Sam Altman - Startup Investor School Day 1

I'm going to turn it over to our first speaker, Sam Altman, the president of Y Combinator, who actually had the original idea for this course, so I'm pretty grateful for that. He's also the man who has said, "You want to sound crazy, but you want to actually be right."

Sam, thank you.

Jeff, and thank you all for coming. It's cool to see so many people in the room. So, I want to talk about why, how, and what to do to invest in startups. I'm only going to talk a little bit about why, especially given how many of you already invest in startups. But I did do something to get ready for this class, which was I asked some of the best investors I know why they invest in startups. And I contracted.

Oh, Steven, I think it's automatically advancing. No problem. So, and then I compared that with reasons I've heard from other people who I don't think are as good. I think it is worth thinking about why the people who have sort of done the best in the field, what their motivations are to invest.

Perfect. This was one I heard from a few people—these exact three words—from a few different people, and this resonates with me. The thing that I like the most about investing in startups is that it's energizing. I feel like I am constantly working with people that are not burned out on the world at all. They have unlimited energy, they have new ideas, and they have the sort of the beauty of inexperience. They don't know they're your people that are doing things for the first time, are willing to do things that anyone who has got a few more battle scars won't try.

That is incredibly energizing to be around. "Help shaping the future" was something that the very best investors said again and again. And as part of that, the leverage on time, the ability to work on multiple things came up again and again and again. Most of the time, you lose one extra money, but occasionally you do get to make a hundred or a thousand extra, and that, for the same reason slot machines satisfying, is incredibly addictive.

It's also satisfying because every once in a while, a founder of a great company, who is, you know, now super famous, will tell you, "Hey, that thing you did for me eight or ten years ago was this make-or-break difference to my entire career." And that's deeply gratifying. You get to be around some of the most talented people in the world.

There is this sense of just endless optimism around the future that is really important to my own personal happiness to be around. I haven't found that in too many other places in the world besides startup founders. It's incredibly humbling. I write on the back of every stock certificate on a post-it note my confidence interval and what I thought was gonna happen with the company, and you get used to being wrong a lot.

That framework, that mental adjustment, that you know you're usually gonna be wrong, that has been helpful to me in everything else I've done in life. You do learn a lot, though. If you treat this as something that you're gonna try to get better at and sort of deliberately practice, you can learn a lot, and you can get better at this really quickly.

Alright, so now I want to move into the two main sections: the how and the what. The number one mistake that I used to make when I started investing was actually not understanding the power law. The number one mistake was that I cared too much about what other investors thought, and the sooner you can free yourself of this, the better.

I think this is a very common mistake that people make when they start investing. You get very swayed by what previously successful investors think. The first question that most people ask when they see startups is, "Who else is investing?" Around people totally outsource—I would say 80% of investors outsource 80% of their decision-making to what other people think about an investment opportunity. The problem is everyone does that, and so there's this weird schooling effect where a company gets hot for no discernible reason, or it fits a trend, or whatever, and then everybody wants to invest in one company.

It's just because a few people decided they liked it. So, the number one mistake I made was to be swayed by what other investors thought good and bad about a company. After I corrected for that, the second biggest mistake I made was not investing in the power and not understanding the power law.

The power law means that your single best investment will be worth more to you in return than all of the rest of your investments put together. Your second best will be better than 3 through infinity put together. This is like a deeply true thing that most investors find, and this is so counterintuitive that it means almost everyone invests the wrong way.

The question that you should be thinking about—the question that most people think about when they start angel investing—is, "Can I hit a bunch of singles?" In most other kinds of investing, that's the right way to do it. You know, if you're going to invest in stocks or bonds or whatever, that's how you do it.

You're just compounding singles for a long time. But angel investing is a business of home runs, and you want to look for things that can be potential home runs. We're going to talk about that again and again through this, but this is, I think, the most important thing to learn and the thing that most investors get wrong.

