Lead Edge Capital 创始人 Mitchell Green：传统VC为何失灵

1. Why AI Infrastrcture is the Worst Investment to Make

I think investing in AI infrastructure today is like investing in websites in 1997. The incumbents usually win. It’s customer distribution. The idea of a single-person AI company, I think, is comical at best. I think the venture industry was about to be in for a rude awakening, and then AI showed up. People didn’t learn a damn thing from 2020 and 2021. It’s shocking.

Mitchell, I’m so excited for this. When Nigel Morris messages me and says, “Hey, you’ve got to spend time with my friend Mitchell,” I’m like, “You know what? This is going to be a fun one.” Thank you so much for joining me.

Absolutely. Thanks for having me on. Nigel’s a legend.

2. Biggest Advice to Smaller Emerging Managers

He is a legend. It always makes me feel very lazy, though. He’s so athletic. He’s also the hardest-working man. I joked with him the first time I met him, “How’s retirement?” Then he showed me his Outlook calendar, and I thought, “I think you work more now than you did when you ran Capital One.” By the way, never, ever go on a bicycle ride with him.

I would never.

3. How Bessemer Taught Me The One Golden Rule of Investing

Before we dive into Lead Edge, there was Tiger and there was Bessemer before. When you think about your takeaways from those experiences that shaped how you operate and run Lead Edge today, what are the 1 or 2 that really shape how you think about Lead Edge?

What I would tell you is that my time at Bessemer was very formative for why and how we do everything at Lead Edge. For a bit of context, when I joined Bessemer—I think this was in 2005—Bessemer was this legendary early-stage venture fund that was very “Shark Tank”-esque.

Every year, 1,000 entrepreneurs would walk in the door. At the time, they had 5 partners, and it was very much like “Shark Tank.” They were wondering why Insight was finding these $15 million-revenue companies that had never raised money. They were personal friends with the guys that ran Insight, Jeff and Deven and the others. All Insight was doing was replicating what Summit and TA did, which was hiring 22- to 24-year-old knuckleheads—which my now-partner Brian and I were—and pounding the phones, calling companies all day long.

You realize that if the company calls you back, the company sucks. It’s the CEO you talk to every 2 days for a month. How do you know what a good company is? Over 2 years, talk to 10,000 bad companies.

When we got there, a week into the job, they said, “Okay, next Monday you’re going to come and present your best companies.” We got there and said, “We found this great company. It’s $2 million in revenue. It’s going to be the next Google.” They said, “No, it’s not. This company sucks. Find us companies that meet $10 million of revenue.”

The next week, you’d find a company that met $12 million of revenue but grew 10% a year, and they’d say, “No, no. Find us companies that grow 50% a year.” Then you’d find a company that had $20 million of revenue, grew 40% a year, but had 30% gross margins, and they’d say, “No, no. Find us businesses with 70% gross margins.”

They had 5 criteria, and over a period of 6 to 8 weeks, they built this very rigid framework. They basically said, “On Mondays, when we do our pipeline meetings, we want you to never bring a company that meets fewer than 3 criteria. If it meets 5, you better already have the meeting set up for the next meeting and start the pipeline meeting with, ‘I spoke to Company ABC. It meets X number of criteria. Here’s what it does.’”

It was a very rigid framework in a world where you can call companies all day long and have an unlimited universe. We took that framework and expanded it to 6 criteria. Now it’s the Lead Edge 8, and that defines everything we do.

I love that in terms of how it defines everything you do, and I love the framework structure. I had someone from Spark on recently, and he said that, bluntly, this form of spreadsheet investing—respectfully, and I hope you don’t mind me calling it spreadsheet investing—is outdated in a world of AI and the next generation. A banker-like approach will not work in the next generation. Is that fair, and how do you think about that? I’m concerned that it is the case.

Look, we speak to 10,000 companies a year. We have a team of 20- to 24-year-olds that speaks to 10,000 CEOs a year. If I say I need to meet all 8 of these criteria, it’s about a 1% yield. Of 10,000 companies, 100 meet all 8 criteria. To do 5 to 7 deals a year, that’s too small of a pond to fish in. You wouldn’t end up doing anything.

What we find is that if you say, “I need to meet 5 or more of these criteria,” it’s objective. You’re 23 years old: Does the company have $4 million of revenue or $18 million of revenue? After speaking to probably 70,000 companies over the last decade, about 10% meet 5 or more criteria.

You do diligence on 1,000 to 1,200 companies. How do you go from 1,000 to 150 to 175? Most aren’t looking to do anything because you’re calling them; they’re not calling you. By the way, the good ones don’t call you back. The good ones you call every 2 days for a month. That 150 to 175 leads you to do 5 to 7 deals a year.

In terms of the AI response, we find the companies. Seventy percent of the stuff we invest in, the guys at Spark have never looked at. They’ve never heard of it. Why? They’re investing mainly on the coasts. Less than 10% of our companies are in the Bay Area—not because we don’t like Bay Area entrepreneurs. We love Bay Area entrepreneurs. They can build some of the biggest companies on the planet. I just think it’s very hard to make money investing at 100 times revenue.

If you look at our companies, less than 10% are in the Bay Area. Seventy percent of the time, we’re the first institutional investor. Now, is AI going to disrupt all this stuff? When DeepSeek was announced—which I find quite funny, because a lot of people who invest in AI infrastructure didn’t even know about it—I think investing in AI infrastructure today is like investing in websites in 1997.

You and I could have taken $50 million, bought Sun Microsystems servers, and built a website. Today, for £10 a month, we can build a website that’s 50 times better than that. The same thing is going to happen. Prices are going to plummet.

The stock market actually acted pretty rationally that day. What happened? NVIDIA stock fell, and the software stocks went up. Why do I mention that? We’re investors in a company in Toronto called Gravity. It’s the world’s most boring company. It makes budget-planning software for small local governments.

If you’re the water district of Avon, or if you’re the Palo Alto Police Department, you need to post a budget online. It helps you plan the budget and post it online. It’s roughly a $10 million business, grows very nicely, and had never raised capital before we came in. It had never had a salesperson.

We came in, brought in a new CEO, brought on a CRO, and partnered with a guy who had built a $200 million gov-tech business. My point here on the AI stuff is that they have, I don’t know, 10 or 12 software engineers. They can use companies like Cursor and Copilot to help their 10 engineers become 30 or 40 engineers. You’re going to see this at Salesforce. You’re going to see this at Workday. The incumbents usually win.

Since the iPhone came out in 2007—or 2006, whatever it was—there have only been 3 companies built that did not exist before that became $100 billion companies: ByteDance, Pinduoduo, and Uber. Who won? Facebook, Google, Microsoft. Incumbency wins. It’s customer distribution.

4. Why it is Comical to think there will be $BN one person companies?

The idea of a single-person AI company—I think that’s comical at best. Why? Unpack that, because everyone is saying, “We’re going to have billion-dollar companies with 1 person.”

These software companies are not that complicated. This isn’t rocket science or technology that people are solving. It’s sales, distribution, GTM, regulations, and go-to-market.

Exactly. It’s that kind of stuff.

5. One Question Every LP Should Ask Their VCs

If I had a dollar for every time somebody said to me, “Microsoft’s just going to do this,” I would have never invested in any software companies, and nor would anybody else. The great thing is that people ask us, “You must run out of companies to call every Monday morning.” New companies come in every Monday that we’ve never heard of.

When Marc Andreessen said 10 years ago that software was eating the world, I thought, “This sounds crazy.” But he was right. It’s changing every sector and every industry.

I could be totally wrong, but when you look at technological trends over the last 50 years, people always overestimate them in the near term and underestimate them in the long term. AI is going to completely revolutionize the world over the next 10 to 20 years, but it’s not going to be because we create a new call-center software company.

