[0:00:06] And so our first speaker, I'm very excited to announce, is going to be Sam Altman from Y Combinator. So Y Combinator is a startup accelerator which provides seed funding and boot camps for startups. He's an entrepreneur, a programmer, a venture capitalist, and he's also a blogger. And he's seen as the leading voice on entrepreneurship, technology, and innovation in Silicon Valley and internationally.
[0:00:30] Over the past 10 years, Y Combinator has funded over 700 startups, some of which are here in New Zealand. And with more than 1,400 founders, it actually has a company valuation of over, and it just confirmed this number literally two minutes ago, of over 50 billion New Zealand dollars. Sam was co-founder and CEO of Loopt, which was funded by Y Combinator back in 2005, and then eventually was acquired by Green Dot in 2012.
[0:00:56] And Sam was named one of the best young entrepreneurs in technology by Business Week. And this is my favorite fact about him because he was listed as one of the five most interesting startup founders between 1979 and 2009. So I went and Googled who the other five were, right? Because I'm like, oh, how cool is this list? One of the other five who was on that list happened to be Steve Jobs. So we are in great company today as I bring to the stage Sam Altman.
[0:01:34] All right, thank you very much for coming out today on such a nice day. We've been in New Zealand now for a few days, and it's really just been awesome how great the startups that we've gotten to spend some time with are. So I actually found out that I have 25 minutes, not 20, so I will talk for a little bit longer than 20 about what we learned about startups. We try to fund startups at mass scale. We found that the community, having a dense community of startups is so important to success, and the more startups we fund, the more they help each other.
[0:02:04] And we've now built this network around the world of Y Combinator founders that work together and help each other build these great companies. So we have a lot of pattern recognition at this point. And what we've learned is that startups are not impossible. They are very hard. But the good news is we understand at this point what you need to do to be successful with a startup. And we are able to help startups do these things.
[0:02:32] And the other piece of good news is that there are only four things that you really have to be great at to be successful with a startup. And these are a great idea, a great product, a great team, and great execution. And that sounds manageable, right? That's only four things. And the founders that we funded that have gone on from total unknowns to these huge successes in their industries have only had to do these four things really well. And I think these are four things that all of they require a huge amount of...
[0:03:02] Intensity and determination, almost anyone can do. So I'm going to talk a little bit about what we've learned in each of these areas about what startups can do to be successful. And the good news is, as we've been spending time in New Zealand with Kiwi companies, I think it's already a foregone conclusion that there's going to be a successful startup ecosystem here. And now it's just a question of how much, because all of these things are already happening.
[0:03:30] One thing that we've learned is that great companies have to start with a great idea first. You really want to have a great idea as the driver, not a company. So a company should be the best way to get an idea that you're obsessed with out into the world. But you shouldn't start a company for the sake of starting one. The best companies are almost always mission oriented. So you have a particular thing that you're so obsessed with.
[0:03:58] That you can't allow yourself to work on anything else because you believe this has to happen for the world. You feel a moral obligation to get this done. And even if other people think it's crazy, or even if it doesn't sound that important to anyone else, that actually turns out to be sometimes a good thing, not a bad thing. But this idea that the company is going to be a vehicle to carry out a mission is really, really important. As I mentioned, it's okay if other people think your idea is bad. In fact, it's good.
[0:04:27] We went back and looked at this once, and our most successful companies had a lot of people when they started saying it was a bad idea. If something's obviously a good idea, a lot of people are trying to do it, including big companies with a lot of resources. So what you want is an idea about which you can say, I know this sounds like a bad idea to most people, but it's a good idea for this reason. And those are the things that make magic startups. Another question that we ask every founder we fund is, why now?
[0:04:59] You want to time your market perfectly. You want to have an idea that wasn't possible two years ago. Two years in the future is going to look obvious and too late. But right now is the ideal time to start this company. And getting the timing right for a particular idea turns out to be really important. Many founders have told us that they were thinking about the idea that became the seed of their company for many, many years before they actually started it. This is not an area to skip.
[0:05:31] So once you have a great idea, you'd like to get a great company, but there's this middle step that's really important that a lot of founders forget about, which is building a great product. This is the hard part, right? This is what you have to do before you get to build a great company. If you just think about the companies that you respect the most in your own mind, very few of them don't build great products.
[0:05:59] The best founders are obsessed with great products, and great products really do create good companies. So the way to build a great product, this first thing sounds obvious, and we're going to talk more about it in a second, which is build something that users love, and the other is talk to users to get this feedback cycle going. So what you would like to do as a startup is build a product that a lot of people love. But you can't do that. Those opportunities are too rare.