So it's all about the magnitude of your biggest success. It is not about the failure rate. Most investors talk about their failure rate. Still, you know when we have people that are trying to build angel firms come talk, the question—the first question they ask is, "What is your failure rate? What's an acceptable failure rate?" Totally the wrong question—totally the wrong way to think about angel investing.

You can have 95% of your investments fail if one of them returns a billion dollars, and you'll be totally happy. So this is what you want to think about. The first question that I try to ask myself when I meet a startup is not, "Why is it going to fail?" It's not, "What could go wrong?" The first question is, "How big could this be if it works?"

Can I imagine this founder, this idea, this market supporting a, you know, massive, massive company? And then I think about all the things that could go wrong, but I found that if I thought about what could go wrong first, I filtered out the companies that could be giant. The companies that could be giant are at this intersection of "sounds like a bad idea is a good idea," and because that's a very narrow intersection, and because they sound like a bad idea, the best investments are the ones that are easiest to talk yourself out of if you start off thinking about why that could go wrong.

So then there's a question about how do you find these companies? How do you find the companies that can be the handful of companies that get started every decade that are responsible for almost all of the returns? Just to put some numbers on the power law, YC has funded, I think, around 1,600 companies, maybe 1,700. Our top five companies represent about two-thirds of the value that we've created, and our top one company represents about one-third.

So this is like nearly one-third, so this is like this very extreme, very counterintuitive thing. Alright, how do you find these companies?

Another thing that I think is surprising is many of these companies that are sort of generation-defining companies are started by people who are out of network, who are not well-known, who are not sort of the people that you know can start a company and get a bunch of press right away.

And so you have to find these, and the best way to do that is from other founders. This is part of what we try to do at YC: we try to get our founders to like us so much that they refer all of their friends to us because they say, "These, you know, YC creates so much more value than it takes. You got to go work with them."

This idea of like word-of-mouth as the way to find companies to invest in has been great. If you're just starting out, I think what I have seen the most successful angel investors do is just start helping founders for free, realizing they may not get to invest in those companies, but that they'll get the referrals down the road.

So much of this is about people connecting you to other people that you don't know. We talked about this at YC—this value of an open network. A lot of angel investors like to brag about how difficult it is to get a meeting with them, how you have to have a connection; otherwise, they'll never take you seriously.

It's got to be from, like, someone they worked with a bunch, and you've got to be well-known and experienced. We just make it really clear: anyone in the world can go to our website, click apply. We try to respond to emails from people that email us. We take people seriously that have no personal brand, no reputation, no network.

We ask people that are in our network to connect us to the most promising people they know that we don't already know. But this idea that we're available and we're open, I think this was in the two or three most important secrets of YC. This was something that we did that was really different. Other people hadn't done it before. In fact, people bragged about the opposite, and we went totally in the other direction.

I strongly recommend this: be open to that random email that comes in, be open to the introduction of someone that on paper doesn't seem like, you know, that there's someone you want to meet. Nine times out of ten, you waste your time; that other time makes it totally worth it. So one big shift that's happened in the last ten years, I'd say, or maybe 15 even, is there are now way more people that want to invest in good startups than there are good startups.

So founders have really become in the driver's seat. Founders of good companies have a lot of choice when it comes to investors, and founders talk a lot now. The network has gotten big enough that the asymmetry that used to exist where investors had a lot of leverage is gone, and I don't think it will come back anytime soon.

Your reputation matters a lot. It is way more important to your future success as an investor that founders like you and say that investor did the right thing by me than it is that you squeeze out a few drops of juice from a failing company. The number of investors that I have seen do incredible long-term damage to their reputation by fighting over the carcass of a company that was never, because of the power law, going to matter to them anyway, but try to get out, you know, ten thousand dollars or whatever from a dying company.