There’s going to be some type of company that AI enables that nobody else could do before, and that changes things. It’s going to be guys like you, Benchmark, or Sequoia that find that thing at the very early stage. My guess is that it is not just some infrastructure software company that the world knows about right now.

I completely agree with you. I want to take it in turn because you mentioned Gravity, this company that you said is around $10 million a year. I’m intrigued, because it’s a very different world from the one I inhabit. What does that deal look like in terms of price?

We’re able to buy businesses. We’re able to buy bootstrapped companies. I think we bought the business for around $50 million or something like that. We own the company. It grows around 50% to 60% a year now. By the way, it will never be an IPO in a million years. It will never be an IPO. We want to build a business.

I’ll give you an example. We just sold a company called SafeSend. It makes a verticalized version of DocuSign for tax returns. There are a bunch of reasons DocuSign isn’t very good at it. It also has the tax organizer that people get, asking, “Did you get married this year? Did you have kids? Did you move?” and all those sorts of things.

When we invested—we bought around 60% of the company in 2021—my partner Nimi did the deal, along with my partner Brian. That business was introduced to us through cold calling. It was based in Ann Arbor, Michigan. It was a bootstrapped business that had never raised capital and had been around for 6 or 7 years.

It was COVID-enabled. What do I mean by that? Before COVID, a bunch of people used to literally go to their accountant’s office and sign their tax returns. That sounds totally insane, but after COVID, you couldn’t do that. It was all electronic. It turned out that it stayed COVID-enabled, unlike a virtual-events company like Hopin, where during COVID people couldn’t go to events, so they went to virtual events.

It turns out people like to go to Las Vegas, drink beer, and get away from their husbands, wives, and children. Everything went back to Las Vegas. At these events, you’d never have gotten a DocuSign and said, “I’m sorry, please send me a paper copy.” It was the reverse. This thing was cool.

We invested in that business at around $13 million of ARR. It was growing around 50% to 60% a year. It was a control deal. We were buying 60% of the company, and we bought 60% for around $90 million of equity and $20 million of debt. What is that? I don’t know, a $130 million to $140 million valuation.

In the middle of 2021, that was insanity. Had that deal been backed by Benchmark—by Bill Gurley or Peter Fenton—doing a minority deal based in Silicon Valley, it would have been a $500 million deal.

Yeah.

Let’s go find stuff where we don’t have to play the same game. Let’s find the boring stuff that’s not going to be the next Snowflake or Datadog. Let’s find stuff that we can just build. Invest in $10 million- to $20 million-revenue software businesses and exit them when they’re $60 million- to $80 million-revenue software businesses.

In 3 and a half years, we built the business to around $47 million of revenue. It was very nicely profitable, and we sold it to Thomson Reuters for a great return. It was publicly announced what it was.

6. Why TikTok Does Not Matter to ByteDance and It Is a Screaming Buy

That was a business where, if you had read our investment memo, the word “IPO” would not have come up. We were saying, “We’ll grow it from $13 million or $14 million to $60 million, $70 million, or $80 million, and we’ll sell it to a mid-market private equity fund because it has 90%-plus gross-dollar retention. Or we’ll sell it to a strategic.”

If that deal had been backed by Benchmark, Bill Gurley, or Peter Fenton, it would have been a $500 million deal.

Exactly. We paid $130 million to $140 million for it in the middle of 2021. Let’s go find stuff that we don’t have to play the same game with.

The immediate response to what you’re talking about would be that your founders aren’t generationally defining founders.

100% correct.

Bill Gurley, Peter Fenton, or Benchmark would say, “We have to back founders who reshape categories—true visionary innovators.” Here, you’re talking about a control deal where you’re bringing in a team, which is amazing, but it’s a very different scenario.

That’s totally fine. I don’t have to do that. We have these 8 criteria. Some of them are generational companies.

Look, we were buying ByteDance late last year. We were paying 5 times earnings for it. It grows around 25% to 30% a year. That’s probably a generational company. It’s already a gigantic company, but it meets the framework of what we do.

We own a business called ExaGrid that was started in 2002. We own around a third of the company. It last raised money 15 years ago. We bought out Lehman Brothers. It’s a $165 million to $170 million-revenue business that competes with HP and Dell in storage devices.

It’s a 70% gross-margin business. It did $26 million of EBITDA last year. We bought our stake at a $130 million valuation. I’m going to build it into a $250 million-revenue business doing $70 million of EBITDA, and I’m going to sell it for 10 times EBITDA and make 4 times my money.

No, it is not generational or category-defining. But it is technology investing, making a really good return.

7. WTF Happens To The Cohort of SaaS Companies With Slow Growth, Not Yet Profitable and $50M-$200M in Revenue

I’m really worried, because we’re seeing the shakeout now. I’m getting old. You said that I was 32.

I’m 28, Harry, just to clarify that.

No, I meant the guy from Spark.

Thank God. I was thinking, “I’m not that old.”

We’re seeing this generation of SaaS companies that raised a lot of money but aren’t profitable. Growth is in the mid-teens yearly—10% to 20% growth—and revenues are $50 million to $200 million. They’re not big enough for private equity, but they’re living-dead companies. What happens? Teach me.

There’s a fundamental problem that happened. We saw this at Bessemer as well. Back then, you would basically exit a company through an IPO or a strategic. Those were the 2 options for how you could get out of a company.

If the IPO window didn’t show up, and you got to $50 million, $60 million, or $70 million of revenue and stopped growing, you’d think, “What the heck do I do?”

In the 2008 to 2010 time frame, mid-market private equity firms like Nautic Partners, GTCR, and Charlesbank would buy industrial companies, manufacturing companies, and services companies. Some bought consumer and some bought healthcare, but none bought software. That was just as Vista, Thoma Bravo, and Francisco Partners were starting to get going.

Fast-forward to today. Those big software private equity firms have gotten very big. You also have mid-market private equity funds that still buy industrial companies, manufacturing companies, and services companies. Their portfolios grow at GDP plus 2%, so if I bring them a company growing 15% a year, they think that’s really fast for them.

Now, it’s not 100%, but 50% to 60% of these mid-market private equity firms also buy software companies. They have a sleeve to do software. If you look at all of our exits, a third of our exits have actually gone to private equity.

I’ve got some companies in our portfolio—don’t worry, we did some real stupid stuff in 2020 and 2021, as everybody else did as well.

No, exactly. None of us were perfect.

You’ll have companies that have $130 million of revenue, 18% growth, don’t burn that much money, and have $130 million of cash. These companies effectively went public in 2015 and 2017. When companies went public, they’d raise $100 million to $300 million.

In 2021, they would go and raise $100 million to $300 million. These companies completed IPOs without actually going public. You have to get them to—I mean, we’ve drilled into a couple of our companies that have this situation—the rule of 40, because that’s the only way you’re getting out.

A strategic isn’t going to come and buy you, and this company is never going to go public. You need to get it to the rule of 40 to either sell it to a private equity fund.

By the way, I’m sorry, but the last round was $3 billion. You’re $120 million of revenue growing 18% a year. If we can get it to the rule of 40, you might be worth 5 times revenue.

You can try to pivot all you want. There are so many VCs who like to waste their time on company boards. We don’t understand it. We’re just saying, “Yes, we have a preferred, so we would get our 1x, but if you told me today I could take a 7x just to get out of it, I would happily do it.” I would happily do it.

8. What is the Biggest Problem with the IPO Market

There are hundreds of these companies out there. The problem with the IPO market is that the IPO market is actually totally fine. If you look at IPO performance, companies have actually done pretty well versus their opening-day prices. Look at Reddit and some of these other companies.

