Why Margins Don't Matter for Early-Stage Startups | Gili Raanan

Ready to go? Gili, we’ve been friends for a while. We did a show remotely, and it is just so much better in person, so I’ve been looking forward to this. I also love shows. You’ve got to remember why I do this show: I do it because I love to learn from people who are so much wiser than me. It’s the greatest joy to have you here with me, so thank you for doing this, dude.

1. Does the Venture Business Even Work Anymore?

Happy to join. If I knew you were coming with shorts to the interview, I would have shown up earlier.

Dude, do you not realize that we have the table so I can hide them? I look professional from above.

I didn’t mean to disclose any state secrets, but—

It’s fine. We know that my shorts are a little too short.

I want to start, dude, with the conversation we literally just had: Does the venture business work anymore when we have entry prices of 150- and 100-times ARR, as we’re seeing today?

You could say that there are multiple answers to that. First of all, the venture business as a whole doesn’t work. It doesn’t work. It shouldn’t work. Return distributions are not divided equally between players; otherwise, it would be too easy, and there wouldn’t be winners and losers. It would be boring. None of us would be playing that game—we would do something else.

The expectation that the venture business would work out is setting yourself up for disappointment from the get-go. It doesn’t work. Now, it worked for some people, and it worked for some people for some time. The number of people for whom it works for a long period of time—let’s take our favorite friends from Sequoia Capital, Andreessen, Benchmark, Greylock, and Lightspeed—is super small.

If you look at the amount of money that’s flowing into the markets right now and has been flowing in for the past few years, no, I don’t think it’s going to work. I think it’s going to end up with some serious catastrophe for many of the players. If I’m a limited partner and I have distributed my venture allocation evenly, I wouldn’t sleep well at night.

Again, as you know, I’m focused solely on cybersecurity, so I know very little about other domains that you probably know way more than I do. But cybersecurity is probably an interesting enough market to talk about, and it’s sizable enough to talk about.

The flow of new players into cybersecurity has been quite steady for the past 20 years. You’re looking at around 350 to 400 new teams that get funded every year across the United States, Israel, and a little bit in Europe. Unfortunately, it should be more, I guess, and I hope it will get to a bigger number over time. But that number—that’s 100% of the cybersecurity universe.

Think about it: Over the past decade, there were about 4,000 new cybersecurity startups. In the next decade, there will probably be 4,000 to 5,000 cybersecurity startups in the world. That’s a large number.

Over the past few years, the entry price, as you rightfully mentioned, has been going up for many of those startups. When I wrote the first check to Assaf Rappaport at Adallom, the first company he started in 2012, it was done at, if I’m not wrong, a $15 million post-money valuation. Many of those deals are now done at much higher prices.

That’s the incoming stream. If you look at the outgoing stream, you look at the exit prices or even the likelihood of a cybersecurity company becoming a unicorn. Take Israel, which is probably 40% of the market. We have to multiply the Israeli number by 2.5 to get the global number.

Do you have any guess what the number of companies that became unicorns in cybersecurity last year—2025—was?

16.

2.

2024? I’m going to go 5.

1.

Oh, that was a bad year.

So, it’s 2 or 1 since 2022. The only year that was an outlier, an exception, was 2021. In 2021, there were around 7 companies that turned into unicorns, but that changed the mindset of investors.

And were they synthetic? Were they artificially inflated? Have they persisted as valuable companies, or was that a 2021 bubble?

In a way, it doesn’t matter.

Mm-hmm.

It is what it is. Those are the statistics. We can argue about the reasons and the drivers, and we’re probably going to be wrong because we have all kinds of biases. But the fact is that, out of around 150 new companies in cybersecurity in Israel, the likelihood that you’ll hit a successful company is still 1%—one out of 150, maybe 2 out of 150.