If you're playing a long game, that's not worth it. Reputation, especially reputation when a company is going badly, is super, super important. I think that is the secret at this point, in sort of 2018 Silicon Valley, to winning deals. The thing that we tell founders to do, and the thing that founders do anyway when they're trying to choose between a number of investors that all want to invest in their company, is they do reference checks on you just like you do on them.

More and more, the thing that I have seen founders use as the criteria to make the decision about which investors to work with is what the other founders that investor is fun to have to say. So I think this will continue to be important. There are other things you can do.

I think helping before you close the deal is good. Deciding quickly, being clear about your reasoning, being responsive, being available— all the things you want from a founder—those all help, too. But this reputation of being good to work with goes a long way, and people remember that for a long time.

The other question I get all the time is, "How do I get a good bargain? Like, how do I get better terms than everybody else?" We recently had a company; they had, let's say, you know, 10 investors that were going to join their seed round. Every one of the 10 had asked for advisor shares. Every one of the ten had said, "Well, unlike all those other investors, I really do work super hard. You know, I'm the only investor in a round that usually gets advisor shares, so you need to give them to participate."

And everybody asked for this—all ten. Same thing: I'm the only one; I need advisor shares. I think, you know, a lot of people are just looking for a good deal because valuations feel high. They have felt high to me for eight or seven years now.

I went back and looked at my best investments. The very best investments I've made have either been a ridiculous deal because no one else wanted to invest or a deal that felt incredibly expensive. The more I was willing to sort of overpay in my mind to invest in a company, often the better it did. And especially when I felt like I was getting screwed—if it was like a huge offer three months after some other round—it was painful.

I did it anyway, I think because the companies that work sometimes work so fast, and you're anchored to like what a fair deal is—you got to watch out for this. But my experience investing in startups is my best investments, with one or two exceptions, have been the deals that felt the most expensive to me.

And the one or two exceptions were companies where I understood something that no one else did, and thus there was no competition whatsoever. Other than that, the companies where I like tried to value invest have not been as good. I think value investing is not a winning strategy when it comes to being an angel investor most of the time.

Alright, I will try to go kind of fast through this to leave time for questions. So there's a big question of what to invest in, and here's the framework I use: I will consider anything that I believe could be a 10 billion dollar company, and that is such a tight criteria. I have no other rules. So, I am willing to look at any stage, I'm willing to look at any sector, I am willing to look at any business model.

There are other investors who have this like, "Oh, I only do this one thing at this one stage," and maybe they make that work. I have never figured out how to do that. The good, the great companies, the companies that are in that number one spot on the power-law are so rare that I suggest you only select for things that can be there.

Other than that, be really open-minded. Speaking of the really big companies, I think—I don’t know if this was always true; I suspect it may have been—but I think today it is easier to start a hard company than an easy company.

This sounds super counterintuitive, but if you're going to build a really big company, you've got to convince people to come work with you, to pay attention to you, to write press articles about you, to care about you, to advise you. If you are starting the 20 mm photo-sharing application, it's really hard. If you are starting a nuclear fusion company, a lot of people want to help with that.

I think especially for the companies that end up breaking out, this is really important; you know, this thing that is so interesting. People proactively want to help you for free, want to come be part of your team—whatever this is something to look for. So, like, is this a company that I believe will be able to recruit hundreds of really talented people who could otherwise start their own companies?

This is a super important filter that I don't think people think about enough. One thing that we've learned at YC is to mostly pick the founders. It is difficult to hear an idea at the very early stage and say, "Yeah, this idea has what it takes to be a ten billion dollar company." You can say an idea doesn't have it, which we'll talk about, but it’s difficult to say this is for sure the big idea.

However, I think you can, with practice, identify founders, and I'm going to talk about how that have a chance at creating one of these companies. Pabu, one of our partners, made a list of the four traits that he thought founders that go on to create giant companies have, and they are obsession, focus, fortitude, and love.