The issue is that the good companies—Grafana Labs, Databricks, Stripe—they have so much money and so much cash that they don’t need to go public. You have this whole other sector of companies that can’t go public.

I’m going to make a contrarian statement: I do not think that in 5 years the majority of companies that could go public will go public. I think being public will be an unfortunate consequence of scale, and that’s really bad for the venture capital industry and LPs if we don’t have a very developed or mature secondaries market.

Do you think guys like Don Valentine, Mike Moritz, or John Doerr would be putting guns to these founders’ heads today, saying, “You need to go public. Don’t be afraid of the 27-year-old HBS analyst. You’ll be fine”?

We actually think it makes companies better. I get both sides of the trade.

How does it make them better?

likely John Collison said the other day, “If you’re a public company or a CEO, and you think that having an analyst at Goldman saying, ‘You need to improve your margins,’ is going to increase the discipline within your company, then you clearly do not have a great company.”

Nobody says you have to be a public company. Chanel, Tata, Amway, and Koch Industries are all big private companies. You do not have to be public.

9. Quick Fire Questions

But I do think that if you take venture capital money from people, you should be very clear: “I don’t think we’re ever going to be a public company.” I think the Stripe guys were saying very early on that they didn’t want to be a public company. If you invest in us, just know that we probably will never be public.

If you’re very open and honest with investors, I think that’s totally fine. Where I do think being public can help is that there are 2 different ways of looking at it.

The quarterly cadence of public companies is a little nonsensical. However, if you ask Mark Benioff or the Google guys whether being a public company made them more disciplined, or made them prioritize one thing over another, they’d probably say that it did. They might say it was a necessary evil. They had investors who wanted liquidity, and they needed investors to get out.

I do think a company like Zoom went public because it’s a very profitable business that was growing fast. They went public because the private companies they competed with were constantly saying, “Zoom is a tiny business. Why do you want to use them?” They used that as a negative. Once you’re public, you can say, “Okay, go look at our balance sheet.”

The quarterly cadence is a little ridiculous, but some companies don’t care as much about it. Their stocks are going to be more volatile. We have no problem with companies that go public and want dual-class listed stock. I’ve got one in TransferWise.

Do you think that private-market investors and venture investors are advantageously positioned because of asymmetric information and historical information to manage the book once it goes public? In other words, is roll-up in the evergreen fund the right structure, or do you think it should be handed over to LPs, who are as well positioned?

It depends on what you tell your LPs your mandate is. I believe most private-market investors are very good company pickers, but a good company and a good investment are 2 fundamentally different things because of valuation.

When a private company goes public, if you’re an early-stage investor, you should get off the board and sell the company. I suspect that’s what you’ve told most of your investors you do. You spend all your time picking small companies and growing them into big companies. When they go public, get off and call it a day.

We are relentless in our focus on trying to make 2x to 5x in 3 to 7 years. When we do it, we move on to the next company. Our business isn’t trying to get 100-baggers while having a bunch of zeros. That’s just not our business.

We feel that if you can cut the downside scenarios—the zeros—you can generate really good returns. People have always credited us with very high DPIs, and it’s just a relentless focus on selling.

I’m enjoying this so much. You said “a relentless focus on selling.” What have been your lessons from that, and what does it really mean?

We have a disposition committee at Lead Edge. We meet and look at the portfolio, asking, “How’s the company doing?”

An investment committee is where you talk about companies that you want to invest in. You analyze whether you should invest in the company. A disposition committee is exactly the same thing, just in reverse. We’re already an investor in the company. How should we think about getting out?

Is there a secondary? Can we find a secondary? Is there an early investor who might want to buy more of our stake? Is there a crossover hedge fund that would want to buy our stake? Why might we want to sell? We might think the market size is too small, or we might have lost confidence in the management team. There could be a whole range of reasons.

We think there are a lot of really good funds that are really good at investing. We think there are a lot of people who are not very good at selling. By the way, I might blame LPs for this just as much as GPs. LPs have to hold GPs accountable.

One of my longtime LPs refers to some private equity firms and VCs as “pigs at the trough.” It’s like, “I ate the food. I spent all the money. Now give me more money to spend again.” They couldn’t give you money the 3rd or 4th time if you haven’t given a lot of the money back from the 1st or 2nd time.

There’s just a lot of complacency in this industry. All GPs need to do a better job of getting money back to LPs and figuring out how to do it.

You said “disposition committee.” I’m often told that companies are bought, not sold. Do you agree, and how do you reflect on that sentiment?

In order to get bought, you have to be known. I’ve generated a lot of returns by putting companies up for auction and selling them.

There are a lot of things companies don’t do that they probably should do. It’s very hard to get bought if strategics don’t know who you are. We encourage all of our founders to get to know the biggest strategics in the space and the private equity funds that could eventually buy them.

If you think you’re going to do $50 million of revenue this year, up from $30 million, tell them you’re going to do $40 million and then beat the number. It’s about building relationships and partnerships with people.

You mentioned duration. You said 2 to 7 times in 3 to 5 years.

We’re trying to make 2x to 5x in 3 to 7 years, which basically blends to a 20% net IRR curve.

Totally makes sense. The thing I hear there is duration. I’m a money manager at a large endowment or pension fund, and I’m okay with that duration. If we compare that to a 15-year duration for an early-stage venture firm, I can compound my money with you almost 3 times over and get that same blended 3x.

Yeah.

Or I can go to an early-stage venture fund, which is probably outsized in terms of actual fund size, and maybe get a 3x. I’ve seen the data. There aren’t many 3x funds. How does venture earn its place in a lot of money managers’ books?

I think it’s very hard. I think there are too many venture funds.

My advice to people starting early-stage venture funds is to look at what Fabrice Grinda has done. Have you ever interviewed him?

I’ve interviewed him twice. I really like him.

Fabrice has been an LP of ours for 15 years. He’s a very good friend. Those guys have figured out the game: invest in the seed or the A, and sell a bunch in the B or C.

That’s a fantastic game to play. You can make a ton of money doing it, generate DPI back to your investors, and still ride your winners. You’re providing both things.

The 15-year-duration issue is totally true, and it should scare more people. It’s actually shocking that the number of venture funds over the last 5 to 7 years has increased, given that exits are getting longer, not shorter.

What I believe emerging managers and people starting venture funds need to do is take advantage of the secondary market and the fact that these growth funds have gotten so big. There are crossover hedge funds and public funds that want to get into private investing. Start selling off stakes.

Do the seed, do the A, and sell some in the B or C. You’re not selling the whole position. Just start returning money to people.

I get you. I just had a GP on from Emergence who outlined the different fund returns they’d had if they had sold or not sold positions. They sold their Salesforce position at IPO, and had they not, they would have made another—I’m probably butchering it—$20 billion to $30 billion in gains. Bessemer sold its Shopify position at IPO and lost billions.

My point is that if venture is a power-law game, that’s true.

Yeah, but you’ve got to stay in business. I think the faster a venture fund can get to a 1x, the faster that fund can probably grow its assets quite a bit.

I’m also talking about emerging managers who need to stay in business and raise funds 2, 3, and 4. They’re not people like Bessemer, which has been in business for 80 years.

For every Shopify, ask them about 1999 and 2000. Ask how many billions of dollars were lost in 2020 and 2021 by not distributing positions.

You mentioned doing stupid stuff in 2021 and 2022. What was your most stupid mistake, and what did you learn from it?

Our stupidest mistakes were overpaying for a couple of companies and assuming the exit multiple would be higher than it actually was.

I credit my partner Nimi, who has been with me since the beginning. In 2018 or 2019, we really started to shift our business away from Silicon Valley-based companies and from thinking every company needed to IPO.

We started saying, “Go find these Gravities. Go find this SafeSend. Go find companies that are not going to be the next Snowflake or Datadog.”