2. Are We Just Being Boomers? The Counter-Argument on Outcome Sizes

The prices, the entry prices where you buy stock at the seed stage, are going significantly higher. That means the market is not balanced. It means that a lot of that cash flowing into the market would be wasted. It means that, not just as a limited partner but as a founder, you have to pick your financing partners more wisely, because the numbers and the probabilities are not working in your favor. They’re working against you, and it’s just getting worse and worse over time.

Can I interrupt you and say this? I can say it because we know each other, and you know there’s so much love and respect for you. Do you think you’re being a boomer?

What I mean by being a boomer is that the alternative argument would be: Gili, we are seeing labor displacement like we’ve never seen before. We’re seeing outcome sizes expand like we’ve never seen before. We’re seeing CrowdStrike and Palo Alto Networks reach market caps that were never before thought possible. Of course, we can pay more on entry because the outcome sizes are so much bigger.

You can say that. That would be a legitimate argument, and I would accept it with all humbleness. It would not change the probability facts around this game. Venture is a game.

We know very little when we get into investments. We analyze—if we analyze—product ideas and markets. Mostly, we analyze smoke, because the founders would change their minds in just a few weeks, and it would be a different product, a different market, and many different things. So we know so little.

You’re right: Some of the outcomes, definitely in cybersecurity, became very massive because the pain points in cybersecurity are massive. This is not an argument for why we should invest less in innovation in cybersecurity. On the contrary, we should invest more in innovation in cybersecurity, for reasons we can discuss in a minute.

3. Will Mega Funds Be Able to Return Venture Economics?

But we should be super realistic as investors and limited partners about the ongoing and lasting impact of entry prices when we invest in innovative technologies and emerging teams.

What we haven’t discussed, correlated to that increasing entry price, is the increasing fund size associated and correlated with it. Fund sizes have ballooned. We both love your Sequoias and your Andreessens, and now we have $10 billion funds. Andreessen is combined, so that’s a little bit misleading, but David George has a $6 billion to $7 billion pool. It’s a lot of money.

Do you think the mega-funds will be able to return venture-like economics in this generation of venture, given what we just discussed?

The funds that have the tradition, the textbook, and the guardrails to make great investments will continue to do well. Would I invest in those funds personally? Yes.

We should admit that we are looking at a massive opportunity ahead of us. This is not criticism of the opportunity. The opportunity is here, it’s real, and the investment in innovation is justified. Those companies—especially those companies that are growing very, very fast—need a lot of cash, more cash than before.

Cloud, code, and AI, at least in the next few years, would not change that materially. It takes a lot of money to build large companies. So, yes, I encourage founders to raise a lot of money if they want to continue and build significant companies. You can correlate fund sizes to that.

4. How to Tell If a Company's Growth Is Real or Engineered

My concern is around entry prices and whether that would limit innovation at some point in time, because disappointment would show up.

What did you turn down because of price that you later regretted, and what did you not see if you do a postmortem?

We are exercising the science of greed.

So, almost by design, we need to be selfish, and we need to be greedy. Those are good traits for an early-stage investor. Those are not bad traits for anybody who’s dealing with early stage.

Price is an important consideration. Whenever I see an inflated price, or see a deal where essentially it’s a bet on a team, I get more skeptical. Whether I turn it down or not depends on many other factors.

I think one challenging thing about where we are today is that so many of our prior assumptions or beliefs are being questioned. One of them is around growth. The growth of companies today is so much more significant than it has been in the past. How do we value companies when the growth trajectory and pathways are so very different?

I believe, first of all, that trajectory, velocity, and growth rates are the most important indicators for a healthy business. I think that part of our job is to look at that growth and try to sense whether it’s been engineered or whether it’s being achieved organically.

There are ways to engineer growth, but whenever you see a company that’s growing very, very fast, it’s a good company, as a general statement. That’s the best predictor of a company that does well. Over time, I learned that whenever a business gets to a point where it’s growing super fast year over year, it becomes part of its DNA.

It would not slow down just because of averages and things like that. There needs to be a significant external event to slow it down. If a company grows fast, it will continue to grow fast. It’s part of the DNA. They probably do something very right at that company.

