Hussein Kanji，Hoxton Ventures 创始人：为什么 AI 让伦敦有机会与美国竞争｜E1248

1. Should Governments Be Funding Venture?

There is a correlation between how much money goes into a company and what the probability of success is. The average is about $300 million to get to unicorn status, and your best path to scale from a financing perspective is America. The rounds are bigger. Do not do a fundraise for the size of the fund; do a fundraise for the time of the fund. Give yourself 90 days—whatever you get, go start investing.

Hussein, dude, we did this 9 years ago. It was a webcam on Skype, which is aging both of us. Thank you so much for joining me today.

Yeah, I think I remember. I had my laptop on a pillow in my bedroom, staring up at me. I was like, “Who’s this Harry kid interviewing me?” I think everyone was thinking, “Who the fuck is this Harry kid? Why wouldn’t he leave us alone?”

2. What Is Hoxton's Right to Exist?

Listen, I want to dive right in. I remember Keith Rabois telling me on a show that every fund needs a right to exist. When we think about Hoxton, how do you think about your answer for what our right to exist is?

Yeah, it’s a good question. When we first started—by the way, I think the venture world does not need yet another fund. We have a lot of them. They’re coming down in numbers, but the world had a lot of people playing VCs. Eleven, 13 years ago, when we first started—we’re 11 years old, but we started fundraising a little bit before then—the world did not have that many VCs in Europe. The US had a lot of them, China had a lot of them, and India had a lot of them, but nobody was here in Europe.

In fact, the seed funds of record here—you won’t even remember the names—were Eden and Pond. They’re bygone, right? The people who raised money in the dot-com boom mismanaged their capital all the way through the collapse and kind of left. So the world really needed a venture player in Europe, and that was the thesis of Hoxton.

3. Does Hoxton Do Outcome Scenario Planning

If you look at where we are today, the world now has quite a few venture funds in Europe, but there are not that many old-fashioned venture funds left in this industry. I think most of us have become momentum investors in this industry. We write the check largely to get the next markup, not to build the long-term, durable, big company of tomorrow. I don’t think there are that many people in Europe who do those kinds of things.

Why do you think that is? Why have we shifted to this heavy momentum?

We went into a market where money was effectively free, and the way you get promoted inside most firms—remember, we’re exceptions to the rule because we own our own firms; these are our businesses, so we think like business owners, not like employees—if you’re the general employee, you optimize for getting to the next career ladder.

How do you show that you can get to the next career ladder? You do a deal, and then General Catalyst or Index or Kleiner or Sequoia or Andreessen—I mean, there are so many of these great firms—mark it up at a significant premium, and then someone else, Tiger Global, et cetera, marks it up after that. All of a sudden, it doesn’t make a difference if you haven’t made any money: you look like you’ve picked a hot company.

4. Does Having Tier One Investors Really Matter in Fundraising?

I would actually argue, even for us and for those who own their firms, that if you have to fundraise, it makes life considerably easier. People are like, “Oh, DPI is all that matters.” It’s not true. If you can show a cohort of companies that have great tier-one investors following on, it is meaningful to LPs.

Yeah, I would say even for us, we have a challenge when it comes to LPs, and this is not us-specific—us in general—which is that the entire industry looks at what the next markup is: who’s following your deal, who’s marking it up, is the company well-capitalized for the future, and really, is it a signal of quality when a Sequoia ends up writing the check? Have you picked a really good company?

5. Spicy Questions

I think that’s true in general. The problem is the venture world is not a general-type industry, right? The averages and the medians are very deceptive in our industry. That’s not where the returns are. So, weirdly enough, we’re in this strange predicament in the industry where you kind of have to do things that are a little bit off-piste. You have to build for the big outcome, and you have to be a little bit contrarian. Then, very quickly, about a year or 2 later, the world has to recognize that you’re right in order for you to really get credit.

What does it mean to do things that are off-piste today, though? You know, vertical SaaS—you think in a world of AI and agents, vertical SaaS has never been hotter. We just did a, bluntly, very boring vertical SaaS company that had 13 term sheets.

Yeah, so I think the whole industry has massively grown, so there’s a lot of money to be made. But if you think about the big, iconic, household-name companies—the Googles, the Facebooks, the Ubers, the Netflixes—they were all mostly brand-new category creators. That category didn’t exist. There wasn’t an Uber before there was an Uber; there wasn’t a Netflix before there was a Netflix. There were search engines, but none of them really succeeded. Then Google became this thing. There was also Friendster before Meta, but it never really succeeded. These were inventing new categories.

I remember even when Facebook went public, when we were fundraising for Fund I, people were really skeptical about how it was going to make money. The transition to mobile was hugely questioned for Facebook, and Facebook hadn’t done the transition to mobile yet, right? How does Facebook really make money? It captures all your attention on the mobile phone and, as a result, has the right to be able to serve you ads. That’s their durable moat. But that wasn’t clear even all the way up to the IPO.

6. Approaching Reserves in Venture Investing

These new categories are really fuzzy up until they’re not, and then, when they’re not, you see really big outcomes. I don’t think people in Europe think in that kind of way. I think people in Europe are largely trained in private equity. They think about, “How do I minimize my downside? I will do the vertical SaaS company because I know I can’t lose money on it. The metrics are really good; I can understand them; I can characterize them with the preferred stack where it is. I only need to clear 12, 15, and there’s upside to 750. Great.”

How many investors have you had come on who think in this kind of language: “I will worry about my downside and the upside will take care of itself”? But the venture industry is all about the power law, all about the outliers, all about those kinds of outcomes that I was talking about, and Europe doesn’t have those.

Do you do outcome scenario planning? Do you think about the whole range? If this company falls into trouble, how do you think about what happens?

Yes, we think about the whole range. If this company falls into trouble—we had a strategy in the fund where, if the founder gets hit by a bus… We had an incident in one of our companies where the founder got diagnosed with bipolar disorder late in life. That explains a lot about the founder, but he had taken himself out of commission and was getting medicated.

I was like, if something like that happens—some weird externality-type event—what do we do with the company? Every quarter, we have an immediate shopping list, which is: if this company falls into trouble, I’m picking up the phone, I’m calling this person at this buyer, at this level of the organization, who wants this kind of product or technology. I can grease the wheels to get an acquisition done.

I hope I never make that call, but I’m mentally prepared for making that call. All of us are.

So you will prepare a list of 3 to 5 names of people who would buy in the case of X happening? And not even just the company—who’s the buyer, what’s the division, who’s the person who’s going to be that protagonist?

What if you can’t identify them? I will quite often sit with the team and I’m like, “Here, I totally see who the buyer is here. I don’t know; it’s a bit fuzzy.” What if you can’t? Is that a red flag?

7. How Long Took To Raise Fund One

We don’t do that from an investment perspective. What we do is once we’re actually involved in the company. The nice thing about being a seed investor is these companies have a little bit of life—12 to 18 months—before they have to worry about this kind of crisis mode. But then, at that point, we start building in this resilience.

Got you. Okay. When we think about Fund I, how long did that take to raise?

Fund I was a bear. Fund I took 39 months for us to get going—over 3 years.

And you mentally think, at 24 months, did you have an anchor?

No. We had a lot of friends, and we had 2 buckets of people we went to. Americans, mostly individuals, were like, “We don’t really understand this Europe thing. It doesn’t make any sense to us. We don’t understand why you want to be in Europe, but we like you. We’ll write you a small enough ticket—which turns out to be a decent-sized ticket—but a small enough ticket where, if you lose all this money, and we’re mentally prepared that you’re going to lose all this money, we’ll still invite you around for dinner every once in a while.”

