss a start. The grind that took, and people talk about records in sports and all of the high-flying athletes that people recognize and their accomplishments.

Here's a guy who was a grinder and played third base, a hard position to play in baseball. I know you're a cricket guy, but study it up. He went 16 years without missing a game. Of course, he's in the Baseball Hall of Fame and celebrated in that way, but he was a good player—not the best. The grind that must have taken for him to get to the major leagues, not being the most talented guy, and then to play and not miss a game for 16 straight years for one team, at 38, when he retires—that, to me, is a different kind of intestinal fortitude and a set of massive steel balls that you just don't get in generations today.

What that guy accomplished is insurmountable. There will not be another baseball player, in my opinion, in the history or future of baseball, who will accomplish what he did.

It's very akin to what you're saying about listening to the show: 3 shows a week for 11 years. There you go. No one giving a shit for 4 or 5 of those years.

Yeah. It's the grind and the willingness to have faith that you can do it, living and breathing the challenge and making it become your identity.

Why do the early shows where nobody was listening?

Because I loved it.

Because you loved it. It was your passion. Still is.

And that's where people should find their work-life balance: when your work is your life, your life is balanced. I couldn't do this today if I didn't love it.

No, of course not. Why? You don't have to go far. You're in your house. You go swimming next door and walk in the park.

Oh, did I not tell you? We do a session in the sauna afterwards. We record that, too.

Yeah, that's a little weird, but I'll go.

We're European. It's fine.

Yeah, I know. I know.

What would you most like to change about G Squared that you can't?

I think, personally, our people would like me to chill out just a little bit. If I could be a little chiller and grind less, I think people at our firm would probably enjoy being there more than they do. The money's awesome for everyone. People don't leave. We only have forced attrition.

That aside, I spend a lot of time thinking about how to change my delivery at this point. We are survivable. We will have the next vintage, and we're going to have many more vintages. There's no fighting just to fight. It needs to be altered without losing the edge. I wish I could change it more quickly than I have, and I think it's unfortunately led to some really good people moving on who otherwise I would still like to be on the journey with me.

What's the most memorable LP meeting here in London?

A guy committed on the spot and signed this—said, “I'll sign the forms for $20 million.” Wow. The first meeting, he said, “I'll do it.” It had never happened to me before, and it was here in London. He's a longtime LP who's still with us today. He probably didn't even care about the strategy. I think he liked me and said, “I'll do it.” That was the most memorable for me.

What worries you today that you don't think many people are thinking about?

I think fundamentally our industry is searching for solutions to the fund-life problem. They're searching for a solution because it's fundamentally broken, and they don't want to say out loud that it's broken. So, evergreen structures, interval funds, quasi-liquid strategies in privates, and continuation funds.

Continuation funds, though, I find a bit more interesting today than they were in the early inceptions, because now they'll only buy your good assets. They've really gotten smart.

I think that's a really interesting space right now.

If I'm the CIO of a large endowment fund, what advice would you give me today about allocating in this environment?

I would say where endowments miss the boat—and, by the way, I'd love to have many of them, but we don't—is that I think they chase the wrong statistics.

What do they chase that they shouldn't, and what should they chase?

My belief is that TVPI and MOIC are not the statistics that people should be focused on, and they are. I think they're fake numbers. They should just chase DPI. It's the only thing you can use to buy food.

A penultimate one: when you think about yourself as a leader, what could you do to be better today?

I think I need to continue to grow and be more open-minded, which I have been. I have full trust now in a co-PM of mine who I think is one of the smartest people I've ever met and a fantastic investor, and he has my ear. I think I could do a better job of having more people have my ear. I could listen to more people than I do.

That's a challenge for me in perfectionist mode. It's just my nature to be productively paranoid, and I should listen to more people and the advice they give me. That's something I've worked on more extensively for the last 5 years: gathering more people around me, digesting it, and then acting versus just acting.

Final one. When has being paranoid helped you, and when does it hurt you?

I told you about Theranos. That helped me. That was productive paranoia, and so was the productive paranoia of closing and investing. The money's there: close it, invest it. The money's there: close it, invest it, because you don't know when the next one is coming.

That's a bit of just my DNA as a human being and how I was raised. It's productive paranoia: things are great today, but the wheels are going to fall off. You have to make sure every day that you tighten the wheels, and you have to be focused on the right things. I think that driving every day, having that paranoia no matter how good things are, and always worrying that it's going to get worse across all things in life is, to a degree, healthy.

Dude, I so appreciate this. I so appreciate your willingness to talk so openly. I loved it on our first walk. I really did. It was very special, which is why I was so keen to make this happen. Thank you so much for doing this with me.

Well, thanks, Harry. Congratulations on what you've built. I'm a huge fan, and I think you're only in the beginning of what you're building. I'm going to be cheering for you along the way.

Time for the sauna.

Yeah, let's do it. Sauna, here we come.

Keine Badehosen. There we go.