He said this sort of in passing in the meeting like two or three years ago; I've thought about it a lot since. There are obviously other obvious things that everyone screens for, but pay attention to these. Speaking of the obvious things that everyone screens for, intelligence is really important. You can give a founder an idea, and they can start a company; the problem is they need to come up with new ideas for a company, basically like every week.

You have to come up with crazy new ideas—big changes all the time. We tried an experiment once at YC. We funded two teams of strong founders that didn't have ideas but were otherwise really good, and what we learned: they all failed. What we learned is that the good founders are the people that have ideas all the time.

So there's an intelligence component to this. There's a creativity component to this. There is an ability to think independent thoughts component to this. But whatever you want to call this, this idea of this particular kind of intelligence that leads to seeing problems in different ways and thinking of ideas that don't yet exist but should—you've got to have that in a founder.

Communication skills, I think, are one of the most important founder qualifications that people don't think about enough. So much of your job as a founder is about communication. You are, every time you hire someone, every time you go raise money, every time you try to sell the product, every time you try and set a direction for the company, um, you do it.

A huge amount of the founder's job is being an evangelist for the company, and if you don't have really strong communication skills or if you don't develop them quickly, you're at a big disadvantage. Think about it. There are obviously famous exceptions to this, but if you think about it on the whole, the founders of the really super successful companies tend to be great communicators.

Execution speed—there's a lot of ways to measure this, but we talk about this a lot—the need to sort of relentlessly execute as a founder. This is incredibly correlated with success. So one way we test this during YC is between office hours, which we have every week or 10 days: how much progress do the founders make?

How quickly do they take a new idea and try it and say, "Hey, I came back and I tried that, this didn't work, but this other thing did"? In the process, I had these three new ideas. I tried those. Just this relentless cadence of execution is incredibly predictive of success. We had a joke once where there were all these founders who were incredible on paper—they never actually delivered quickly.

They always have great reasons for why they didn't, but they still never go and be successful. Then there are these people who just, like, get an amazing amount of stuff done. Their iteration speed—the speed with which they can have a hypothesis tested and implement it—is unbelievable. That's really correlated with huge success. The rate of improvement of the founder…

So if you look at a founder who comes to meet you for a seed round and compare that founder to Brian Chesky, you will be disappointed 100% of the time. That is the wrong comparison; you will never write a check. However, just like startups, you look at the growth rate—you should look at the growth rate of the founder as well.

So, one thing that we can often tell during the ten weeks of YC is how fast a founder is improving. This is different from how fast the business is improving. Humans always underestimate exponential growth; we're not well evolved for that.

If you notice a founder who is improving incredibly quickly over the couple of months you get to know that founder, pay a lot of attention. Again, you won't get Brian Chesky in a first meeting, but you can find people who are on a trajectory to develop into a Brian Chesky, and that is super valuable.

This is one of those things where I have seen it you know maybe like ten times in my career so far where I just knew that this founder was going to develop into an incredible leader, and it's basically been right every time I've felt that. I really do trust this rate of improvement metric.

I think one thing you have to be increasingly aware of are the wrong motivations. Starting a startup is a very long-term commitment. You know if it's gonna work, it takes more than a decade. It's really hard. There are a lot of days where you just want to give up, and there are a lot of people now who start a startup because they think it is a way to get rich quickly, and unfortunately, it's just not.

So as startups have become the new default career trajectory for ambitious people, there are a lot of people who are doing a startup as a resume item or as a way to get rich quickly. This does not work. The amount of pain that you have to suffer for a startup—you realize at some point, "You know what? I can do pretty well in any series of other jobs with much less risk and much less negative effect on my life."

So you really do want to stay focused on the mission-driven founders. Again, if we look at our own success and failures in our portfolio of YC, every time we thought a company was going to go really well and didn't, the company—the founder did not have this deep sense of mission. So it's something we really look for as a primary motivation.