[0:06:29] Too many people go after them. So most startups, almost all startups, have to choose to start with either building a product that a lot of people like a little bit or a product that a small number of people really love intensely. And the area under the curve is basically the same in both of these cases. So the total amount of value created for the world, the total amount of love is the same, but it's distributed very differently.
[0:06:56] The mistake that we see startups make all of the time is that they go towards making something that a lot of people really like. Because it feels like, hey, it's a startup. It needs to scale. We need a lot of people. And that's a wrong instinct. It's far easier to take something that a few users really love and get something that a lot of users really love than something that a lot of people like and make them love it. So we really work with our startups to make sure they're not just building something that people are sort of...
[0:07:24] You know, somewhat excited about, but really wait until they find something that they've made that people just say, I cannot live without this. I'd be so bummed if it went away. And getting this, getting to this point is like 80% of the work towards building a great product and something that we push our startups to do all of the time. You know, startups often get scared when a competitor has more funding and has, you know, a snazzier marketing campaign and more salespeople and whatever else and better PR.
[0:07:56] But great products will eventually win. Yes, you have to get distribution and sales and marketing right. You don't have to get that right on day one. But if you can get a great product right at the beginning and then do all these other things later, that almost always wins the day in the end. So fanatical first users, fanatical early adopters are one of the things that we look for most when we're looking at startups. We're much more interested in how much, how intensely your users like you and how many you have.
[0:08:28] And I think that's something that's really gotten lost in the sort of global conversation about startups, where it's all about how many users do you have and how quickly are you scaling. Companies have to build value over 10 years or longer, and this is what will sustain it over the long period. Part of this means you can get your customers manually. You don't need that many to start with, but you do need a very close relationship to get the product to the point where people really love it.
[0:08:57] And so everyone's obsessed with, you know, I need a scalable customer acquisition strategy. We actually tell our startups to do things that don't scale. You know, the way that you get your first 100 users is almost always, and in fact, I think it should be, very different than the way you get your next 10 million users. You really want to have a close relationship. You really want to have a small community that you can engage with very, very deeply and will help you refine your product and make it great.
[0:09:26] So this turns out to be really important advice that is counterintuitive to most founders. Most founders don't want to do anything that won't work in year 10 of their business for whatever reason. But year one is year one and you're allowed to do and you should do very different things. So you want to get this cycle going where you build product, you show it to this small type group of early users, you get feedback from them, and then you get it back into the product and you keep the cycle going.
[0:09:54] And the best startups are almost always the ones that make the cycle as short as possible. So, you know, it used to be, I think, in, say, 20 years ago when people were shipping software in boxes to CompUSA or whatever and then selling that. The cycle was measured in years and then it got measured in quarters. And then the first version of the web came and it was measured in, you know, months. And at this point, it's measured in hours. They're really great companies.
[0:10:23] Get this cycle down to literally hours. And that used to only be the case for web companies. But now it's the case for hardware companies as well, where people are 3D printing versions of their hardware every few hours and giving it to their customers. If you only make something 1% better each cycle, but you can measure your cycle time in hours, your product gets really great really quickly. So the two things that we tell startups are spend little money.
[0:10:52] And get your cycle time down as quickly as you can. In any industry in which you can do those two things, I think you can build a great startup. So the third thing that you have to be really great at is team. And I'm going to talk about two areas here. How to think about co-founders and how to think about the people that you hire. And on the co-founders thing, the phrase that we use to describe great co-founders are people that are relentlessly resourceful. And I've come to believe over a lot of time,
[0:11:26] That this is more important than anything else, including intelligence, when it comes to co-founders. As any of you who have been working on a startup know, it's really, really hard and there are a lot of times when you want to give up. Most of our failures that happen in the early stages of startups happen because the team falls apart. One co-founder decides to quit. The co-founders don't like each other anymore. When the going gets hard, they want to give up. And this is usually fatal to startups.
[0:11:56] I think more important than anything else when you're selecting your co-founders are people that are relentlessly resourceful, people that are just going to keep going, people that will find a way. Someone that can learn, someone that's dedicated, determined, will find a way is way more important than anything, than any amount of domain expertise. The other thing on co-founders is you really want someone that you've known before. We actually just did a study on this of our 800 plus companies.
[0:12:22] The co-founders that came together for the sort of express purpose of starting the company fall apart almost 100% of the time. You really want people that have some history together. And again, it comes down to the same thing, actually. When things get tough, and they always do, there is some point in the startup when the expected value dips below the x-axis, usually in the first year or two. And the rational thing to do at that point is say, you know what? This has negative expected value. I'm not going to work on it anymore.
[0:12:52] But if the co-founders know each other, there's this sort of obligation to not let the other person or the other people down. And that often keeps these startups going and sort of like violates this law of nature because everyone's trying to sort of not be the one to give up. And that's really, really important. And then this is what to look for when you start hiring people for the team. There are three questions that I always ask myself when I hire someone as an employee. And that's, are they smart? Do they get things done?