What caused that shift? I go through your fund ones, and it’s Alibabas, Spotifys, Ubers—fantastic companies, but all venture.

It was caused by looking around and saying there was no possible way every one of these companies could grow to be as big as they were. The law of compounding, when you’re investing at a $1 billion or $2 billion valuation, is just harder. It’s the law of large numbers.

We also looked around and asked, “Who has money?” Mid-market private equity funds. There are hundreds of them. None of these guys used to buy software companies. They’re now starting to buy software companies, so there’s fertile ground.

For us, we can go buy a company for $40 million with $10 million to $20 million of revenue, growing 40% a year. We can grow it, 2x to 3x the revenue, and then it will be growing 15% a year. That’s fast for these buyers. We can run an auction, sell the business, and get 20 people to bid for it.

This is my point: Those companies growing in the mid-teens with $50 million to $100 million in revenue have an exit. They just need to pivot the business and realize they’re not building the next Datadog or Snowflake.

They need to get to the rule of 40. The most important thing in getting to the rule of 40 is high gross margins. If you have high gross margins and 90%-plus gross-dollar retention, it’s a shame. You can let the private equity fund try to sell it when you’re losing money or breaking even.

What some of these growth equity firms and venture funds should do is do it themselves. It’s not that complicated. If you have high retention rates, make the hard decisions and get the thing to profitability. Turn it into the rule of 40.

You said one of the mistakes was that you paid up for things a little too much. How do you determine the difference between a stretch on price and a mistake where you stretched too far?

We build a 5-year model. The model is wrong, but we try to put a reasonable exit multiple on it.

Is that valuable to do? Exit multiples vary so much over different durations. If we look back at 2021 and 2022, the multiple would have been so much higher than today.

You need to use something reasonable. For software, revenue multiples are just completely out of hand. At the end of the day, if you build a software company, and it’s growing 15% to 30% a year, you should assume an exit multiple somewhere between 4 and 7 times revenue.

Our bands are generally 4 to 8 times revenue, and maybe sometimes 10 times at the absolute highest if it’s growing 30% to 40% a year.

I credit the guys at Iconiq with a huge amount. They were underwriting deals in 2015, 2016, and 2017. They thought they’d exit at 10 to 12 times revenue, so they bought the best assets. They maybe paid 20% higher to get access to the best assets, and then multiples went to 20 to 30 times revenue.

The best way to make 4 times your money is to 2x the revenue and 2x the multiple. The reverse happens too. That’s what happened to all the stuff in the 2021 vintage funds. Multiples got cut in half.

If you 2x the revenue and halve the multiple, that’s called a 1x.

What is the stupid stuff we’re doing today that not many people are talking about?

We’re paying 100 times revenue for companies.

We like to ask ourselves, “If I invest today, grow it for 18 months, and assume it’s still growing fast, am I in the money? Or do I need to grow for 4 or 5 years until I even get in the money?”

Toast is an example. When we invested—I think it was in 2016 or 2017—it was around $25 million of revenue, growing 250% a year. What would that be? A billion-dollar-plus exit. That would be a billion- to $1.5 billion revenue multiple today.

We paid 20 times revenue. It was a $500 million valuation. That was expensive, but we thought, “In a year from now, we’re probably in it at 10 times revenue.” For that growth rate, that’s pretty reasonable.

I encourage people to ask, “I pay this price today. In 12 months, am I still in it at 80 times revenue or 50 times revenue?” The prices being paid are totally insane.

I also think not enough investors focus on gross-dollar retention. I’ve had numerous entrepreneurs tell me, “I don’t look at gross-dollar retention. Net is the only thing that matters.” I think, “Oh, really? Okay.”

If you’ve got a business that ended 2024 at $10 million of revenue, and you say, “I have 50% gross-dollar retention, or I have 200% net-dollar retention,” you might say, “My $10 million became $20 million from existing customers.” But if you only have 50% gross-dollar retention, you actually lost half your customers. Yes, you had a few that really liked it, but a huge amount of the business was experimental.

When you’re a really small company, the difference between 90% gross retention and 50% isn’t that much because the pond—or the bucket—you have to fill isn’t that big. But when you get to $100 million, $200 million, or $300 million of revenue, that creates a huge hole in the bottom of the bucket.

That leads to sales and marketing inefficiency. Your burn rates are much higher. Not enough people focus on it. We look at a lot of these AI software companies, and their gross-dollar retention rates are really, really low. It’s actually shocking.

Pretty much all of them?

Not every one of them. Lovable got 85%, which is pretty good—better than ChatGPT. But again, it’s not 90% gross-dollar retention.

We have a business in our portfolio that will not change the world. It makes cardiac-monitoring software. It’s a very small market, but it has 99% gross-dollar retention. That means you can run the business very capital-efficiently over time because you don’t have to keep spending money on more and more sales and marketing.

How do you think about that capital efficiency in the future-dilution element when investing? As we said earlier, Uber and your Alibabas are incredibly cash-consumptive businesses, while many of these newer businesses are incredibly cash-efficient lean machines.

Alibaba was actually very cash-efficient when we invested. It was a billion dollars of profit. There may be fewer shares of Alibaba. I credit Jack Ma and the team at Alibaba for actually managing stock-based compensation dilution well. The amount of stock-based-compensation dilution in a lot of public companies is totally crazy.

Most people, including ourselves, massively underestimated over the last 15 years the amount of stock-based-compensation dilution and dilution generally that we all took. Uber was totally insane.

Over the last few years, we’ve dramatically increased the amount of dilution we assume. We assume 20% to 30% dilution, and if you’re investing earlier, it could be a lot more than that.

We have what we call capital efficiency. Warren Buffett would laugh at us because it sounds kind of stupid, but it works. Are your revenues today greater than your historical cumulative cash burn? Not raised—burned.

If you’ve raised $80 million but only burned $20 million, and you have a $40 million-revenue business, that’s a great business. We’re looking for a 1-to-1 ratio or better. We just think it speaks to so many qualities of the business.

We were lucky to be investors in Benchling. We’re still investors. When we invested—Benchmark did the early rounds, and Thrive was an early investor—it had around $13 million of ARR and was growing well north of 100% a year.

That must have been expensive.

It was. I think it was in the low 3s. It was expensive. However, the company had only burned around $10 million. It had burned very little capital. Why? It had amazing gross-dollar retention rates, and the CEO thought a lot about the value of a dollar.

When it got to $20 million, it had not burned $20 million to get there. They valued the value of a dollar.

We have definitely missed businesses that were not capital-efficient. It just wasn’t for us.

What business most sticks out to you with that in mind?

Snowflake, massively. I mean, when we looked at Snowflake, it had horrible gross margins. Again, we looked at it and were completely wrong—100% wrong.

How many businesses actually have that perfect profile that fits yours when you look at Ubers and DoorDashes?

We looked at DoorDash in the past. It’s fine. Again, we’re going to miss stuff. That’s fine. We have a framework.

Of 10,000 companies, 100 will meet all 8 criteria. A lot of those valuations would be totally insane, if you can even get into them or if they even want to take money. Of 10,000 companies, 10% will meet 5 or more criteria.

We live in such an interesting asset class because it’s one where the supply chooses the demand. In other words, the company chooses the source of capital in a lot of cases.

My question to you then is this: We find one that meets all the criteria, but they then have to choose you. If they meet all the criteria, they probably have a lot of options. With absolute respect, Mitchell, you’re not as romantic around company creation in the way that a lot of Silicon Valley VCs are, or the way other investors are. The founders resonate with the kind of financial founders you want.

Our pitch is very simple, and it’s the reason that a huge number of our founders have invested in our funds post-exit.

The founder of Duo responded and then DM’d me after I tweeted about having you on. He said, “Amazing, amazing. Now I’m an LP. It’s even more amazing.” It was cool to see.