That was our base, and we got to about $8 million on the basis of that. Then we had a lot more to go. We had to get to at least $25 million for the first one to make it viable.

Okay, so 39 months—we have $8 million from that. How did the rest come together?

It was a grind for the next 2 to 2.5 years. The early money was easy, and then it was a grind for 2 to 2.5 years. Finally, we found the family who believed, who then ended up writing a check to us. Then, about a couple of years later, they wrote a check to Isomer.

What was the biggest check that you had in that fund?

$10 million—no, sorry, $15 million. $15 million out of $15 million, and they split it between $10 million for Isomer and $5 million for us.

Wow. $5 million from the family, $10 million from Isomer. That’s a big check in a small fund.

Yeah, and that basically made our fund. We would not have a fund without them.

Is there anything you would do differently on that fundraise when you look back now?

Yeah. I got this advice from Mike Maples, who said, “Do not do a fundraise for the size of the fund; do a fundraise for the time of the fund.” In other words, give yourself 90 days. Whatever you get, go start investing. You’re a smart guy; you can figure out how to do portfolio construction with a smaller amount of money or a bigger amount of money. Go deploy it, go put points up on the board, go prove that those investments are actually really good, and then go back into the market to raise more capital.

Now, again, Floodgate had 4 big LPs: Princeton, Yale, Harvard—I forget who—and I think Notre Dame was the fourth. It was a $75 million first-time fund. To me, it was like, “This is great advice, but it’s kind of luxury advice,” because you have to look at your LP base and the size of your fund.

I ignored it, but the advice I give every emerging manager is Mike’s advice, which is what you ought to do. Give yourself a finite amount of time. Do not do what we did—39 months to basically do nothing with your life other than beg, which is what you’re doing. That’s a lot of time wasted.

Were you terrible at fundraising?

Awful. I think we were selling a story that nobody believed in: “Europe is going to produce really good outcomes. Prove it to me.” Well, there was no data. You look in the rearview mirror and there are no outcomes. Now, 15 years later—13 years later—it’s obvious that there are outcomes here. It’s much easier to tell that story now. People shop for this product, but you’re basically selling a product that nobody’s designed to buy.

Did you come out of the gate fast? Often, when it takes a long time to raise, it’s like, “Woohoo, we have cash now.” Did you deploy fast?

No, we were pretty methodical all the way through. The only time we probably sped up was 2021, but I think the whole industry was speeding up at that point. Then we slowed down intentionally in 2022. We used to do about 4 to 6 a year.

8. The Best Investment From the First Fund

Which is pretty small for a seed fund—pretty concentrated. In terms of first funds, what was the best investment from that fund?

The best investment on paper—not on paper, realized—was Deliveroo. It was about a 34x return on the first check. Whatever we deployed out of the fund, we put in just around $1 million of that first round.

The first round was weird because Will came to me when he was still a graduate student doing his MBA, and I tried to talk him out of it. Most of the founders that we end up writing a check to are very missionary-like. They’re not worried about money or fame or status. What they’re trying to do is solve a problem that they think is really broken, and that’s what they want to do. That’s what they want to bet their career and their life on. That kind of becomes their project.

Will was a distressed-debt guy, and I was like, “You’re going to make a lot more money in London working for a hedge fund or working for a bank. Why do you want to get on a bicycle or a scooter and do these delivery drops? You realize this is going to be you doing the deliveries in the early days as a startup? You’re going to be taking food from a restaurant and schlepping it to someone’s house.”

He was like, “It’s broken.” He was pissed off that he couldn’t get food delivered, and that’s what he did. I tried to talk him out of it when he was an MBA student. He came to us in the summer and said, “I’m going to do this round.” We were in the middle of closing Fund I, so we couldn’t do it. Then he launched.

To be fair, I was somewhat skeptical. I was like, “The world doesn’t need this problem. This is a top-1% problem. High-income earners who have lots of disposable cash want to get food from restaurants and have a driver come drop it off. I’m not so sure this is a mainstream product.” There were lots of other delivery companies back then.

Then, about 4 or 5 months later, it was very clear that he was the most methodical and thoughtful about the operations of the business, which is kind of the core. He built an Android stack that tracked all the drivers. There was real technology in it, and he was growing 50% week on week.

We came in and said we’d write the first check. Then Index came in and gazumped us.

What do you mean, they gazumped you?

The round was supposed to be a $1 million to $1.5 million round, and it became a £3 million round. £3 million was like $5 million back then. That was big enough that the small seed funds like us—which were really the only seed funds interested in this—couldn’t really write or match that check. So they ended up winning the deal, and then Will fought, and we ended up co-investing with Index.

You have that first $1 million. It returns the fund, great. When we think about preserving ownership, how did the preservation of ownership look in that?

We followed, but as a $28 million first fund, which is what we were, it’s hard to follow your capital. Then we had a weird scenario in that particular company where our pro rata rights got taken away from us. In the legal documents, they changed the definition of who would get the pro rata, basically singling us out. There weren’t that many other seed investors that owned above that bar and below the number that they said, and they forgot that we bought common stock from angels.

We politely didn’t comment on the legal documents. Then, in the next round, we said, “We’re going to exercise our pro rata,” and then we were told, “You can’t.” I was like, “Yes, we can.” Then they realized the mistake and changed it.

The seed round was called an A, so it was named by the letter A. We did the B and the C, and then we didn’t do the D, which is when DST came in.

9. How Long Took To Raise Fund Two

How do you think about reserves? There’s the theory that a company will never be as cheap as it is today, so just buy up as much as early as possible. Then there’s also the theory that you see your winners evolve over time, and you should double down and concentrate capital effectively. How do you think about which camp to be in?

We’ve gone from a $28 million fund to an $89 million fund to a $214 million fund, and we haven’t really changed all that much in terms of portfolio construction. What we do, though, is that we are now super aggressive about doubling down. If we see early traction in any one of our companies, we will figure out a way to put more capital in.

These days, for our best companies—and our best companies are much higher concentrations than our average company—we’re getting closer to 15–20% ownership pretty consistently over time. Not on the first check; the first check is the first check. But we put in a second check, and sometimes it’s by the third check.

Those second checks—you’re kind of making the round happen. Are you proactively going to them and saying, “Hey, just take $3 million more in a note”?

We’re finding ways to take more. We can’t—we don’t want to screw up the downstream investors who then have to write the check, and we don’t want to be cherry-picking our best companies. But we find ways to get more capital into these things.

Sometimes it’s as easy as, “Whatever you’re doing the next time around, we want to do more than our pro rata,” and it’s a handshake agreement. In other cases, we’ll do a SAFE on top of our first check. We’ll find a way to put more capital to work when ownership is still inexpensive in the grand scheme of things and build up the ownership.

Will you ever do an uncapped note?

No, not an uncapped note.

Will you ever do common, not preferred? I’m seeing this more and more.

Weirdly, no, because I think in the UK it’s very easy to convert common, if you buy common as a secondary, into preferred. Generally speaking, we’re old-fashioned: we think the preference matters, even in these large outcomes, because there could be volatility downstream. I don’t think we’ve ever really bought common.

10. When To Sell an Investment

The other really challenging element is when you sell and how you manage that. With Deliveroo, when it IPOs, do you just sell then? How do you think about that?

We’ve learned this the hard way. In the case of Darktrace, which we also took public, Darktrace went public at £2.50, traded up to £4, and by the time the lockup expired, it was around £6. We did not sell. Had we sold anything? No, because I was super long-term and was all the way until the end.

How much did you have in that?

We would have been a 10x net fund on Darktrace at its peak.

A 10x net?