5. Wiz vs Sierra: What Fast Growth Actually Looks Like

We can analyze that and backtrack it, attributing it to all kinds of founder traits, market dynamics, and things like that, but it doesn’t change. If you look at the way companies like Waze or Sierra have grown, you see that same pattern.

At Cyberstarts, we do the same type of product-market-fit exercise with all our companies. We call it Sunrise. We spoke about it the last time we met, and it’s an attempt to get into some sort of alignment between the pain point in the market and the solution you have.

You really sell something that people would use, love, buy more of, and recommend to their friends and colleagues. That’s product-market fit.

At Waze, when you look at the first year of selling software, their first quarter was $1 million. Their second quarter of selling software was $2 million, then $8 million, and then $24 million. That’s an amazing year. That was in 2020, or maybe partially in 2020 and then early 2021.

When you see that level of growth, this is not a one-time event, and the company continued to grow. By the way, we had the records of companies like Palo Alto Networks or ServiceNow, which are part of the Sequoia Capital portfolio. I had access to the numbers, and this is an insane pace that Waze demonstrated.

Do you think great companies are up and to the right? I was always of the belief that companies zig and zag and bounce up.

I’ll give you the other example. Sierra had an amazing start. They sold probably $500,000 in their first quarter, then $1 million, and then they sold $0 for 2 quarters. Literally zero. It was, “Okay, what’s going on?”

As an investor, you look at yourself and say, “Okay, I really screwed up.” Then the team—and I really attribute that to the founders, Yotam Segev, the CEO, and Tamar Bar-Ilan, the CTO—analyzed what was going on. They made some modifications, and in the next 12 months, they sold $12 million of new business. They went from $2 million to $12 million.

I don’t like to disclose all the numbers because it’s still an active company that’s making terrific progress, but it continued to grow extremely fast. When you see a company that grows that fast, it’s part of the DNA. There’s something about the company that makes it grow fast.

6. Does Market Size Determine Whether a Company Plateaus?

It may be amazing execution on go-to-market. It may be weakness on the competitive side. It may be perfect timing with the market. It’s probably product-market fit. There are reasons that you can analyze, but the thing that makes them move so fast typically, in most cases, would not simply fade away.

One of my biggest lessons is the importance of market size and just having mega-markets. To your point about being up and to the right, if you hit target and continue to hit target, I honestly thought, “Yes, but so many companies plateau.”

They hit $20 million, $30 million, and then the markets are just not as deep as we thought. They’re more crowded than we thought, or the market is not what we thought it was. Am I wrong? Does one great quarter compound into the next great quarter? How do you think about that? The majority do plateau.

I don’t think you’re wrong. I’ll give you 2 contrary examples, and that’s the beauty of our profession: it’s made up of the exceptions. Who cares about the rules?

Take 1 of my portfolio companies in Fund 1, called No Name. It was focused on API security. It was an amazing company. They did $3 million or so in the first year and $15 million in the second year. That’s amazing, and they did slow down.

Why? The market for API security was a niche segment within application security. In order to really sustain that growth, the company had to reinvent itself into a much bigger product vision and market vision. It was super hard, and eventually we sold the business to Akamai for $500 million or so. That was the end of the story.

7. Island - Building a Market That Didn't Exist

On the other hand, another company in our portfolio, founded in the same year, 2019, is called Island. They have amazing founders. Mike Fey is the CEO, and Dan Amiga is the CTO. The company is basically selling browsers, and their idea is the enterprise browser.

Believe me, in 2019, the number of customers, the number of CISOs, and the number of chief information security officers who told us that they needed an enterprise browser equaled the number of CIOs or users who told the market in 2007 that they needed an iPhone. It was a market that didn’t exist.

Still, the company is growing super fast. It’s a $5 billion company today by valuation, selling and growing very, very fast in a market that they actually defined. The market is growing.

I can’t talk about the specific customers they have, but they have tons of financial-services and Fortune 100 customers. Think about a bank using an Island browser instead of Google or Microsoft browsers. That’s unbelievable, because you’re essentially competing with free, which is tough competition.