And then another way I used to get tricked a lot is there were founders that I didn't think were that good, but they had stumbled on a nice business or, you know, they had this metric that was growing pretty well or all of my other investor friends were investing, and so I got scared and did it anyway.

But I think this focus on truly exceptional founders—like people that I'm like, "Wow, I want to go work for him or her," is really important. I've never once made a lot of money backing a founder that I thought was only okay, but a business that was otherwise good.

I talked a little bit about this, but we have a word at YC called "scenesters." This is different from people who just want to make a lot of money. There are an increasing number of people who just want to be around startups and go to startup parties and talk about being a founder. Treat that as a red flag. Obviously, low integrity people—that doesn't work out either.

Okay, this is maybe the third biggest misunderstanding I had, and I think for many people, it's their number one biggest misunderstanding. People always say that what matters is not a startup's current revenue but its growth rate, and that's true.

However, in the same sentence, investors will say, "But the only thing I care about is the size of the market today." This is obviously ridiculous on its face, right? If you think about the biggest companies today, ten years ago, many of those markets did not exist. If you think about the size of the social networking market when Facebook started, if you think about the size of ride-sharing apps when Uber started, that's a really bad metric.

Unfortunately, it has become dogma among investors that, you know, size of the market is the most important thing. Even really good investors say this. I think they either—I think they actually mean what they care about is the size of the market in ten years, but they don't say it that way.

What you should care about, of course, is the size of the market in ten years. If the market is huge today, first of all, you probably have a lot of big competitors already, going after big companies doing that. Second of all, you don't get to surf this wave of this new technological change that pulls startups along and creates a ton of value in a short period of time.

But you should prefer a small market growing super quickly to a very large market today. Super counterintuitive. If you chase the things that worked in the last set of companies—which is what most investors do—Facebook works; they all want to fund more social networks. Uber works; they want to fund more ride-sharing apps.

That is much harder to do the second time; it's far better but more difficult if you can identify the next rapidly growing market and invest there. This is where independent thought is really important. If this is not something where you can just sort of follow what everybody else says, by definition, you've got to learn to form your own thoughts about what the next really big market is going to be.

One way that I like to do this, one way that I like to say, you know, is the market growing really quickly, is to think about this question of whether something is a real trend or a fake trend.

And I'll talk about that in a second—actually, right now. Okay, so it has almost become a joke to make fun of angel investors moving like a school of fish after one, you know, declaring somebody's a hot trend and then two years later never talking about that again and saying, "Well, that just didn't work."

But so there's this question—sometimes they're right. You know, investors in Silicon Valley as a whole, for example, got mobile right as a thing in a big way. But then they got most other things in the last ten years wrong, and I think every time someone talks about a big trend, my first reaction is skepticism, and I suggest yours is as well.

Okay, but how do you differentiate between a real trend and a fake trend? A real trend is one where, although not that many people are participating yet, the people who are use the platform a lot every day and tell their friends spontaneously how great it is.

So when the iPhone came out, most of the mobile industry would make fun of it because Apple only sold a million or two million, whatever it was, in the first year. But if you talk to anyone who had an iPhone, they would say, "This is the greatest piece of technology I've ever had!" People used it for many hours every day; it was absolutely life-changing.

So, even though the number of people that had it were small, you could identify that as a real trend because the people who had it were not only like daily active users but hourly active users. They were the best free advertising Apple could ever have hoped for because they told everyone, "This is the future; you've got to buy it."

If you contrast that to something like—I got to pick on somebody; I'll pick virtual reality. If you contrast that to virtual reality, everyone talks about it as the next trend. It may be in the future, but today, if you talk to people who have VR headsets, they don't use them every hour. They don't use them every day.

Most of them don't use them every week; they sit on shelves. That has clearly not become a real trend platform yet. It may in the future, and the point at which you know people that you know, or a lot of people you know, even start putting their headsets on for hours every day and telling all their friends they've got to buy one, it's the greatest thing in the world—that is the time to start investing heavily in VR.