Yeah, so our numbers for Darktrace are way higher than our Deliveroo numbers, but we mistimed it. Then you get pressure when you don’t sell at the top. To be fair, this was also in 2021, when the market was just euphoric in general.

When did you sell?

We sold it about a year later because we were coming up to the end of the life of the fund. We sold it around £4 a share, maybe not even £4—I think closer to £3.50. Then I distributed it in specie, so I got a bunch of stock from Darktrace, and I held it all the way until the Thoma Bravo acquisition, which was roughly around £6, I guess.

That 4x’s the fund, though?

Yeah, we made a lot of money on Darktrace, but we probably should have sold programmatically. I think the formula that we now have is: at the time of the IPO, as soon as you’re out of lockup, sell a third of it; sell a third of it 6 months later; and then sell a third of it another 6 to 12 months after that. Just make it a formula, because I think there’s too much human error in this. Long term, I was right, but the markets and what you think long term don’t always map 1-to-1.

Did you do a third, a third, a third on Deliveroo?

No, we sold out of Deliveroo at the IPO. We thought it was very fairly valued at the time. It was at £3 a share; it went public at around £3. It was supposed to go public at £3.90, went public at £3 a share—sorry, after lockup it was £3 a share—and then it came down to about £1 a share. So we looked really smart for selling Deliveroo on the eve of the lockup.

Were LPs grateful?

Yes. I think distribution—I mean, we have real distribution in the first fund. It has distributed multiples back of the fund, so DPI right now is a real topic. But a $28 million fund, even if you multiply it by multiples, doesn’t turn out to be that much money for the LPs.

Do you invest differently when you’ve delivered real DPI? What I mean by that is, bluntly, you’re not downside-protection thinking. You’re not thinking, “Oh, fuck, I’ve got to put numbers on the board.” You’re able to see greatness more easily, having proven yourself.

Yeah. We just had our AGM yesterday, and we told most of our investors, “Do not pay attention to TVPI for the time being,” because what we have been doing is proactively finding ways to put more money to work inside of our best companies.

We know what the best companies are. It’s about a third of the portfolio in the second fund; that’s now shifted to a little bit over 50% of our capital being in the top third of the fund. In the third fund, it’s getting closer to about 60–65% of the fund. The money is going into the best companies.

11. Missing out on making $400M in Darktrace

When that happens, you’re obviously putting money to work at slightly depressed prices. You’re not sending them out to get ridiculous markups, because you don’t want ridiculous markups on those companies. If the founder wants it, then we’re along for the ride. But if you can find a way to avoid the ridiculous markup and put more money to work, you’re buying more ownership. If you’re right 3 to 5 years later, that will make a material difference in DPI.

When you review the best companies, are the best companies the hottest companies early on?

Not necessarily. Darktrace was not a hot company for a good chunk of its existence. In fact, I know that they talked to a bunch of the good and the great, and people were skeptical about them. As a result, they were buying opportunities for us.

The biggest regret we had in Fund I was that, at the Series C, we brought KKR into the cap table of Darktrace. We introduced KKR and kind of brokered the introduction there. There was a bit of a miscommunication at one point, and we put some social capital in to smooth things over.

KKR put up a $40 million round at a $400 million post-money valuation on a company that was doing about $4 million a month in revenue and scaling. I would argue that was a pretty fair price. The company said, “We will give you $10 million of that $40 million. Go raise it as an SPV from your investors. We think you will not get rich enough off Darktrace, and you guys are doing so much work behind the scenes helping us that we want you to have more skin in the game.”

This is a weird scenario to have, by the way, right as a small fund. We went around—and remember, it was February, cold Berlin—and I was going around to all these family offices. We obviously couldn’t say KKR was leading this because the term sheet was confidential. It was like, “A major investor that you would know, a household name, is seriously looking at this.”

We raised zero. At a $400 million post-money valuation, it got privatized at $5.3 billion. This is all dollars, so $400 million to $5.3 billion.

It would have been a net 10x net of fees. We left money on the table, and it was painful—really painful. We have another company, I think our best company right now in our second fund. It’s an AI discovery company, and we wrote the first $1 million check.

Nobody believed in it. This was at a time when people did not understand tech bio; it was before the term “tech bio” was coined. It was brand-new market creation. We were convinced, so we put our chief scientist on the phone with the company, and we were pretty convinced these guys were onto something.

Worst case, the 6-person team would get acquired for the technology chops they had. It would be an acquihire, so there wasn’t that much downside in our case. Then all the data started lining up the right way. The check after us, less than a year later, was $40 million from Bessemer and F-Prime.

We used to own 18% of this company. It’s in an $89 million fund, so that’s a $7 million to $8 million check. It’s a big check to write out of an $89 million fund. You can’t really go to LPs at the Series A and say, “I think this is the next big thing,” because it’s super early. The company has only raised another round on the basis of some data.

We ended up coming down in our ownership. We own about 13% or 14% of the business, post-18%. If that company goes where I think it’s going to go, I think it could be the iconic company of tomorrow.

Have you done many SPVs nowadays?

In the run-up to the Darktrace IPO, we did a bunch of SPVs with our investors. We put more money to work in Darktrace than the size of Fund 1. We think the lowest-performing IRR for us was net 66%—all realized, by the way—and the best-performing was net 155.4%.

So you did over $28 million in SPVs?

We did something like $35 million or $40 million.

What was your blended entry on those?

I think generally we made somewhere between 1.5x and 3x net for our investors, over a year to 2 years. That’s why I looked at the IRR numbers. The IRR numbers are a lot more indicative because there were very short holds.

I’ve been bullish on Darktrace ever since I wrote the very first check, because I saw the numbers. This is the delta: when you have inside information, when you’re close to the company, you know how it’s doing, you see the buying opportunities, and you see that they’re fair prices. These weren’t overly inflated prices.

How do you think about 2 things? One is bias. Let’s start with bias. You love the founder, the numbers look good, but you just really like the founder. They’re messaging you late at night with great ideas, and you have a connection with them that you don’t have with someone new.

We have a fairly trained growth investor on our team who isn’t in these companies and can look at the data on a pure data basis and give a view. We basically assemble a different team other than the person leading the investment, saying, “Take a look at this, figure it out.”

Going back to Darktrace, the company did $4 million a month, or $48 million annualized, at a $400 million post-money valuation, not pre-money. That’s a pretty fair price for a SaaS business.

No, dude, that’s ridiculous. You never see that.

I’m understating it. It was a good deal. It was a great deal. You don’t have to think about the founder sometimes, or the vision. You can just look at it from a numbers perspective, and it’s a pretty good company in the making.

The delta was that they had gone from about $1 million a month at that point to about $4 million a month. It was super-exponential growth in those early days. When the company got privatized, it did $732 million of revenue, but when we invested, it was doing something like $10,000.

12. Doubled Down on a Reserve That Didn't Work

Have you ever had a reserve check where you really doubled down extensively and it hasn’t worked out?

Yes. We’ve learned a couple of things along the way. People don’t really need much from their investors when things are going well. They just need money and for you to get out of the way. Whenever there’s a hiccup, they usually end up picking up the phone and calling their investor, and we’re usually the ones working on it.

I think this is a big transition right now in the seed world. Sometimes, when those calls are being made, it’s not the Series A guys or the multistage funds that are doing the work. There used to be a time when you wrote the check as the seed firm, then the big boys came in, you exited politely, and the big boys ran the business.

They did the board work, the hiring and firing, and the acquisitions if things had to happen. These days, with the growth of those firms, it’s all call options for them. They invest into something and see how it plays out so they can write the $30 million or $50 million check when it starts to get meaningful.

For us, it’s always meaningful, so we usually end up doing all of this heavy lifting. As a result, there’s a bias that comes into this, which is that you think you can fix a lot of things.