The conclusion, again, in my mind, is that we are exercising the science of exceptions. It’s good that we share those lessons, but if you just take those lessons and apply them linearly, I think it would be very hard for you.

You mentioned 2 incredible businesses there with Wiz and Island. I’m interested because when companies are on a trajectory like they are—and I’m not choosing them, so I’m just talking about a trajectory that’s amazing, with fast-growing companies that are clearly looking like winners—capital concentrates, and what happens often is the funding funnel explodes.

Do you worry that too much money goes in too quickly and the founders become defocused and distracted?

I’m never worried about that. Never worried about that.

Why?

Because it takes a lot of money to really build those companies, and if we don’t need the cash this year, we need it next year. So, I’m not worried about that.

The contrary example is engineering growth. If you’re taking good money and your magic number is horrible, and for every dollar you spend on sales and marketing you generate $0.10 in new ARR, you’re in a horrible business. Yes, you can take that money, throw it in, and your efficacy, your yield, is so low that you would not be able to sustain it.

But if your yield is good—if you’ve built a product that fits what the market needs, you’ve got a go-to-market team that executes in a decent, or decent-plus, way—your yield would be significantly higher. It may not be as good as you’d like it to be, say, $1.40 on the dollar. Maybe because it’s early, it’s going to be $0.65, growing into $0.80 on the dollar, but the yield would be decent.

You can see how you can turn it into a profitable business. Why would you care that you have another extra $200 in the bank?

I think the concern is that you have a brilliant but young founder who suddenly brings forward a product roadmap, does 4 things instead of 1, becomes defocused, opens up new geographies too soon, hires too aggressively and poorly, and then suddenly the core business that we liked, that we loved, is all over the place and we need to rein it back in.

Intellectually, I get it. I respect that, but I don’t have that concern. I’m not in the business of babysitting founders, and for me, this is like babysitting the founder.

8. Do Gross Margins Still Matter in the AI Era?

If we trust them to build, in my case, an important cybersecurity company that’s critical to all the major banks in the US, and you put in their hands the safety of our nation’s most sensitive information, then you tell them, “Okay, and you can’t handle the idea that you have some extra cushion in the bank, and you’re going to get sloppy and lazy?” I don’t buy into that.

You mentioned engineering growth. One way that you can engineer growth today is actually in your COGS: spending on inference and allowing for a reduction in margin. I was always taught that margin matters, but we're seeing margins degrade in a wave of AI as more and more is spent on inference. Do we just appreciate that margins will come good eventually, or do we appreciate that AI is just a different margin profile that we have to get used to?

I'm not sure what the right answer is, because I don't think that we have seen enough healthy, profitable AI businesses to really derive the important vital signs for a healthy AI company. Who knows? I can tell you for sure that the vital signs for a healthy cybersecurity company involve healthy gross margins.

My instincts are that gross margins matter. Now, are they important? How much do I discuss—how often do I discuss—gross margins with my early-stage companies? Never. Part of the journey and part of our job as investors is to really help the founders realize what challenges and problems they need to tackle right now, this year, and, let's say, in 2026, and what challenges and problems they would tackle in 2027 and 2028.

If I were lucky enough to have you become a founder of a young cybersecurity company in the Cyberstarts portfolio, I would tell you, “Gross margins are important. Let's talk about it in 2029. Go, and let's build the foundations of a healthy business, assuming that we will get to deal with gross margins.”

That's true for cybersecurity. It may not be the truth for AI businesses. As I said, I don't think that we as an industry have enough track record and history with that, but I suspect that gross margins will continue to be important.

Has your expectation of the growth rate of companies changed? Before, triple-triple-double—from 3 to 10—was good. Now, with a Lovable, with a Legora, with a Harvey, you need to go from 50 to 200 in 2 years. The growth rates are so different. How have your expectations changed?