Can you pick that back up? So this question, you know, are people actually using the platform, I think is a really important one when you're trying to think about the next technology wave. You do want to try to figure this out.

Certainly not all, but most of the biggest technology companies get created soon after one of these massive platform shifts. Sequoia says this thing that I've always liked, which is you cannot create a technology wave that is well beyond the capability of a small company to do, but you can surf one if you can find the wave.

I think that's really important, and I think it's actually, if you use this framework, pretty reasonable to evaluate. Okay, so I mentioned this a little bit earlier. What you are looking for are good ideas that look like bad ideas. These are things that you can articulate; there is a reason that this is going to be huge that most of the world is missing.

Unfortunately, what most people end up chasing are bad ideas that look like good ideas. I would say this is where like 90% of all angel capital in the startup ecosystem goes, so this is something that is worth trying to avoid.

One very common way that people make this mistake—bad ideas that look like good ideas—are chasing the thing that worked two years ago. If you ever find yourself doing that, be very skeptical. If you find yourself tempted to invest in one company where there are hundreds of others working on the same thing, be very skeptical.

If you find yourself tempted to work on something that the founders work super hard to convince you is not going to be a long-term commodity, be very skeptical. The more people talk about—like it is absolutely true that you want something that has real pricing power that comes from a network effect, a moat, or barrier to entry, whatever it is—but the more founders try to sell you on why they are super differentiated and why they have the long-term competitive advantage, the more skeptical you should be.

But I have found this framework, just trying to think about, "Is this a good idea that seems bad, or is this a bad idea that seems good?" I've found that has helped me make good decisions a bunch of times. This is sort of YC's mantra, but it's so important that I want to talk about it again.

The best companies all have great products. Unfortunately, the current fashion in Silicon Valley, I think, has gone a little bit too away from this, and it's too much about growth hacking and sales and marketing machines and everything else.

And that does work for a while. You know, you can get away actually for a pretty long while by executing really well to grow a mediocre product, but you don't usually create a Facebook-sized company by doing that.

I think asking—it can become—most of the time, they won't have a great product. By the time you're making an angel investment, if you don't believe they can't or won't get there at some point, I don't think it will be a huge company most of the time.

And here's the very simple framework I use for this: If I think about all of the most successful internet mobile startups, I heard about those because—this is like true for enterprise and consumer apps. I heard about those because they were so good that one of my friends spontaneously told me about it. They were not being incentivized to. They didn't market to me; didn't advertise to me.

It was just like someone I trusted said, "You got to try this new thing; it's amazing!" If the startup is not gonna get there, I think they will not be at that number-one spot on your power law returns. So I think this is like a really important filter.

Related to that—and I mentioned this earlier a little bit—human intuitions about exponential growth are terrible. We clearly had no evolutionary need for this. We can like visualize linear growth very well; we can like, you know, visualize the trajectory of an arrow very well. We have very few people that I've met, maybe none, sort of tell me, "What, you know, like let's say 25% monthly growth?"

1.25 to the 36th power: where will this company be in 36 months? It's really hard, and it is how these companies get super valuable. So I long ago learned to stop trying to trust my intuition on this, and I just model it out. I try to model the decay rate about how I think growth will slow down.

But I try to say, "Okay, you know, given that this company is growing by word of mouth, how big can it be in five years?" And, yeah, I've learned not to trust my intuition on that.

I mentioned this a little bit earlier, but I wanted to mention this near the end of the presentation. There are a whole bunch of words—a whole bunch of different ways that people talk about this. This is one of the most important concepts in startup investing. This is one of the things that differentiates investing in startups from investing in small businesses.

You are looking for a company that gets more powerful as it gets bigger. You are looking for a company that gets increasing pricing power as it gets bigger. You are looking for a company that has an easier time getting more users as it gets bigger. That gets harder to compete with as it gets bigger.