We had a company that hit a stumbling block. We doubled down—not with a lot of capital, but we doubled down, rolled up our sleeves, and started working. The other investor with us was another big venture fund. They wrote it off, but weirdly enough, they still held onto their board seat so they wouldn’t get recapped. They defended themselves, but they wrote 0 additional checks.

The company needed about $2 million to turn itself around. We were able to put in $1 million, and we assumed that if we did the first million, someone else would come in and do the other million. We ended up raising about $1.3 million of the $2 million, ran out of cash, and the business had turned around. Someone else bought it out of insolvency and has been flying with it since then, because all the heavy lifting had been done.

I got a call about 2 months ago saying, “I heard that you were such a good board member for this company. I want to give you stock options and have you back in the company on the board. I’m giving it to you for free.” I can’t say the company because it hasn’t closed. We’re in the middle of the legal process, and there’s a lot of back and forth.

Basically, my company went bust, someone else picked it up, and it’s now starting to fly. They heard from the company how much work I did and called me back up, saying, “I want you on my cap table.” They realized that I was tapped out, that our fund had no more money for this, and that I had tried as hard as I possibly could to rescue it before we ran out of capital.

The lesson for us is that this one feels like it might actually work out. This hard-work stuff is something you need to be well capitalized to do, and sometimes it’s not our place. Even if we could do the work, we don’t have the capital base to be able to do these things.

13. Has Series A Product Quality Declined?

Has the Series A product worsened over the last 3 years?

14. Is the Criticism of Europe’s Venture Scene Fair?

I don’t know if it’s worsened, but I think we’ve gone to an era where people are writing checks and then letting things play out. There’s a great transcript oral history that I do a lot of reading around. The Computer History Museum in California has gone back and interviewed all of the good and the great in our industry—the top 35, the founders of our industry, the early venture capitalists.

They did these oral histories, 8- to 12-page PDFs that you can read. If you read those transcripts, the way venture looked in the ’60s, ’70s, ’80s, and ’90s looked really different from the way it does in 2024. These people give you another story.

Back in the day, Dave Marquardt was the only investor in Microsoft. Very few people know this. In the early days of Microsoft, it was structured as a partnership, not an Inc. It was not a company; it was a partnership. There was weird tension between Bill and Paul, so they needed a third party to come in and clean it up.

His firm complained that he was spending a lot of time helping these 2 kids out. They were both in their 20s at this point. The industry was really young, even as it is today. He did that work for about a year before he got invited into Microsoft, and he ended up owning 10% of Microsoft.

I got a lecture from my COO when I did this with another one of our companies. We did a next-generation AI law firm that runs as a law firm, and obviously, if you’re running as a law firm, there’s a whole bunch of stuff that you have to do beyond the tech. I was giving free advice to the founder. We own 20% of the business, and the company is on fire; it’s doing really well.

I spent about a year problem-solving and troubleshooting. Every time the founder had an issue, he would come over to the office, and we’d sit down and work through it. My COO was like, “What are you doing? Your time is really valuable. We have a whole portfolio that you’re supposed to be working on, and you’re brainstorming with this guy.” It’s fun, but sometimes you have to do those kinds of things in order to buy the goodwill to actually be able to write the check.

We ended up coming into the deal at 9%. The company is on a £1 million monthly run rate right now.

Wow.

That work pays for itself, but I think the industry did all of these things in the ’80s and ’90s. We’ve gotten so much bigger now that we’re honestly asking whether it’s worth the time to do this for $1 million, $2 million, $3 million, or $10 million. You’re much better off raising a $500 million or $600 million fund and putting $50 million to work.

Is this good for founders, though? They get a $10 million to $15 million check from a Series A player who’s got a $1 billion, $4 billion, or $5 billion fund. They say, “You’re a total call option. We’re going to give you $10 million or $15 million, get out of the way, see if you’re interesting, and we’ll come back and give you $50 million if you are.” Is that good or bad for founders?

If you’d asked me this question in 2021, I would have said it was good, because if the market’s on the way up and everything is pulling you up, all you need is money from the investors. You don’t need all that much more. It’s nice if they invite you to things, but you don’t really need very much out of them.

If the market stumbles—the market stumbled in 2022—there are a whole bunch of companies out there. If you look at the public markets today and go look at all the SaaS companies between the 1st decile and the 10th decile, everything other than the 1st decile is both growing and profitable. Both, not either or.

That means that if you’re a $100 million vertical SaaS company that’s private today and you’re at that kind of stage, and you’re one or the other but not both, you have a long way to go before you can go public. Those companies need work to go into them. You probably need someone, even if it’s just a sounding board, to sit down and have that conversation with. That’s probably your venture person.

15. Do the Best Founders Really Not Need Help?

If you’re in a call-option business, it’s not worth the venture fund’s time to do all that stuff. I think it depends on whether you’re in bull cycles, bear cycles, or somewhere in between. I think we’re sort of in between, because we’re bullish on AI and bearish on a lot of other things.

Keith Rabois, who’s a friend and has been on the show a couple of times, always says, “The best founders don’t need your help.” Simple. Do you agree?

Yes, until there’s a hiccup. There are hiccups. People forget that even some of the massive outcomes in our industry have had hiccups. Google did not have any hiccups—maybe now it has a hiccup with some regulatory challenges—but Facebook had hiccups.

It wasn’t easy to raise some of the rounds of Facebook. There’s a reason why Microsoft ended up on the cap table. Companies don’t have a linear path from 0 to success. It looks like that because you stretch out the curve and miss all the volatility.

It’s like your glucose monitor. You see ups and downs on your glucose monitor, and then you see the trend line. The trend line goes up when you eat, but you do see ups and downs. You forget about the ups and downs with history, but it’s in the downs where you need someone around to have the call. Maybe not on the ups, but definitely on the downs.

Another thing Jason Lemkin taught me is that, honestly, giving founders true feedback on why you’re passing isn’t worth it. They’ll just argue; they’ll just think you’re a [expletive]. Just don’t bother. There’s no upside.

I’m kind of in the same camp. There’s no upside from arguing with people. You argue with your founders when they’re in your family—you’re an investor, you’re in it for the long term—but there’s the outside world.

One of my founders says your single biggest flaw right now is that you have an overactive [inaudible]. One of our LPs says the same thing: overactive Twitter, or X. Why are you leaving it on there? Do you give a [expletive] what people think?

I was in the gym the other day and you tweeted that some kitchen-utensil company had blocked you. I was going to respond to you, “You know I love you, but even the kitchen-utensil company doesn’t work.”

It was too good. I got blocked from buying frying pans.

What did you do?

I bought a frying pan 2 holidays ago for my wife, and it never turned up. I complained, and eventually it turned up. Apparently, the response to my complaint was to blacklist me. I’m not allowed to ever buy a frying pan from this company again.

Does it affect you? Does it upset you? I think you know the truth, which is that you’re not the most popular dude.

I don’t know if we’re shooting for a popularity contest.

Does it ever have a materially adverse impact on funds if you’re too straight, too controversial, or unpopular?

I think so. Who wants to work with people who are too controversial or too unpopular? Especially when things are going up, you want people who are going to be cheerleading as much as possible.

Fund 1 took 39 months. How long did Fund 2 take?

It was better, but marginally better. It took 28 months. We went out to raise Fund 2 after finishing Fund 1, which was a $28 million fund. Somewhere around 2017 or 2018, we went back out into the market to raise, and we were going from $28 million to $89 million.

We were trying to get to $100 million. We thought $100 million was the right number for a seed fund back then. We now think the closer number is $150 million to $250 million, but back then it was about $100 million. If you were going to play this game well, that was about the size you needed to be.