I think that exceptional companies have traditionally grown at an extremely high pace. For me, an extremely high pace is, let's say, in the first 5 years from the moment you start to sell until the fifth year afterward, if you go 4x, 4x, 3x, and 3x on new ARR—new ARR, not ARR.

In the second year, you do 4 times the new ARR that you did in the first year. I'll save you the math. That's 144x after 5 years, which means that in the first year, even if you have booked $1 million of new ARR, in the fifth year you'll book $144 million of new ARR. That's a nice company. That's a nice company.

If you've done $2 million in the first year and you follow the same velocity, you'll do $288 million of new ARR. That's an even better company. So, I don't think there is a limit on what great is. I just gave you the amazing numbers of Wiz, and I'm confident that 5 years from now I'll be able to show you—I will be able to demonstrate—another team showing that, actually, Wiz was a slug and they can move much faster and they've done whatever it is.

But the bar for real greatness for companies, I think, pretty much stays the same. Now, you can do higher than the bar. That's great. Good for you. Do that. You can go from 5 to 50 to 200. Please do that.

9. Why Public Markets Are Crushing Software Multiples Right Now

But even if you do, I don't know, in ARR, 1, 4, 16, 48, those are terrific numbers. You'll do well. You may not be the most iconic company ever, but you'll be a very, very nice company.

They're terrific numbers, and they're even more exciting if the multiples at which we value those companies are good. Kidding. I look at my public-market book, and I used to think I was so good. It used to just be green, green. Now I look at it and it's all red. I mean, Google and Nvidia aside. I look at it and I'm like, “Oh, maybe I wasn't so good.”

The multiples are so low. You have Monday trading at about 1.5x. You have Wix trading at 2.5x. They just announced a buyback, which is enormous at that $4 billion market cap. What do we do in a world where these multiples are so low and public markets don't value what we always sold?

I'm not always sure I understand public markets, and sometimes I'm confused. I'm baffled, exactly as you are. My guess is that markets have expectations about growth rates, as we have discussed. For whatever reason, if they believe that the growth rate of a company will decline, then, in those specific cases, my guess is that there's an expectation that autonomous programs would displace and eat part of the business of those companies.

10. Is the Extension of Private Markets Fundamentally Good?

Then you would see the multiple declining. But again, I'm not sure, and I'm not confident about what I'm saying. That's my assumption. The multiple is just the market's anticipation of your growth rate. But if those companies continue, regardless of the market, to grow at an incredible pace, the multiples would rebound back to where they are.

With that extension of private markets, because I think so many people are so baffled by the public markets that they don't want to go public—your Stripe or your Canva of the world—do you think the extension of private markets in the way that we're seeing is fundamentally good?

11. Using Secondaries to Retain Talent & Return Capital to LPs

I think it's functional, and I think it's sustainable. For me, going public is not a financial event. It's a branding event. It's an occasion where you tell your customers, your partners, your employees, and your future employees, “I'm here to stay.” That's an IPO, because typically it's not a financial event. It's not a liquidity event. On the contrary, it's the opposite of a liquidity event.

You get shackles on your hands. You cannot sell stock. You've got all kinds of limitations. It's hell for liquidity, but it's an important marketing event. So, I believe that many founders and many companies would still choose to go through that exercise and pay the price for lack of flexibility and lack of liquidity just to gain the long-term value of that marketing event.

An IPO by itself is not a financial event. It's not liquidity. It's the contrary of that.

With that extension, we have the ability to sell in secondaries and, bluntly, sell into much higher-priced rounds. How do you think about your responsibility, or the importance, of selling in secondaries much later on and providing mega-returns to LPs in these very highly priced rounds?

I think about secondaries, first of all, in the context of retaining talent. That's, I think, the most important consideration I have in mind when I think about secondaries, because it doesn't just take a lot of cash to build important companies, and specifically important cybersecurity companies in our case. It takes longer.

With the current market, you typically grant employees stock for 4, maybe 5 years. Then, yes, you can do some new refresh and new allocation, but typically those are fractions of the original allocation, because the company is bigger, there are more employees, and it's in a different stage.