What do we do? Why is that?

When we started Lead Edge, my partner Nimi and I sat around—Brian hadn’t even joined yet—and asked, “Why would we take our money? Why are they going to take our money?”

We thought, “If we were the global head of HR at Pepsi, and Nimi had been the global head of HR at Microsoft or Dell, we could probably cold-call HR software companies and say, ‘Let us invest in your business, and we’ll introduce you to a bunch of global HR executives.’ That would probably be believable.”

We didn’t have that. We had never been the global head of HR of anything. We had never done anything except cold-calling analysts.

We thought, “How are we going to get into companies?” We decided to make our competitive advantage our LPs. We raised money from world-class executives and entrepreneurs.

If you look on our website, 80% to 90% of our LPs give us permission to list them. These are people who run and have built some of the world’s largest companies: the former CEO of Charles Schwab, the former CEO of Kimberly-Clark, the former CEO of Colgate-Palmolive, and others.

We go to companies and say, “If you invest with us, we’ll give you access to our LPs, who have built, run, and advised some of the world’s largest companies.”

If you’re a software company that sells into the pharmaceutical and biotech space, and you sell R&D software, we’ll say, “Would you want to meet the former CEO of Pfizer? Would you want to meet the former CEO of Biogen?” We introduce them during our diligence process. They act as our own version of McKinsey and help us. That’s how we get into deals.

Great. Sounds great. Respectfully, though, that’s a model used by a lot of people. I have around $5 billion of founders in our funds. They’re less traditional company founders and more software founders, but I use that as a weapon, so to speak, to win. Founders say, “That’s great, but Index has that too, and Sequoia definitely has that too.” All the great firms have this armory or weaponry of great entrepreneurs who invested with them.

To what extent is that differentiating?

A lot of theirs aren’t entrepreneurs. They’re executives at plain, boring, vanilla non-software companies.

We constantly leverage our LPs. We can track introductions. We track every introduction.

Everybody says they help. Very few people do. I think there’s a reason that 80-plus former and current portfolio-company executives whom we’ve backed are investors in our funds.

I don’t know. Everybody says they help. Some people probably do more than others. We just do what we say we’re going to do. It’s not that hard.

I love that. You said you track introductions. You had a very viral tweet—I can’t remember who did it. I think it was Pitti Dashi?—that tweeted your criteria around the hierarchy of bullshit.

How CEOs lie to us.

We’ll go into how CEOs lie to us, but one of the points was that when funds are fudging numbers, they track introductions. What does that mean?

We believe in it. There’s a reason Dug Song, who sold Duo for $2.35 billion, put money with our fund. We drove tons of introductions. There’s a reason VCs from some of the biggest venture funds on the planet are longtime investors with us. They’ve seen us in action.

I don’t know why more funds that say they have these incredible networks don’t help people more. I have no idea why.

I have a feeling a lot of funds scale, and it’s also that a lot of people are just not as proactive. A lot of our LPs are executives who are plus or minus 5 years away from retirement, or 10 years away. They’ve been retired for the last 5 years, or they’re about to retire in the next 5 to 10 years, and they want to help.

A huge number of our LPs are not from Silicon Valley. We find them a company and get them involved early in the process. It’s how we do diligence.

If you’re a payments company—we invested in Wise—we very early on, before investing, said, “Would you want to talk to the former CFO of PayPal? Would you want to talk to the former president of Visa?” Any good entrepreneur will say, “Yes, those sound like interesting people.”

Then we call them and ask, “What did you think? How do you manage that?”

How do you manage that from an infrastructure perspective? That sounds amazing, but it’s difficult.

It’s difficult. Every introduction we make—I wish I could tell you it’s automatically logged—isn’t. We BCC an assistant, and it gets logged in Salesforce. It’s a highly customized version of Salesforce that we’ve spent millions of dollars customizing over the last 15 years.

Do you have heads of network, network managers, or community managers?

We do not. Every person on the investment team has access to all the LPs.

If you’ve worked at Lead Edge for 2 years—or even a year and a half—and you’re going to Seattle for a wedding and say, “I want to stay on Monday,” we’ll say, “Wonderful. You should meet these 4 LPs, and we’ll pay for your plane ticket.”

We encourage everybody at the firm to get to know our LPs and spend time with them. If I were a vice president at Index—I could pick any firm on the planet—and I were flying to Seattle to meet a company, I’d go meet the company, maybe meet another prospective company, and then fly home. I wouldn’t spend 6 hours meeting 4 other individuals who are LPs.

That’s not their model. These funds are primarily backed by the largest endowments and pension funds in the world. Our model is that we’re 95% backed by individuals, and we treat those individuals like gold. We communicate with them.

10. Why LPs are More Important than Founders

A lot of VCs today say, “Founders are our customers. LPs are not our customers.”

I would tell you that we have 2 customers: founders, but more importantly, LPs. If you do not have LPs, you do not have a business.

I 100% agree. I think it’s really arrogant to suggest LPs are not your customers. There are 2 customers, as you said: founders and LPs. I’m astonished by this unwillingness to recognize them as customers that we have to provide a great product for.

We run our business trying to figure out how to have 97% gross-dollar retention with LPs—not net, gross.

What do you do? We communicate with them. We do lots of events. We do quarterly calls. We walk people through the portfolio. We tell people how things are doing.

Some companies will be doing well one quarter, and some companies will be doing badly one quarter. When you grow 30% a quarter on average, some will be doing well and some will be doing badly.

The lack of transparency in this industry toward LPs is shocking.

To what extent do you think your re-up rate is determined by engagement and communication versus performance?

It’s both. Without good performance, you can have none of it. But people want the nice guy, the good guy, the person who communicates, to win.

If you have 1 or 2 bad vintage funds, I think LPs are more likely to stay with you.

That line doesn’t quite add up. You said they won’t come back without performance.

Wrong. They do. We’ve seen it.

There’s a very well-known, very large private equity fund—it’s not in software—that historically raised funds and literally told LPs, “You have 3 weeks to get your documents in. Take it or leave it. These are the terms. We’re not changing anything.”

They had a bad fund vintage. They’d been in the market for 3 and a half years. The fund they just closed was a fraction of the size of their previous fund. I asked their LPs what the problem was. They said, “You know exactly what the problem is. They’re jerks.”

If you’re an LP, they give you the money. Without you, the GP doesn’t have a business. I don’t know. I get you.

We’ve seen a lot of fund returns or performance numbers leaked in recent months. I’m not going to name the funds because I don’t want to publicly single them out. Their numbers have been poor, to be blunt—mid-teens IRRs at best for early-stage funds.

They’ve scaled AUM and had excess supply of LP cash. Is performance even relevant?

That’s why LPs are to blame too. I have many LPs come into this office and say, “Harry, I have $500 million a year. Where do I go?”

I say, “We go to these tier-one managers, but you can only get $20 million in each. There’s $100 million.”

They say, “I have $400 million left, so I have to put $75 million into a multistage fund. I just have to. It’s my budget.”

The smartest ones say, “If the opportunity isn’t there, let’s figure out where else to put it.” It’s not as if they have to put $1 billion into venture every year. Maybe $1 billion in venture is too much. Maybe they should be doing smaller funds.

There’s a guy, Eric Cush at Mercer, who runs research there. He took a bet on us very early, before Fund 3. People laughed. Our first institutional investor in Fund 2 was the University of Virginia.

We literally started the meeting by saying, “You’re not going to invest with us. Why would you invest with us, you knuckleheads? We don’t have a brand. We have nothing.”

A lot of investment consultants are very brand-name-focused. You don’t get fired for hiring IBM or McKinsey. I credit Eric Cush. He took a flyer on us very early, and he’s done it with a bunch of other young managers.