We went out for $100 million and got a commitment from the European Investment Fund. It’s a long process with the EIF, and they were the anchor LP of record in Europe, especially in that era. We asked them about Brexit, and they said, “We don’t see a problem. Article 50 hasn’t been invoked.”

We got to the final terms, but we hadn’t entered legal documentation. We were doing the jurisdiction work when Article 50 got invoked. Everyone who had a check or a commitment from the EIF at that point lost the commitment because they were no longer able to invest in the UK.

We also had a big insurance company that was going to be the co-investor and co-anchor. We lost the insurance company as well, so we basically reset back to 0 and had to start the fundraise from scratch in 2018. We eventually got to closing in 2019.

What was the big breakthrough moment? One or 2 big commitments?

We got British Patient Capital, which is part of the British Business Bank. Instead of the EIF, we got British Patient Capital, and it was a good chunk of the fund—about 40% of the fund. I don’t think we would have had a fund without them.

We’re always told—and I know this is an incredibly luxurious position to be in—never have anyone over 10% of your fund. You don’t want to be too concentrated. Is it better to get the right-sized fund with an imbalance of LPs, or the wrong-sized fund with the right balance?

If you’re going to deliver a ton of returns for your investors, get to the right-sized fund.

You mentioned British Patient Capital. Should governments be funding venture?

I think in Europe they’ve had to, but I have a controversial take on this. The worry about governments funding venture, especially at those kinds of concentrations, is that you end up with governments having market power.

I’m a big believer in capitalism and in markets. If you have someone like the EIF representing 30% of the aggregate capital of LP commitments, it’s too big. What you should really have is what happened with AT&T in the United States, where they broke it down into the Baby Bells. You had 5 different Bells competing with each other in telecom. You probably need 5 different EIFs competing with each other in the market.

If the government is going to step in to help, you don’t want it concentrated in 1 big power, because then you end up with weird terms that may not be market terms, and it’s really hard for the market to function the way it needs to function. If you’re going to do it, do it competitively.

Or it could actually force the hand of pension funds, which sit on the side in the UK and do absolutely nothing, which is a disgrace.

I looked this up. The UK pension funds, in the defined-contribution schemes, have about 10% of their capital invested in the top tech names in America. About 5% of the pension funds is invested in UK equities, so they’re actually pretty long tech as far as pension funds go. They’re just not in venture.

I have a genuine worry. It’s really hard to be a VC, it’s really hard to be a founder, and I think it’s also reasonably hard to be an LP. If you’re coming in from scratch with no knowledge, I don’t know if you really know what you’re doing. It takes money and time to train a VC, and I think it takes money and time to train an LP.

It’s super hard. So, pension funds now decide to allocate capital to venture. Who’s going to do it? Where’s the talent base in the UK of experienced LPs who know how venture works and know what kind of funds to bet on? There aren’t any.

No one is talking about this. You’re going to have a problem. There’s going to be no one. We already have too much cash in Europe.

This is the other problem we’ve had since 2021. A number of people entered the industry. We used to have about 10,000 people doing tech investing, venture, and so on. It went up to around 35,000 and has come back down to the historic norm, but there are still a lot of people doing this stuff. I don’t know if people know how to allocate capital to who’s good, who’s exceptional, and who’s average.

We have a lot of fund managers through the shows and everything, and a lot of LPs come to me and say, “Here’s my book. Tell me how I should size it.” They talk to me about their annual budgets. The annual budget is generally about $300 million to $500 million for U.S. endowments.

$300 million to $500 million annually in venture is almost impossible if you want great returns, because you’re going to get $20 million in your top names and there are probably 3 or 4 top names. We’ve got $80 million out. Why are you going to put $220 million into everything else?

You have to bet on the right emerging manager, or you have to play the index. You can’t even do that. You could do 5 emerging managers at $10 million each. Great, now you’ve got another $50 million out and $170 million left on the small side. Then you’ve got to put $40 million into Andreessen, because where the [expletive] else are you going to put it?

I thought the Tiger playbook was fascinating in 2021. Let me not sell a product that’s designed for returns. I’ll sell a product for capital deployment and just buy the index. I’ll hoover up not $50 million commitments, but $200 million commitments from people who have to deploy into tech, and I’ll just buy the index.

It didn’t work because they were overpaying in the market, but I could understand the appeal to the LP base. Where am I going to put this money? I’ve got a group that’s going to take not a little bit of money, but a lot of money, and be able to play the market for me.

The thing that’s so challenging with that strategy is that you assume the outcomes are equiprobable, independent of how much cash goes in. What I mean by that is that you’re saying, “We’ll pay up, but it’ll still go to 3x. It may not be a 5x.” That was the mistake.

I do think there’s a correlation between how much money goes into a company and what the probability of success is. This is why, if you’re going to be contrarian—and we take pride in being contrarian—you have to make sure your companies get capitalized. If they don’t, they don’t have it.

The average is about $300 million to get to unicorn status. There are some companies that do it for $200 million, but you have to raise that kind of quantum of capital.

The biggest structural problem we have in the UK and Europe is that the conversion rate between rounds—from seed to Series A, Series A to Series B, and Series B to Series C—is basically on par with the U.S. these days. But the capitalization of our companies from seed to Series A and Series A to Series B is far below what happens in the U.S.

There’s a statistical correlation between the size of a seed round and the probability of becoming an outlier. If a seed round raises $100,000, the probability of it becoming an outlier is very, very, very small. If that seed round goes up to $10 million, the odds between a $5 million seed and a $10 million seed basically double.

I would always assume that $3 million to $5 million is optimal, but $10 million actually becomes detrimental.

Yes, but there is a reason for companies to raise the right amount of capital at the stage. Too much capital becomes too much of a wash, but there is a number where you’re freeing up the capacity of the founder and of the company to try to achieve greatness. You’re shooting for greatness as fast as possible.

16. Is Hoxton Price Sensitive?

I don’t think people fully grasp this in the European venture ecosystem. The big problem we have in Europe is that we raise small rounds. People will take the risk, but they’ll mitigate the risk by writing a small check.

The inverse should be true. If you believe in this thing as a seed investor at $1 million and you have the fund size to do it, you should believe at $3 million and you should believe at $4 million. It doesn’t make sense to believe at $30 million, but there is a number that frees up the capacity of the founder and the company.

Are you price-sensitive?

We care about ownership, but not when it comes to the size of the check. Most of our deals are contrarian. They’re contrarian even within the table. The rest of the group doesn’t get it. We don’t see an obvious reason not to do it, but we don’t see it.

Some of our best deals are usually like this. Instead of downsizing the commitment, we say, “We don’t really get it, but they’re raising $3.5 million. If they’re going to make a real run at this, maybe they should have $4 million or $5 million. Maybe you should go in there and buy an extra few points of equity for that money.”

That’s the right way to play the power law, but it requires a fund size bigger than the $100 million we initially thought was appropriate.

So they’re contrarian even in the partnership? “We’re partners, dude. I want to do this deal, you really don’t, and you think I’m nuts. Can I do this?”

We do it on the basis of whether there’s a red line. Is there some flag that we can throw down that says it doesn’t make sense? I look at the cohorts and the early cohorts. If all of those cohorts are deteriorating, then you may think it’s really good because it’s growing exponentially, but the data suggests that maybe it isn’t.

That’s a real-world scenario. I got super-excited about a company in Portugal, and 1 of my partners looked at it and said, “Hussein, you missed a trick here.”

This is why I love working in partnerships. I think partnerships are much better than solo GPs because you get an error-correction mechanism from other smart people. But if the error-correction mechanism is just that they’re blocking you for no good reason and running interference, then it’s really difficult.