You get to situations where your best employees, your most important employees—your best engineers, your best product managers, your best salespeople—are already fully vested. Structurally, you're unable to allocate equally large or equally tempting grants, and you actually force them out of the company.

For those employees, assuming they're not born super wealthy, they're lucky enough and happy enough to be part of a company that's doing extremely well. They're fully vested, and now most of their family's wealth is actually attached to that 1 company, so it's very logical for them to consider diversification, exactly as we diversify our portfolio by going and joining another team and hoping to build a diversified portfolio.

The antidote for that market's built-in weakness is the secondary. That's the reason, by the way, that at Cyberstarts we created a vehicle we call the Employee Liquidity Fund, which is focused not just on one-off types of secondary deals, but on creating a recurring program with a portfolio company where we provide liquidity to its employees every year.

What we do is underwrite a tender offer every year, so the employees of that company know that they're getting liquidity. The very same type of liquidity they would get in a public market, they would get in a private company. That would help our portfolio companies retain talent.

How does the rest of the cap table feel about that? You have ROFRs, obviously—rights of first refusal.

I'm happy to let others participate with me. I do not object to that. We just announced that we've done our first type of secondary program with Cyera. I think that we're buying—probably not mentioning the exact number—but it's many, many millions of dollars from a few hundred employees of Ayera.

How do you set the valuation on those? Is it a premium to the last round, and you just have a kind of blanket valuation mechanism, or is it?

It's an ongoing process with management. You have to price the round. So, liquidity is—back to the topic.

I gave you the example just to show how we practice this theoretical argument: it takes more time for companies to mature and get to the public market. If they get to the public market, it becomes a strain on their talent pool, and secondaries are actually the solution for that. The story I just told you, or the example I just gave you, is a way to solve it. I’m sure there are other ways to do it.

12. Did Gili Sell Wiz Too Early & What Did He Learn?

Secondaries can also be a way for early-stage firms like Cyberstarts to return capital to limited partners. It makes the system and the markets more sophisticated, and with that extra sophistication, you can create better solutions, first of all, for employees, founders, and limited partners. Overall, I think that’s a highly positive element in a business.

So, you will lean into liquidating some parts of positions and providing cash back in earlier situations?

It’s not a secret that at Cyberstarts, we have sold secondary shares in companies like Wiz early on. By the way, I regret that I sold every single share in Wiz. I regret it because, if I sold it right now, I would show better performance for my limited partners. But at the time, it looked like the right and responsible thing for us to do, and we did it.

What did you get wrong? Obviously, you made millions and millions—it’s incredible—but when you do a postmortem on that, what did you not see that you would have liked to have seen?

13. GP/LP Misalignment: The Hidden Problem Nobody Talks About

It was the early days for Cyberstarts. We talk about startups at Cyberstarts or at 20VC, and those are businesses with business plans, teams, and clients. Early on, I thought it was a good thing for us to show our limited partners that we had an incredible, super-high paper value in the portfolio, and I wanted to demonstrate to them that we could actually drive some liquidity to them.

Do you think there are core misalignments between GP and LP? We can take that as an example, where an early GP will want to distribute and show great DPI because they want to go and raise—I’m not saying you here at all, but in most cases, because they want to go and raise a bigger fund sooner. Actually, if I’m an LP holding that position, I want you to retain it, and I don’t want you to do that. There’s a misalignment there. Do you think there are other misalignments that we don’t talk about?

Potentially, but even for that quote-unquote misalignment, it’s always easier to look at it over time. Back in the early days, when I made a mistake and sold the Wiz stock, if I knew where it was going, I would hold on to that.

14. Quick-Fire Round

The reaction from my limited partners—and I still have a sophisticated, smart set of investors—was overwhelmingly positive. They were cheering for that because, for them, it was a new GP, and that was a positive event. I had only 1 or 2 super-smart LPs who reacted differently. I remember 1 of them called me and said, “Hey, I’m not investing here to diversify or to hedge my risks. I actually like to take more risks.”