And, you know, this is often fairly obvious; sometimes it's not; sometimes you really have to think hard. But, you know, once you do, you can come up with a story for it. This is something that a lot of people get wrong because they get caught up in, "Oh, this is going—this is so cool today."

This company has discovered this wonderful thing. Almost all of the value in a startup is the revenue, the earnings it's going to generate in years 10, 11, and beyond from now. If you can't answer this, be pretty skeptical.

And then finally, one other question that I like to think through before making an investment is, "What do I understand that other people don't?" There can be a lot of answers to this, but this comes back to not basing your decision too much off the decision of other investors.

I like to understand in axiom: if I don't answer this question, I don't feel like I have any competitive edge in that particular investment decision. Sometimes the answer to this question is just like everyone's bullish on this company, but I'm more bullish because I understand a specific thing about this market.

And everyone thinks it's good; I think it's even better, so I'll pay this very high valuation. But this is a question that I have found helpful for me in many scenarios.

Alright, unfortunately, I took all the time, but maybe I can do like five minutes of questions. We definitely have time for some questions. I just wanted to remind all the folks who are live-streaming that the hashtag for questions is YC S; it's YC S. I think the beginning of the livestream might have missed that, so please do send some questions.

We started a little late, so perhaps running a little long, and then we'll take a quick break. Thanks.

Alright, yes, what can at least age investors do to add a lot of value to founders? Almost all, if you ask founders this (which, you know, they're the customer here, I think that makes sense), almost always the number one thing they want help with is hiring.

So, helping them find really good people, help them interview—you know, back when I was sort of an active angel investor before YC, I would tell founders, "You can use me as much as you want for interviews. I will help recruit; I will help source people; I will help close." I think people really like that. Help with future fundraising and then help with just sort of like everyone wants to provide the big strategic advice, and that is really valuable.

I think one of the things that I did well when I was an angel investor was I would just try to be available all the time for tactical advice. So I would meet for the big strategic, “What can this become?” advice, and that's fun. But I think a lot of the value comes just from being available at 11 o'clock on a Friday night when a founder needs a two-minute phone call for some emergency.

So, super availability for tactical advice I think is good. What flaws are acceptable in a founder in a, you know, series C or series A stage? A lot, I think. Like, don't compromise on the things that don't get better—like, don't compromise on a founder's integrity.

But if you think the founders are improving fast, I think a lot, so I like I bucket this as traits that I believe can change and traits that I believe can't. If the founder is improving quickly and it's something that I think is changeable, you know, we've had many, many very unsophisticated founders but that were smart, wanted to learn, and were, you know, doing this for the right reasons.

Super mission-oriented come through these doors, and they have just progressed really fast. So, you know, there were founders who I think had like no domain-specific knowledge about—like, one funny thing is when you are negotiating an investment, a lot of the time a founder who's otherwise very good will have no idea because they've never done this before, and it’s kind of spooky.

So you're like, "Wow, you don't know what like evaluation is." And that's the kind of thing that's like, "That's a founder; that's okay." How do you judge for integrity with a founder that comes out of the network?

So, in our experience, we've gotten this wrong a handful of times, but we have prevented ourselves from this mistake hundreds of times because even in our 10-minute interview, if you give like a founder a chance to sort of tell you about the unethical things they do, they will often do it and yet surprisingly often.

So I think the answer is you just listen to the decisions they've made so far in building the business, and if you’re like, "That's not a decision I think is okay," you can expect more of them in the future. But just listen in the first few meetings, and you'll be surprised.

We definitely have been fooled by plenty of people, but we also make our decisions in 10 minutes. Results-based observations about when to exercise pro rata and when not to: several venture firms have done very sophisticated studies of this and they have all come to the following conclusion, which is if the company is raising an up round led by a good VC—say a top quartile VC—you should always exercise pro rata.