There’s always a reason to say no to a company. There’s always a cohort that’s off, a conversion rate that isn’t there, or a retention metric that’s down.

17. Why $150M to $250M is the Optimal Seed Fund Size?

So you said $150 million to $250 million is where you’re naturally thinking the optimal seed fund size is. That’s bigger than most people would suggest. I’m in your camp on this; my new fund is $125 million for seed. Why do you think $150 million to $250 million?

If you’re going to try to do 20 of these, your check sizes are going to be $3 million, $4 million, and $5 million. Sometimes these jumbo seeds are now up to $10 million. Ed Sim has done a bunch of work showing what the path is for these jumbo seeds, and they’re more and more common. Jumbo seeds have increased 6 or 7 times in volume from where they were a couple of years ago.

The seed rounds that are $5 million-plus are a fifth of the industry. You do those jumbo seeds at inception. We did 1. It was a $30 million round where we took $10 million off the table, locked down the price, and put in $10 million.

It was an AI foundational-model company. It didn’t necessarily need all the money for compute, so $10 million into a foundational model is a really large number, but in foundational-model land, $10 million is still small.

We took it off the table, then everyone wanted in. We selectively let a few funds in. Lightspeed came in, along with others, and the round became $30 million.

A $10 million check is a very big bet out of a $200 million fund. This is why you have to be a decent-sized fund. If you’re a $100 million fund, that’s 10% of your fund right there. Five percent of the fund is very different from 10% of the fund.

I would argue that the number should maybe be $200 million to $300 million if you’re going to double down proactively. If you have another vehicle, you can double down out of that. But if you’re not able to have another vehicle and you’re going to do it out of the same fund, you probably need to be a little higher than $125 million.

What are your capital-concentration limits per company?

Ten percent.

So you say you have another 5% for this company over time?

Yes. I think in this case, we believe AI in the foundational-model space is somewhat binary. It either works or it doesn’t.

The company was funded recently. The term sheet was in March, the wire went in June, and we’re sitting in December. It would be foolish for me to say anything about it. We’ve seen the commoditization of different model providers happen very quickly over time.

This isn’t a generic model. It’s a model for materials science. It’s a bespoke foundational model that builds the next generation of materials using AI. There was a piece of Microsoft Research that came out 2 years ago proving that this kind of thing can work. It’s a very different type of foundational model.

Okay, I get you. Going back to it, you have 20 companies in the portfolio, which is a pretty good picker at 20, by the way. The probability at seed of picking something that becomes a unicorn is 3%.

If you have a $150 million fund, are you doing 20 $5 million checks, minus fees, plus reserves?

Plus reserves.

So, roughly, a 1-to-1 rule of thumb?

18. The Problem with $75M Seed Funds

It’s not quite what we do, but just double it because that’s the easiest mental way to think about it.

What do you think of all these seed funds that are $75 million?

I’m worried. In a bull market, where you’re writing the first ticket and someone else is carrying the slack and picking it up, it’s easy to be the feeder fund for those people and write the small checks. I think we have too many of them in the industry right now.

I think the real opportunity in Europe is that there are a handful of really good venture funds at the top. We know who they are: Index, Accel, and others, all very active in Europe.

Do you think there’s actually a handful?

I think it’s slightly broader than that.

Creandum would feel really annoyed if you didn’t put them on the list. I think that’s true of a bunch of other funds.

They’re a seed fund? I think they’re a Series A fund.

Okay, fine. There are a handful of funds that are bigger and aiming for big outcomes in Europe.

The market here has grown 30-fold. When we first started, about $1 billion went into European venture. These days, about $30 billion goes into European venture. Weirdly enough, people seem to think that markets becoming more liquid and competitive is bad.

I think markets becoming more liquid and competitive is good because the market is actually working. If the market goes up 30-fold, even if there’s more competition, I’d much rather play in the bigger market than the smaller market.

I think there’s a chance now for a few more funds to be on that list. I think you have that ambition. I have that ambition. Ailie has that ambition at Blossom, and some of us are going to make it.

I don’t know if the world needs another emerging manager or another micro-cap fund right now. I think what we need is 5 to 10 dominant superstar venture funds in Europe, the way there are 10 or 15 in the Bay Area.

We do a lot of first-time founders. I think a lot of people in Europe won’t back someone if they look like a first-time founder.

Do you prefer younger founders?

Younger or older doesn’t matter. I think the people who are doing their life’s work usually have the company capture their life’s work. That’s the only thing they do with their career.

Nick from Revolut was on the show, and he said that when they look at the work they do with QuantumLight and analyze founder age, 25 to 35 is the optimal time. They find the best performance there.

I think if you’re saddled with a family, it’s harder because you have dual interests. You end up in this solo mindset of, “Build my company,” and that’s much harder to do when you’re raising a family.

If you’re 15 or 20, unless you’re a superstar and really precocious, you probably don’t have the accumulated wisdom to learn the lessons. I think 25 to probably 35, or 25 to 40, is the period when you can grind and have enough experience to know what to do.

Do you think there are enough high-quality seed companies graduating out of London and Europe for the multistage funds?

I don’t know about the general market, but in our portfolio, for sure. That’s why we’ve ended up concentrating. You are seeing the U.S. players come in with large amounts of cash and spend.

The challenge in Europe is that most of these companies need a little bit of tinkering. You can’t just fully leave them alone. You have to think of Europe the way venture was in the ’80s or ’90s in the United States.

If these companies didn’t have a strong partner on board helping them build, and you read these oral histories, you’ll see what I mean by companies actually getting guidance on what to build. We’re in that mode.

I think the industry in the U.S. has shifted to, “The market takes care of that stuff. I just have to deploy capital.” In Europe, the market doesn’t take care of itself.

Do you think founders are aware of that? They read “Founder Mode” by Paul Graham, they read the U.S. articles, and they read everything U.S. founders consume, but they’re operating in a different environment.

I don’t know if that necessarily always resonates, but the minute there’s a hiccup and you have those hard conversations with founders, people grasp it.

Do you think the criticism levied toward Europe today, which we both see on Twitter like never before, is fair or completely unfair?

I don’t know. From a macro perspective, that’s not what I do. I think of it as the underlying fundamentals for my business.

We’re living in a world of AI. I think this is the big seismic shift for the next 10 years. This is where we’re going, and this is where the next wave of wealth creation is going to be.

I’m looking at the conditions on the ground. When I was at Accel, we used to produce really interesting gaming companies. We were really strong in Europe, and that was probably the only thing we were really strong at.

Then the government lowered the regulation in finance, and we became really good at building fintech companies here—Monzo, Revolut, and others. The FCA made it easy from a sandbox perspective.

When you look at this new big opportunity, you’ve got DeepMind down the road in London and Meta running its AI work in Paris. For the first time in European history, we’re on par with the U.S. in company creation or technology creation—not in a niche field, but in a horizontal field.

I can’t interpret that in any other way than there being an opportunity in this area.

But we don’t have the supply of entrepreneurs.

We do a lot of first-time founders. I think a lot of people in Europe won’t back someone if they look like a first-time founder, but some of the most interesting outcomes in our industry have come from first-time founders.

I’m terrified that the German car industry is going to get wiped out by China and Chinese EVs. I think there’s been sleepwalking. What happens then? I don’t know. I also don’t know what’s going to happen to energy prices in Germany.

I’m worried and petrified about UK growth stagnating and quality of life here deteriorating. But from a company-creation and investing perspective, those things aren’t necessarily related to what we do as an industry.

They are, because if you consistently say, as Keir Starmer and Rachel Reeves do, that we will not have growth for the next 3 years, you’re not putting up a banner for great entrepreneurs saying, “Come build here.”