15. How Gili Has Changed as an Investor

I appreciated that. By the way, I still believe that, for Cyberstarts back then, it was the right decision.

When you look back at the investor that you were and that you are today, have you changed much?

I think I changed a lot. I probably went from zero to one—meaning from no business, a seed investment, or no idea to a real business—probably close to 50 times. That’s a lot, and I hope that if you do that type of journey 50 times, you learn something. I think I’m learning every day.

I think the only constant in our business is the diversity and change of the people I meet and partner with. I think that’s what makes this profession. In many ways, it’s a terrible profession. It’s a profession where you don’t know if you’re good at what you’re doing for 5 or 6 years. Show me another profession where you show up to work every day for 5 years and have no idea if you’re doing any good.

But I really think it’s one of the most exciting professions in the world, just because it gives you the opportunity to share your life with so many amazing individuals. You can gain just a little bit from every team you partner with, but cumulatively, I think we’re gaining a lot, and we have to listen better. We have to become better listeners over time.

That by itself makes us better people, better parents, and better partners. You’re going through a change, and it’s not like a one-time change. It’s a gradual change, and the more you do that, the more teams you go with on the journey, and the more ups and downs you experience in the business, you change. I think you become a better version of yourself.

What would you say to me and to many people in the industry who are looking at frameworks that we used to use and that are kind of out the window—whether it’s your Rule of 40, or your triple-triple-double-double, or your focus on margins in the early days, or whatever these are—and the world seems to be less secure or obvious than it was in a prior generation? What would you say to that younger generation of investors feeling insecure about their skills in this new world?

Learn as much as you can from old farts like myself, but at the end of the day, use your gut to make decisions. Nobody knows better than you do.

Do you have a monopoly on the Israeli cyber market?

I don’t know. I don’t know, and I don’t think about it.

Do you ever have a company, though—and I don’t mean this arrogantly at all, and you don’t sound arrogant because I’m asking you—do you ever have a cyber company in Israel where its seed round is announced and you’re like, “Hmm, I didn’t see that”?

Maybe once or twice over the past 8 years, but there are deals where I’m telling myself, “Okay, I should have done it. That was a mistake.”

Which one resonates most?

It really doesn’t matter, and I probably regret—I probably don’t regret the right one, and I regret the wrong ones. But one thing I learned about the business is that I focus on the deals I’ve done and the teams I’ve partnered with. That’s where I put my focus and energy.

You can’t cover everything, and you can’t get everything. You’re not going to win every battle. If you’re stressed about winning every battle—“I need to be in every important AI company”—I can predict that you’re not going to be in every important AI company. “I need to be in every important cybersecurity company.” You’re not going to be in every cybersecurity company.

So, I focus on my portfolio companies, and I try to do the best with the teams that put their faith in Cyberstarts and work with us.

It’s funny—I remember speaking to Pat Grady about the great companies that Sequoia invested in, and I was saying, “Oh God, you’re such great pickers,” and I was being very kind, as is deserved. He said, “You don’t understand, dude. Every single public company that doesn’t have Sequoia as an investor is a miss. And this is not okay. You understand that?”

It really hit home for me that market share was a core driver for them, and it is for Andreessen. Have you lost a deal in the last 5 years?

Yes.

You have?

Yeah.

Who did you lose to?

Some other amazing investors.

Is there anything else you could have done? I always think, “I never want to leave anything on the field.” That’s what I say to the team: “We could have done more. I could have done another customer call, sent them another intro, hired someone else for them, paid more.” Is there anything else you could have done to win it?

Absolutely. There are always things you can do. If I look at this startup business, we are improving it all the time, and I’m telling my partners all the time that we are always as good as our next investment.

16. How to Build a Great Venture Partnership

You know why it really doesn’t matter? Because if all our investments were amazing and we lost 1 or 2 companies, that doesn’t matter. So, again, everything leads me to the conclusion that we should focus on our own thing. We typically get what we want to get, and we can always improve. I’m very, very happy with the progress.