He digs into it: “I’m in Fund 3. Why did you exit these things in Fund 1? How do you think about it? How do you think about returns? How do you think about treating LPs?” He talks to portfolio companies and asks, “How do you actually add value?”

We tell prospective LPs, “Everybody tells you they help companies. Wonderful. Let me introduce you to 10 companies in our portfolio. Call them as many times as you want and ask them if we add value.”

I put screenshots into decks of introductions to amazing people between our portfolio companies and LPs. You can see the response. It’s tangible value.

Great. I totally get that. Respectfully, when you look at someone like Andreessen Horowitz, they’ve proven that brand is more important than performance at scale.

We’ll see what happens over the next 10 or 20 years.

Does that make you change your perspective on the importance of brand, or do you say, “Fuck that. We’ll stay in our lane and do what we do”?

We stay in our lane. I have a huge amount of respect for the firm. It’s probably one of the best-returning firms in technology investing.

There are a couple of them. Spectrum Equity is an incredible investor. They’ve kept $2- to $2.5 billion funds forever. Benchmark’s fund is smaller now. Benchmark has had Generation 1, Generation 2, and now Generation 3, and they continue to put up really good numbers.

Josh Kopelman at First Round has stayed small. Mike Maples and Ann Miura-Ko at Floodgate have stayed small. The firm that I think has generated the best returns in the technology-investing world over the last 34 years is TA Associates.

It’s become more of a buyout fund, but they used to do tons of minority investments. It’s gotten gigantic, with huge funds, but the DPIs they’ve put up are incredible.

What do you think Spectrum does that makes it so good?

They have a relentless focus on liquidity. A lot of the stuff they do in the buyout world now involves minority sales. They’ll invest in a company and, 2 years later, sell 30% of the company to somebody else and get their basis back. They’re already at a 1x.

They have a relentless focus on liquidity.

You mentioned earlier that one of the things you admire is that relentless focus on liquidity. Doug Leone said on the show that we’ve moved from a boutique, high-margin community to a commoditized, low-margin industry.

I 100% agree with him.

Do you think that’s reversible? Has platformization matured venture into this asset class, or is it reversible?

The hard thing for me is that the cost of capital is so different. I lost a deal recently to one of these large multistage firms. I did 3 meetings with the founder. It was pre-product, pre-revenue, pre-everything, but an amazing founder. He called me the next day and said, “I would never walk out on a handshake, but I got offered $8 million on a $100 million valuation.”

These bad returns will do it over time, but it’s going to take a long time.

I think the venture industry was about to be in for a rude awakening, and then AI showed up. I always say AI was the oxygen that a lot of these venture funds needed.

Yes.

For a lot of these venture funds, I think that’s true. I’ve talked to a lot of legends in this industry. I love talking to people who have been in the industry longer than I’ve been alive, or who were investing in the 1990s and 1980s.

What’s going on in AI looks very similar to the internet bubble. People didn’t learn a damn thing from 2020 and 2021. It’s shocking.

Let’s not even talk about crypto, because that’s a whole other conversation.

Specifically, what should we have learned?

Entry price matters a lot. Don’t overcapitalize companies.

Bingo.

Entry price matters. Don’t overcapitalize companies. The amount of stock dilution really matters. If you run a fund where you need IPOs, you better invest in founders who want to IPO their companies, or have a good plan for how you’re going to get a lot of secondary liquidity out of it.

Entry price matters a ton. These companies are not all Snowflakes and Facebooks. The vast majority of companies are not.

I can put on one hand the number of people who are capable of backing companies like Google and Facebook and doing it more than once. It’s a really, really small number. It sure as hell isn’t me. It’s people like Doug Leone, and there just aren’t many of them.

I love Doug. We need more Doug Leones, Don Valentines, and people who are direct and outspoken.

If you were to advise the several thousand LPs who are listening about investing with managers today, what would you say?

I believe there’s a great question people don’t ask. They should ask any manager who has been around for 10-plus years:

“On September 30, 2021, how much unlocked public stock did you have in your portfolio?” That was the height of the insanity in the last tech run-up.

Then ask, “Why didn’t you distribute it to LPs?”

A lot of funds can distribute stock too. You could have kept the stock. Why didn’t you? Some people will say, “I was on the board.” Shouldn’t you have distributed it? Isn’t your goal to return capital to LPs? Isn’t that the whole job of the business?

It would be shocking how many people had a lot of unlocked stock and did not return money.

I also think LPs should spend more time talking to companies in portfolios that failed. Talk to the ones that didn’t go well or were 1x investments. Find out what the person is really like to work with.

The ones that work really well are the easy ones. I like to focus on what actually didn’t work: How did the person respond? How did they deal with adversity? How did they deal with you?

You said that if you could get a 7x back on an underperformer, you’d take it.

Fuck it. I’d take it all day long. All day long.

How do you feel about the transactional nature of where time is spent? Fred Wilson is an incredible investor—an incredible investor—and he would be in that top 5. He always says reputations are made in the bad companies. But then you also realize that you have a limited amount of time, and you have to manage a portfolio and invest in new companies.

Is it possible to cut your losers elegantly so you can focus on your winners?

It’s a lot easier for me to do it than it is for Fred, because he was there when it was nothing. I came in when it was a bigger company.

There are some VCs who are world-class VCs and cut their losers. There are firms that are known to say, “If you’re a CEO and you don’t perform, you probably won’t be the CEO.”

If you’re a founder coming in, you should just be direct. I tell all my employees, “If I’m not the right person to run Lead Edge, throw me out. That’s fine. Or put me on the side and you come run the business.”

You mentioned ByteDance earlier. We had the head of private investments from Baillie Gifford in yesterday. They’re big in ByteDance.

There’s a lot of public concern in the United States—or excitement, depending on which side you sit on—that TikTok will be shut down or divested. You’ve said before that you’re not worried about that.

When we underwrote ByteDance, we assumed the US business was worth zero. ByteDance North America is a single-digit percentage of revenue. We assumed it would be shut down, and it isn’t profitable in the United States.

Then we saw what happened when they shut it down for a day. Both sides of the aisle came running up with their bags, saying, “Please keep it open.” I just want something to happen. Either shut the thing down for good, spin it off, or do something.

I suspect it won’t be done by April 5—or whatever the date is. They’ll probably push it out again. I don’t know. I think there are 4 or 5 people in the world who know what’s going to happen.

I do think the Chinese government really likes ByteDance. It’s a truly global business. Alibaba is not a truly global business. Tencent is not really a truly global business. Nike is a truly global business. Microsoft and John Deere are global businesses.

What I mean is that you can go into 140 countries around the world and buy Nike shoes. You can probably go into 100 countries and buy a John Deere tractor. China has always wanted to build a truly global business. This is the first truly global business.

If you think they’re just going to let it go, I think they’re very proud of what they built. It’s a huge business. I think they’re going to be one of the foremost AI companies on the planet over the next decade.

What makes you say it will be one of the foremost AI companies?

The amount of technology they have. There’s a reason for the amount of AI they’ve already embedded in the product.

In India, when they were kicked out several years ago, nobody has really been able to build a competitor. Facebook is trying. By the way, do you know who hates ByteDance? Mark Zuckerberg.

If I were running Snapchat, Instagram, or Facebook, I would be all over politicians in Washington saying, “This is horrible. This is propaganda. You’ve got to get these guys out of here.” It’s the biggest threat to those companies.

It’s absolutely incredible what the Chinese have built. If you look at the number of PhDs, science spending, and all these other statistics, China is an incredible country. They’re not worried about what happens next quarter or next year. They think in 50-year blocks.

11. Why We Drastically Underestimate the Power of Chinese AI?

Do you think we in the West underestimate China’s ability in AI?