There is a correlation effect. I don’t know if entrepreneurs, engineers, and tech people are thinking about macro when they’re thinking about doing something, but that is a problem.

Then the question is how you scale the company. If you’re scaling the company, your best path to scale from a financing perspective is America. The rounds are bigger, the chances of success are correlated with bigger round sizes, experience is bigger in the U.S., and the market is bigger in the U.S.

The direction of travel has always been the thesis at Hoxton. Darktrace made more money in America than it did in the UK from the very early days. It had more staff in America than it did in the UK. Our direction of travel is to find the best here, be the bridge to America, and take them over there.

From my perspective, I feel fine regardless of what’s happening in the macro. As someone living here, I’m terrified about the macro because there are real quality-of-life repercussions if policymakers don’t get this right.

I think policymakers are focused on the wrong problem. Everyone here is talking about the London Stock Exchange. I just told you that defined-contribution pension funds in the UK are investing 10% in the big tech names in the U.S.

It’s not hard on a Bloomberg terminal to put in a few extra characters and buy a share on a New York exchange or a Nasdaq exchange instead of the LSE. It just doesn’t matter. The world is global these days.

We need good, strong local liquidity markets, which we don’t have. But why? The LSE doesn’t matter. If you live in a global world, why are you still focusing on one exchange?

I would argue that it only allows the top 1% to flourish. If you’re not Spotify or Revolut, you’re not going public these days. Only the top 1% of IPOs can go to the U.S.

We looked at the data. The bar for an IPO today in the U.S. is north of $200 million to $300 million. Gone are the days when you could take a company public at $100 million of revenue.

That was a 10-year-old world. There used to be a time when Yahoo went public on $10 million or $20 million of revenue and was worth billions in the market. Those days are gone. We now have very deep, liquid, large markets across the industry.

The private markets are so big now that there are other ways of getting liquidity than going public. There are private-equity firms that would probably want to buy LADbible or take it out.

I don’t disagree with you, but I don’t know if this is that much of a problem. Why are you solving a problem that’s really hard to solve and probably doesn’t matter?

If the path to a New York Stock Exchange listing were closed from here and our best companies couldn’t go public, we’d have a real problem. If our pensioners couldn’t access those markets, we’d have a real problem.

Can we access Nasdaq and U.S. markets en masse—not 1 or 2 companies, but hundreds? We have to build those companies and get them to $200 million. We have a company in our portfolio that’s about 2 years away from an IPO. It’s appointing bankers right now and has $150 million to $160 million of net revenue on a run-rate basis. It’ll easily get to $250 million, which I think is where the bar is to go public on the New York Stock Exchange.

The first 2 big tech IPOs on the LSE that were supposed to reinvent the exchange were ours: Deliveroo and Darktrace. Then Wise went public on the LSE, and I tried to talk them out of it. I said, “Go to America.”

I just think the game is different.

Would Deliveroo be performing differently if it were in America?

I’m not so sure about that one. I think Darktrace definitely would.

How so?

I think Darktrace was trading at a huge multiple discount. It had $732 million of revenue and was privatized at $5 billion. That was a premium of, I forget what the number was, 20% to 30% over the stock price. The stock price bumped on the day of the acquisition announcement.

We had C. L. Holmes on the show, and he said what we need is a European Delaware. Why? It’s the same thing. People want to solve “EU Inc.” They want to make it really easy to incorporate a company in Europe. Why incorporate in Delaware? Problem solved.

I’ve seen these petitions. We signed one, whatever it was. I’d love for it to happen because it makes things easier and removes the borders.

Anything that makes it easier, I’m supportive of. If the LSE is able to transform itself and become a viable exchange, great—another path of liquidity. But as a policymaker, if that’s the problem you’re focusing on, I’d much rather have you focus on how we make sure our companies get capitalized the right way and can become the top 1%.

I was at a dinner last week. The Wall Street Journal put it on, and I was told that one of the biggest impediments to housing in the United Kingdom is that we didn’t build reservoirs 30 years ago.

There is a problem. We don’t have enough reservoirs. This is a country where it rains a lot. Water isn’t our scarce resource.

19. An Advice For Keir Starmer

If we wanted to build 10 million additional units of housing for 10 million more people, we wouldn’t have enough reservoir infrastructure. You’re advising Keir Starmer; I am not. I’ve never been called. Hypothetically, what would you advise him?

I wouldn’t have tinkered around with the tax rates. I feel like there was enough own-goal removal already.

I’ve said this publicly on X: I have no problem paying income tax on carried interest. I think it’s income. It’s not my capital at work; it’s because I’m doing my job. I have no problem with that.

But I think tinkering around with things and changing them frequently is a problem. Capital gains have now gone up in the UK. Fine. Are they going to tinker with them next year?

I want stability. I want infrastructure and policy to be long-term, stable, and consistent so I can focus on the really hard things, like building these companies. If policymakers start tinkering with the things I take for granted, it becomes infinitely harder for me and the founders to navigate and build these companies.

The message should be: stick to a strategy. Boring is good for governments. That’s not the world we’re in, especially with social media, where being more exciting gets you more credit, kudos, and attention.

We mentioned that, in terms of liquidity markets, we can just go to the U.S. or not go public at all. We have different players, and the extended window of privatization and private capital is real. You’ve got Stripe, Databricks, SpaceX, Starling, and all these companies that don’t need to go public for the foreseeable future.

Is this extended window a problem? Is it good or bad?

It’s too hard for me to know. We live in a very different world in 2024 than we did 10 years ago, and definitely than we did 20 or 30 years ago. The markets have evolved in such a different way, and they’re so much bigger that I don’t know how this is all going to play out.

I’m a realist. This is just the way it is. This is the world I play in, and this is how it’s going to be. I don’t see it going back anytime soon to the way it was in the ’90s. I’m just going to adapt to this new reality.

Do we have to navigate secondary liquidity markets differently?

I don’t think so. As a seed firm, you build great companies. If someone offers you 50x on your first investment, maybe there’s a reason to take some money off the table. Normally, you would have said that you get your 50x at the IPO stage or at the very late stages, and the same argument would apply.

If you get it in 1 year, the same logic applies. You just do it 1 year in.

Are you ready for a special type of round? We get these questions from mutual friends or people in the industry, and we put down a number. You can either answer the question or donate the number to a charity of your choice.

What number do you want to set as the donation?

What’s the market for this, between $1 and $5,000?

$2,000.

Okay.

You caused a bit of a stir on social media with a post about hiring women. What did you say, and what did you mean?

It’s hard to find people to come into a partner-only organization. You require people to be reasonably well trained. You’re taking a chance on them, but you expect them to be able to hit the ground running.

There are some great women who are very capable, but it’s hard for me to poach them because they’re very well taken care of in their existing funds, for very good reasons. There’s a shortage of them.

One of my LPs, who’s a woman and an individual entrepreneur, sat down with me. The thing I hadn’t thought about, and which she made me reconsider, is that I always used to think of us as a 2-year scrappy startup. It was a brand-new fund, and it took 39 months to raise Fund 1. That’s etched in my brain as hard yards.

But we’re 11 years old. We manage a $200 million fund, and as much as I don’t like to admit it, we’re now part of the establishment. We’re no longer the scrappy startup.

If there’s a shortage of women whom I can recruit laterally because they’re well taken care of, and there’s a shortage of women coming up through the industry, I’m part of the establishment now. I can’t be scrappy anymore. I probably have to invest money and train someone to fix this problem.

The problem falls on my shoulders. I think that’s an interesting lens through which to think about it.

I can’t do that just yet. I don’t have enough partners in my partnership. We’re a small partnership; we’re 3 GPs.