We’ll do a quick-fire. As we think about building teams, you have an amazing partnership and great people on your team. What would you advise me about building a great venture partnership with incredible dynamics and relationships between partners? What should I know that you’ve learned?

Lots of lessons. I’ll pick 1 example, which might be nontrivial. I learned that people are very different, and they bring different talents with them.

As a manager, as an executive, as a managing partner, a very easy mistake you can make is to try to create some sort of guardrails and textbook and bring everyone into the same mode of operation. You typically do that because you think, “Okay, this is what worked for me.” Now, here’s a new partner.

Let’s map the gaps between the way he or she performs and that recipe, and let’s bridge the gap. My view on that is that I would let each team member play on their relative strengths. I would not require them to focus on improving their weaknesses, but actually play more often and stronger on their relative advantages.

On their weaknesses, at best, they can be as good as the market. But on the things that are exceptional, they are creating real advantage, real alpha, some real greatness. That leaves Cyberstarts with a team of people that really enjoy working with each other, but each of us is operating in a different way. And we respect that.

I think you do—you learn that through mistakes. I think I put guardrails on people before, where I constrained them to my way of thinking, and then I realized that that’s a net negative for them, whereas it was a positive for me.

Exactly.

Okay, quick-fire round, my friend. What have you changed your mind on most in the last 12 months?

Founders’ chemistry—how important that is. I always thought that chemistry with the founder was important, but founders are brought to life in all shapes and forms. Focus on the teams that you have the most chemistry with.

Are the best founding teams not often broken up eventually? We always say, “The CEO is amazing, but the CTO…” And I’m like, “Don’t worry. Just focus on the spikiest element being the CEO. The CTO will probably fall away. They might leave. They might be a head of engineering, whatever that is. Just focus on the spikiest person.” Does the founding relationship matter?

Extremely. It matters extremely. Yes.

How do you test it?

I have simple tests. If they truly know each other, were they roommates for a long period of time? Did they work together and go through some challenges? Sometimes you don’t know.

Who do you learn the most from as an investor? If I’m fortunate enough to have people like you, Neil Mehta, or Pat Grady as my mentors, who are yours?

The decade I spent with Sequoia Capital was a formative period for me, and it wasn’t an easy period. I couldn’t do what I’m doing today without learning from Doug Leone, Michael Moritz, Jim Goetz, and Pat Grady. It wasn’t easy. As I told you, it took me a long time to mature as an investor, and it’s hard. It’s hard to really—you know, it’s crazy hard.

You show up to the office every day. You’re surrounded by super-achievers who are building amazing companies. You look at yourself in the mirror and say, “Okay, I’m the shittiest investor in this room.” There are, I don’t know, 10 guys around me. I’m the worst. And the next day, I’m still the shittiest investor. And you go like that every day.

It’s really hard. It’s really hard. You have really bad days sometimes. It takes a lot of greed and determination to keep going and believe that you’re going to figure it out.

What was your hardest day as an investor? I can pinpoint one for me.

When the first company I invested in shut down. I had to shut it down. That was super hard because it’s a very public failure. It’s a failure that you cannot cover. It’s a failure that you have to deal with.

What motivates you more: the thrill of winning or the fear of losing?

The thrill of winning.

What’s the most memorable first founder meeting that you think of when I say that? Not the best founder or something, but just the most memorable first founder meeting.

A really fun first meeting with a founder. I would not mention the name. During the meeting, the founder started to shout, “I’m the best. I’m the best. I’m the best.” It went on and on for about 10 minutes, with him praising himself.

Did you invest?

No.

Did it turn out to be an interesting company?

A public company.

Wow. That’s amazing. Okay, final one. What are you most excited about when you look forward to the next 10 years?

Working with my team and growing amazing companies that can keep on making an impact on cybersecurity.

Yeah. I so appreciate you. I so appreciate the friendship, and I so appreciate the honesty. You’ve been fantastic.

I really enjoyed it. And hey, thank you so much. You should invite me more often.