100%. I’ve seen how hard people in China work at some of these technology companies.

I agree with you. As a result, I get concerned. Do you get concerned by their ability to infiltrate our societies and provide amazing products?

I think both countries should learn to get along. China and the United States collaborating more is better for the world than collaborating less. There are lots of things both countries can do together to make both countries better places.

Peter Gao at Baillie Gifford taught me about the strength of the core ByteDance business in China. I know it’s a global business, as you said, but the core business in China is a monster.

It’s a monster. It’s also a huge e-commerce business there. It’s unbelievable.

We all think, “TikTok is shutting down in the US. It’s over. It’s terrible.” Not really.

My question, though, is that if it is more domestically focused, I don’t understand where the liquidity comes from. It’s clearly not going to list in the United States.

Hong Kong. Hong Kong, Hong Kong, Hong Kong.

You think that’s feasible?

100%. Tencent is a huge company. You can list in Hong Kong. There are giant Asian businesses listed on the Hong Kong Stock Exchange. It’s very liquid.

You could have a trillion-dollar ByteDance. By the way, I don’t know what Meta’s market cap was yesterday, but I think it’s a $1.5 trillion to $1.7 trillion company. ByteDance is the same size of business in terms of earnings and grows faster.

Alibaba and Tencent don’t grow that fast. They’re 5%, 7%, or 8% growers. What do they trade at? 13 to 15 times earnings. You do the math. This is a very, very big company.

We like to buy things when no one else likes to buy them. When the world hates something, it’s interesting.

Doesn’t that go against your statement that the best founders are the ones who don’t call you back?

That’s fair. It’s a fair statement. I’m not always right.

I struggle with that statement too, because I have so many great entrepreneurs in here: the founders of UiPath, the founder of Klaviyo, and many others.

Tobi at Shopify gives me trouble on that one. Fifty VCs said no to him.

Actually, they were calling him back.

They were calling him back?

For every Tobi at Shopify, there are 100,000 founders who aren’t going to make it. But it’s incredible what he built.

Can I ask a final one for you? A quick fire. What’s your favorite deal you’ve done? You’ve done many deals. What’s the one where you think, “That’s my favorite,” and what did you learn from it?

It depends. My favorite deal might be buying LP positions out of a 20-year-old fund, buying something at 4 times earnings, or finding something and getting 4 times earnings.

You buy fund positions?

100%.

There’s a table in front of us. Let’s say this table is a company. If I buy 10% of the table or 60% of the table, it’s the same table. If you own the table, and the chair you’re sitting on is owned by the fund, and the fund owns the table and the chair, I can buy the chair. I just bought 25% of the table.

We view investing in companies this way: We have specific criteria, and then we take a completely flexible approach.

We’ll buy 10% of your company. We’ll go through the front door. It’s like walking down the street and seeing a 3-criteria house. You walk by it. You see a 4-criteria house. You walk by it. You get to a 6-criteria house and knock on the front door.

You can go in the front door and buy around a minority stake. You can go in the front door and buy the entire business. What if they don’t want to raise capital? You can go in the side door and buy out an early investor or early employee in the form of a secondary. It’s still the same house.

Let’s say that doesn’t work. You can’t buy secondary. There’s no seller. For the whole history of Lead Edge, it could be a roadblock, or there could be no seller. Let’s go through the basement window with a pickaxe and find a derivative.

Let’s fund somebody’s co-investment vehicle. Let’s find a 20-year-old fund where 90% of the NAV is in 1 company.

The company I’m going to get to is Workhuman. It’s an awesome business in Boston. It’s been around for more than 20 years. It’s insanely profitable, although it doesn’t grow 50% a year. It’s a giant, stable business.

We met the company, but there was nothing to do. The company didn’t need primary capital. It was super profitable, and there was no secondary to buy.

We found an old fund that was 17 years old. We tendered, and we went to the founding partners and said, “The fund must have a bunch of LPs who want out. You’ve been in it 16 years, right?”

We bought the position at around 5 times earnings. We’ve gotten 90% of our money back through dividends from the company.

How did you do that?

The LPs had been in the fund for 15 years. They just wanted out.

That’s a unique opportunity. It’s an arbitrage on timing.

Correct.

The most special time is when you have a manager who desperately needs liquidity to raise a fund. They know it’s an inopportune time to sell, but they need to sell to get the next fund.

Your duration is different.

Correct.

Do you worry that everyone is doing that now?

It shocks me that more people don’t do this. There aren’t that many people doing it.

Secondary funds are doing a lot of the multi-asset stuff. They’ll buy an LP that’s selling 30 positions in 30 old funds, with 200 or 500 underlying companies.

We’re looking for the stuff where there’s 1 underlying company, it’s an old fund, and 90% of the NAV is in 1 company. We just write off the other stuff, basically.

So you buy the basket and discard the 10%.

Correct.

Have you ever been surprised by the 10%?

Yes. We got back 50% of our money on the 10% in this fund and Workhuman. It was some chip company. We looked at it and said, “The cash balance on the balance sheet is almost as big as the valuation for the entire company. This doesn’t make much sense.”

Maybe it will get something back.

We got 50% of our money back. I love that.

I want to do a quick fire. I’ve so enjoyed this conversation.

Let’s start with this: What do you believe that most people around you disbelieve?

I believe that DPI is the most important thing, and marks are for suckers.

You could buy and hold 1 public stock for 10 years—Microsoft. The pricing power they have is absolutely incredible.

Microsoft is expensive, but it’s an incredible business with amazing pricing power. I think Satya Nadella is an absolutely incredible CEO. It’s a giant business.

If we could get a 30% drawdown, I’d buy Snowflake or Datadog and put them in a drawer and let them compound for 10-plus years. Amazon goes in that bucket too. A lot of them are fairly rich.

What’s the hierarchy of bullshit that companies report?

“Favorite place to work.” If it’s on the 2nd page of your presentation, and you say, “We’re the greatest place to work,” that’s one.

“You know, in this region.” A lot of it comes down to saying, “Our revenue is this.” You look at the chart and start doing the analysis, and realize that was actually total contract value.

That’s a pretty good one too. There are a lot of ways to fudge gross-profit numbers through COGS and all that kind of stuff.

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

I was probably even more skeptical about AI.

Self-driving cars, actually. I went for rides in them, and they were incredible. It’s going to take a long time to get out there.

Ten years ago, everyone was talking about self-driving cars. People always underestimate technology over the long term, but they overestimate it in the near term. The experience I had in Los Angeles and San Francisco in a self-driving car was absolutely incredible.

We don’t shit on people on this show, but we can praise them. If you were to choose 1 seed firm, 1 Series A firm, and 1 growth firm to put your money into as an LP, which firms would you choose?

For traditional growth, Spectrum would be the growth fund. A lot of your audience is not looking at that type of growth. That’s more bootstrap growth than traditional growth.

Either ICONIQ or Meritech. The returns at ICONIQ are freaking amazing.

For seed and Series A, it would be Benchmark or Bessemer.

You don’t worry if fund size is too large?

If you need to own a large fund and write a $500 million-plus check, I think Bessemer is as good as any of the other big funds. I think the guys at Index are amazing investors too.

My problem with a lot of these funds is that you have to buy the basket. Bessemer actually just has 1 fund, which I highly respect. I guess they now have a buyout fund and an India fund, but you always have to buy the basket.

You totally agree?

Yes.

What can I ask you? I thought you were going to ask me about the short on MicroStrategy.

MicroStrategy is totally insane.

Why?

I’m not bearish on crypto, but it reminds me a little bit of the tulip craze. You can’t actually use crypto to go buy things.