Do you feel like you have freedom of speech today?

Yes. You get freedom of speech, and then people call you out on it, which is what happens.

Do you think it was a fair response?

I do not. I reached out to the person who wrote the LinkedIn post multiple times to grab coffee before it went out, on the day it went out, and after it went out. I’ve chased afterward, and she’s never met me.

I feel that’s where society has gotten a little more toxic. I would much rather have had that debate in person. You can have the debate on LinkedIn and call me out. If I say something stupid, call me out. I have no problem with that. I can take it, and it was a fair criticism, I thought.

But then sit down with me, have coffee, and let’s break bread.

You said $2,000. You split up with your partner, Rob. What actually happened?

After 10 years, we grew apart. That’s what happens. I think the nice thing about having success is that we made real money on Fund 1, even though it was a $28 million fund. We did really well for ourselves personally.

You get to build the firm in the way you want to build it. The big difference between Rob and me is that Rob wants to build a different type of firm. We could think about how to make that work within the constructs of Hoxton, but it would be hard.

You’d have to change the character of the firm. You’d almost be like Millennium, where you have people running their own books. Or you could say, “Go gracefully and build it.” I’m an LP in Rob’s fund, and I’ll be the first check in the fund. He has my full support, no restrictions on trade, and so on.

You keep your track record, keep the LP base, and go do it the way you want to do it. That’s basically what we ended up deciding. It took us a while to be able to do that.

We very much want to build the next-generation, bigger, earlier-stage firm. We want to be 1 of those dominant 5 to 10 firms. I think Rob does not want to do that. Rob very much wants to build a science-oriented, deep-tech firm. It’s just different.

Which venture investor do many people in Europe think is great that you do not?

Historically, I would have been able to answer that, but these days I don’t know. Historically, it would have been quick commerce, Amazon roll-ups, and electric vehicles.

That’s an easy one. I should donate anyway. $2,000.

Let’s do a quick-fire round. I’ll say a short statement, and you give me your immediate thoughts.

Sounds good.

What have you changed your mind on in the last 12 months?

That you can actually make money across the spectrum in AI. CuspAI was 1 of these. We would have said we weren’t doing foundational-model deals. They’re too expensive and too capital-intensive. It’s not a place for a seed fund, especially not a small seed fund.

Then CuspAI walked in the door. It’s a foundational model for materials science. We wrote a double check, using our reserves, and made 1 large investment. We own 11%, which is a decent-sized investment.

If it goes the distance and they raise $100 million or $200 million, some of these funds will be fine.

What’s the best investment advice you’ve ever received?

Play the long game and be contrarian. But you have to make sure that the market sees you as right within a very short amount of time. You can’t be contrarian for 10 years or more.

What’s the biggest sin of the zero-interest-rate environment?

So much money went into some of these things, and people just relaxed their diligence.

Are we seeing that today? Do you think we’ll see many more frauds?

Fraud is people not paying attention to details and not turning up to things. It was a mess. I think we’re seeing some of that same stuff in AI. There’s so much euphoria around AI that people feel they have to have some of these companies in their portfolio.

As a venture industry, we have to think about how to create monopolies. The regulator doesn’t want monopolies, but we want monopolies. We want companies with increasing returns to scale and deep, defensible moats.

You build this thing, it has a huge moat, and every additional customer and every dollar of revenue increases the size of the moat. It puts distance between the company and everyone else until eventually it has to be broken up by regulators because it’s too powerful.

That’s what I want to put money into. I’m not sure about these companies that get commoditized super-fast, where there are 20 versions of the same thing. The expression that Brian on our team uses is “a knife fight in a phone booth.”

The reason we’re doing them is that people want to deploy capital and write checks into these things. They don’t want to miss out on the next big thing because they’ll look foolish as a big firm if they miss it.

This doesn’t sound like a retreat. It sounds like it can go the right way, but it reminds me hauntingly of 1995, 1996, and 1997. Do you play the game on the field?

You have to. You can’t sit it out.

Doesn’t that go against what you just said?

Yes, but most of our investments are about figuring out how to be contrarian and still play the game on the field. It’s not always possible.

You have to be prepared to be a little bit lonely, but you can’t be too lonely for too long. If you are, your companies don’t get capitalized. You need the capital to come in, but you also have to be prepared to be a bit of an iconoclast for a while.

Do you think we will go through an AI winter in 2025?

I think we’re going to go through something, but I don’t know what that something is. You look at the dot-com industry, and the household-name companies of the time didn’t necessarily turn out to be the big outcomes.

Yahoo and Netscape went away. Amazon powered through, but Amazon was never the super-hot company. eBay was the hot company back then, and the Googles and Salesforces didn’t exist. They came at the tail end of that period.

I don’t know what AI is going to look like 5 years from now.

Do you think Nvidia is undervalued today?

One piece of investment advice is that you can’t think of something as too cheap or too expensive. You can’t go into a company because it’s too cheap, and you can’t walk away from a company because it’s too expensive.

The big question for me with Nvidia is that its net margins have grown from 10% to 50%. Besides the revenue growth, which everyone looks at, that margin growth has made it a natural monopoly.

But if you look at what everyone is doing today, Apple, Amazon, and Meta are all building chipsets to remove their dependence on Nvidia. I don’t know if those chipsets will actually get anywhere. I’m not a semiconductor guy, so I don’t know what’s coming around the corner that could commoditize it.

A 50% net margin feels high. If that margin comes back down to even a very good 30% net margin, the multiple changes. The answer is that you have to be a technologist to do technology investing.

What do you think is coming around the corner that may or may not threaten that margin? If you can figure out the answer to that, you can play the Nvidia game.

OpenAI at a $160 billion valuation, Anthropic at $40 billion, and X at $50 billion. Which one do you buy?

OpenAI. It has real revenue, and I think there are increasing returns to scale. But the same thing I just said about Nvidia applies to a lot of this AI stuff.

I had dinner with Alex, who founded Wayve, one of the AI companies here working on self-driving cars. In our discussion, we talked about how quickly this market is commoditizing. The tools are getting so good, and other people are building tools.

A couple of weeks ago, we saw that China had a company likely associated with Kai-Fu Lee that had been able to replicate what GPT does with a fraction of the compute, because China doesn’t have the same access to computers. They published a lot around this.

It’s commoditizing so fast that I don’t know how much of this ends up as consumer surplus. We all benefit as humanity because the spending goes in and everyone benefits, but it commoditizes so fast that no 1 company ends up skimming off enough of the cream to become the big winner.

I don’t know where any of this goes. But if you don’t play, you have no way of knowing where it goes. You have to be on the field to learn.

Does Trump open up M&A and IPO markets?

Yes, because JD Vance has made it very clear that, for sub-$500 million deals, the FTC has no business trying to block transactions. That’s phenomenal for seed funds and for recycling capital to put more dollars to work.

It helps that our vice president used to be a venture capitalist.

Final one: where do you want Hoxton to be in 10 years?

You said you want to be 1 of the big generational players. That takes us to 2034.

I’ll answer it a little differently. You know the direction of travel for the company. I would love for us to have built the partnership—I’m thinking actively about this—so that I can hand the reins over. I want to be 1 of a few partners, with someone else running the firm.

If I can do that successfully over the next decade, I’ll know I have a durable firm. I want to build a firm, rather than a boutique or a project. It shouldn’t be a shell around me. I want the firm to be around, and I want to be able to pass the reins over within the next decade.

It’s only been 9 years, so it’s not a huge amount of time. I’ve so enjoyed this. Thank you so much for coming, and I really appreciate it.

My pleasure. Hopefully we do this again in another decade.