And if you do that across your whole portfolio, you'll be happy. Now, this could change if the world really changes. In the world today, there's like very clear data on this. What's pro rata? So often when you invest in a company, you will get—not always, but often you'll get something called a pro rata right, which is your right in future rounds to invest enough dollars in the new round to maintain your ownership level.

Sure. Alright. Do I think security tokens and ICOs will change financings? Probably, but not in the way that most people think. I think this idea that like everyone is going to just raise money from the crowd—I think we'll find out that we have securities laws for a reason and that we want some level of that.

The level of—I think like there are some incredibly important ICOs happening right now, but they are dwarfed by the number of sort of things that are between just incompetence and scams. However, I do think that it's possible that we just find a much better mechanical way to track the investments we do right now.

So that's possible. A bad idea that seemed—sorry, a good idea that seemed like a bad idea? You know, this is like of our darlings, but it's a—it’s an example that sticks with me so much because they were in my own YC class in 2005; it was Reddit. So when Reddit started, I remember very clearly telling my friends about it because I was like, "Oh, there's this site, and you know, you can like find these links."

And I remember people looking at me like, "That's the dumbest thing I've ever heard! Like, there's all these things already on the internet. You know this one is just like pictures of cats or whatever it was at the time, and there's no way you will ever make money on this business."

So that's like there's other more famous examples, but that is the one that for me resonates very deeply because I heard so directly from people I trusted so much. I remember like when they would say it, I would just be like, "Oh yeah, I guess it's not a very good idea."

Like, I had all this conviction that totally went away when people I trusted said something was bad. There's another common version of this problem, which is where there's an idea that seems good in the abstract, but everyone assumes the big companies will crush you. So Dropbox is an example of this, where when we funded them and when they were kind of getting going, everyone was like, "Oh, it's a perfectly nice product, but you know, Google, Microsoft, whatever guaranteed to crush them soon."

Alright, other questions? Yes. You know when Uber was getting going, there would be all of these articles that would come out. It felt like every year where someone would say, "Uber is not worth X; the entire taxi market is only worth, you know, 10% of X." And so, and that just kept going.

And that is really hard, right? Because you don't have a sense for exactly how big the market is because it's growing so quickly. I think one thing you can do is look at like shifts in consumer behavior that are creating new markets. So like, if you thought of Uber as a replacement for booking limo services, that was one thing.

If you then started to realize that people had begun to use it as a replacement for taxis and then public transit and then car ownership, you could project forward, "Wow, this market is actually going to be quite big because all of this other consumer behavior is going to shift here."

Alright, a solo non-technical founder? I wouldn’t say I never would. I’ve done it before; I think it can work. I do like it when that founder learns enough to build an MVP. Where I’ve seen that go wrong the most often is in their ability to attract, evaluate, and retain technical talent.

And so we have a strong, strong preference for founding teams that have at least one technical founder. We also have a strong preference for teams that have at least two co-founders. Again, none of these are absolutes because this is all about the power law. We are always willing to consider exceptions. So I would never answer a question like that and say, "Never."

But I would try to be clear: here's what I've seen work more often. What is the next question? Oh, a good amount of self-awareness. I think a good amount of willingness to take feedback and a drive and a desire to improve is really important.

But, you know, you do also have to sort of believe that you can succeed in spite of all of your flaws, and that’s almost more important to me than like someone who really spends a lot of time categorizing everything about it. If they're willing to listen, willing to improve, I've usually found I can work with that founder.

One more question in the back. How do you think about evaluating your time and resources to different founders? A self-indulgence, but one that works is only fund founders that you want to allocate a lot of time to because if you don't—and if you're like—if the founder is like difficult to work with or doesn't listen or you're just not excited about the business, a, that's probably a red flag for their qualities as a founder, and b, you then won't spend time, and you won't help them, and you won't get this differentiated thing.

So I like I won't fund a founder that I don't want to spend a lot of time helping, and that has always worked pretty well for me. Alright, thank you all very much.