If I could buy a Tesla with crypto, that would be amazing. If I could go to Amazon and use Bitcoin, it would be incredible. You can’t do that right now. Maybe it’s just too volatile for people to use as a currency.

From what I understand, MicroStrategy is effectively issuing debt to buy more crypto. They keep issuing debt to buy more. If that reverses, eventually they have to pay the debt.

It sounds like a house of cards to me. I’m not an expert on the company, and I don’t know, but some of these crypto businesses make me skeptical.

Do you have any crypto investments?

I do not. We’ve looked at things around the edges. We looked at Chainalysis years and years ago and probably should have done it. The picks-and-shovels type of stuff.

We should have done Coinbase. I should have bought Bitcoin too. I never have. Clearly, I could have made a fortune.

What concerns you most in the world today?

Two things. Income inequality, and the fact that where I grew up in Michigan—I worked in a factory in high school—I think a lot of those people today are much worse off relative to wealthy people than they were 25 years ago.

12. Why Social Media is the Most Dangerous Thing in Society

The 0.1% or the 1% has broken off. The printing of money has caused massive income dispersion.

I agree. What happens, Mitchell?

That causes revolutions, to be clear.

The thing I worry about more in the near term, and that’s solvable, is social media for teenagers. It’s absolutely horrible.

I’m an investor in ByteDance, to be clear. ByteDance is highly regulated in the vast majority of its markets. Kids in China go on Douyin to read about science experiments and math projects. In the United States and England, I assure you that’s not happening.

Social-media companies need to be regulated. There’s a reason the BBC, Discovery Channel, or NBC can’t say a lot of the things that get said on social media. They’re highly regulated by governments, and they get massive fines.

Social-media companies need to be held accountable for their content. Suicide rates, depression rates, bullying, and all these different metrics are going one way—up and to the right. Social media is the demise of society.

Australia has banned social media for people under 16. Should we ban it?

Yes, I think we should. It needs to be much more highly regulated.

Penultimate one: When have you questioned yourself most as an investor?

We questioned our existence in 2020 and 2021. We were getting annihilated on prices.

We question ourselves now. Are we totally wrong on AI? Do we just not get it? Are there going to be one-person companies much sooner than we think? Are all of our software companies going to be completely disrupted and go away?

I’m taking a stand that they’re not, but I could be totally wrong. A good investor who says they know the answer to something is probably taking the best route to failure.

You have to stay intellectually curious. My favorite idea is that every investor has a thesis. What the fuck are we, professors? I didn’t know about you, but I’ve never had a thesis.

I don’t have a thesis either.

My thesis is to meet 6 or 8 criteria. I literally don’t have a thesis.

The amount of pontification in the world, especially on Twitter, by people in the investment business is incredible. People should spend less time tweeting and more time investing.

What you don’t understand is that brand is a hack. It’s a hack to get LPs and a hack to get great founders.

I totally agree with you on the pontification.

One thing I love about our businesses is that yours is about 6 out of 8 criteria, while mine is about whether they’re a generationally defining founder.

There are 2 different ways to invest. One isn’t right or wrong.

The best advice I can give any emerging fund manager is: Define what you’re going to do, and do exactly that. Don’t stray from it at all.

Will the tourists get washed out of venture?

100%. When, I don’t know. It might be a slow hole in the canoe, but eventually, yes.

I like to finish on optimism and positivity. When you look forward over the next 10 years, what are you most optimistic, positive, or excited about in the world?

Humans are very resilient. They adapt to change.

I’m glad I became a software investor and not an energy investor. Energy has been a dying industry for the last 15 years.

Energy has never been hotter.

It’s hotter now, but it goes in waves.

I like the amount of innovation and change that’s happening. I get to invest in and meet really interesting founders. Some of them are going to change the world. Some of them are going to sell budget-planning software to the Palo Alto Police Department. That’s fine. They’re building cool companies.

When I went to Wharton Business School, I was involved in the entrepreneurship conference. We’d bring in people who built software companies and consumer companies.

I thought, “Why don’t we bring in Steve Cohen or the people who built giant money-management firms?” Steve Cohen is an entrepreneur. He started with a very small business by himself and now runs one of the biggest hedge funds on the planet. The same thing applies to others.

They’re incredible entrepreneurs. I get to spend my day meeting cool entrepreneurs. It’s not a job. It’s an amazing thing to do.

I hate the term “operator” and the way we use it, because it implies that if you’re not an operator, you don’t operate.

Exactly. I have 80 employees.

The best advice I ever got—and something I try to do every year at Lead Edge—was from the founder of Accel-KKR, which is a very good fund. He said, “You should interview all your employees, from admin to your other partners, and ask them for feedback.”

You say, “If you were running the place, what would you do differently? Tell me everything you do in your job—green, red, or yellow.”

I don’t care what you do that’s green. I care what you do that’s red, because I want to get rid of those things.

I’ve been doing that for 3 or 4 years, and the feedback you get is incredible.

What have you most changed on the back of that feedback?

We started calling more and more non-Silicon Valley companies. Our 18- to 24-year-old associates were saying, “In Chicago, nobody knows who we are. In Indianapolis, nobody knows who we are. We need to get the brand out there.”

We hired an amazing head of PR and communications, Michaela, who had helped build TCV’s brand for Bradley, which had built a pretty good brand over a couple of years.

We asked, “How do we get entrepreneurs in Madison, Wisconsin, to know who we are?”

We have 10-plus—or 15—years of quarterly letters about the hierarchy of bullshit. Other venture funds take them. We literally have a 400-page book. It makes a good doorstop and good kindling in the winter for a fire, but it’s really good.

We want to start releasing that content to the public. We joke that partners at Andreessen Horowitz, Benchmark, or First Round have made the hierarchy-of-bullshit letter kind of famous. We didn’t actually make it famous. We just gave it to other people, who distributed it.

We should distribute our own content. We have 10 or 15 years of amazing letters, so let’s start putting those out there.

Let’s go on CNBC and Bloomberg. Entrepreneurs think that if you’re on television, you’re smart.

We tend to be a very direct firm. Another thing we pride ourselves on is fast nos. If you talk to a company, they’ve prepared for the meeting and done a bunch of work. Tell them quickly if it’s not a fit. Don’t waste people’s time.

I tell our team that when you talk to a company, they’ve prepared for the meeting. Tell them it’s not a fit. Don’t drag it out for a month or 2 weeks.

It’s the same with LPs. If we’re not a fit for an LP, please tell me. There’s no reason for me to meet you 2 or 3 times and give you a bunch of data. I highly respect somebody who has a call and says, “You know what? What you do, Mitchell, isn’t a fit for us.” No problem.

Do you give them the reasoning?

100% we do.

You don’t worry that they’ll argue back?

I don’t care. Some people argue back.

We passed on an analyst candidate. I wasn’t involved in it at all. He emailed me and said, “I’m on the Harvard varsity team,” and listed a few other things. He knew we liked athletes.

My rejection inbox is full. He wrote, “I reject your rejection.”

He got another interview. Most people get rejected, and 99.9% of people get rejected, and you never hear from them again. This guy said, “No, I reject your rejection. Here are the reasons why I think I’d be really good. Here are some interesting companies that I think are interesting.”

He sounds pretty damn good. Why did we pass on him? They said, “We were kind of full for the year.” I said, “No. Get him back in. Let’s meet this guy.”

The thing I understand about people who want to get into venture is that it’s actually quite easy if you give the premium version of yourself.

If I send you 3 companies every quarter that are aligned with Lead Edge’s model, and they’re companies you like—I can see that on your website—and I do that for 3 quarters, or 9 months, then I come in here and talk to you.

100%. Yes, that’s impressive.

Mitchell, I’ve loved doing this. Thank you so much for joining me. You’ve been a fantastic guest.

Thank you. Amazing, dude. I appreciate it.