[0:13:23] And do I want to spend a lot of time around them? And you actually need all three of those. And the times that I've compromised and had only two of those three, I've almost always regretted it. And certainly hiring bad employees, I think, is the most painful thing that founders do. And if you do it in the first few employees, it is often fatal to the company. So it's really important. This is your most important.
[0:13:52] The most important thing that you do as a founder is choose your team. And especially the first few people you hire because that sets the tone and they choose the next people. So really spending a lot of time thinking about the people you hire, how they're going to be for the company, is just an incredibly important thing to do. And these are the questions that I think about. Here are four other things that I think are not as obvious or were not as obvious to me. But now that our companies have hired tens of thousands of people,
[0:14:24] We've learned more about what to look for. So good, clear communication skills are really important to have, especially among early employees. Startups that have really great communication are able to execute at a pace far faster than startups that have unclear or muddied communication. Determination is not quite as important for employees as co-founders, but it's still really important.
[0:14:55] Most checkboxes. But just aren't that determined. They're just sort of lazy or whatever. Try not to hire them. One way that we have to describe this is what we call the animal test. This actually works really well in American English. I'm not sure if the expression is the same here. But if people describe someone as an animal at what they do, like she's a sales animal or whatever, that's a really good sign. And those are the kind of people that end up doing really well at
[0:15:25] You know, at startups. Because you want these people that are just all in and incredibly focused on what they're doing. And then the last thing here is, would you be comfortable reporting to them? So this is actually something that Mark Zuckerberg said the last time he came to speak at YC that I thought was just a really great insight, which is he won't hire anyone that he would not be comfortable reporting to if the roles were reversed. And if you hold yourself to that standard for the team that you hire,
[0:15:55] Then they'll be able to attract great people for their team. So I think that's a really good thing to look for. And then the fourth major area that I want to talk about is execution. This is another area that we see a lot of founders fall apart. Execution is the hard part of a business, right? Because these other things happen pretty quickly. And this is what takes up, you know, if these other things take six months, this is what takes the next nine and a half years. It is the relentless.
[0:16:27] Waking up every day and just sort of banging your head against the wall until things work. And everyone thinks they can hire someone to do this. You know, founders always think, well, I'll hire a COO and he or she will manage the operations for the company and I will get to go off and I'll just think about strategy and meet with investors and go to conferences and it'll be great. Obviously, that fails 100% of the time. The value gets created over compounding execution for years and years and years.
[0:16:56] And it never gets easier. But this is what you're signing up for as a founder. Very little of the value gets created by the idea. If you just think about the time frame, most of the value gets created by this part. And so people say that most of what being a founder is about is not being in love with creating a product, but being in love with creating a company. And I think there's a lot of truth in that. There's a big shift after those first three categories all start working.
[0:17:25] Where you shift from the great idea and the great product to great execution. And this is a shift that a lot of founders fail to make. So these are the five jobs that I talk about a CEO having. And that last one in orange, as I just mentioned, is the one that people think they're going to avoid. Because it's hard work and it's the rolling up your sleeves and getting your hands dirty that people don't want to do. And they think they're just going to do the first four.
[0:17:53] But every great CEO that we've been involved with, and luckily there have been a lot, I think at this point more than 30 of our companies are worth more than $100 million each. So we actually have real patterns. It's not just a few outliers or a few hits. All of those 30 are founders that really get obsessed with building these companies that are great at execution. So you don't get to ignore this job. And we've seen a few things that they do, and unfortunately I'm not going to have time to talk about all of them.
[0:18:24] But I will mention a few that I think are really important. Momentum is basically the lifeblood of a startup. Winning teams keep winning and losing teams keep losing. So if you ever slip, if you ever take your foot off the gas and the company starts losing, it's very, very hard to recover. And you do whatever you can to get momentum back. And by whatever you can, you don't give big vision speeches to rally the troops. You have to save that for when the company is actually winning. No one wants to hear it when you're losing.
[0:18:54] But you just have to get these small actual wins and just let them build momentum on each other. And eventually, over time, it'll become this self-reinforcing loop where people feel good, the company's executing, the company's winning, and they want to keep doing that. It's also about always continuing to grow. And this metric is different for different companies. Growth can just be progress on R&D. It can also be user growth. It can also be revenue growth.
[0:19:24] But it's really important. This is also about continual winning and also it's about not deceiving yourself. So the worst thing that we get in our investor update emails are companies that say, well, you know, we're not really focused on growth right now. We're in this other stage. Most companies, we call that circling the drain. Most companies never recover from that. The best companies grow on whatever their key metric is all the way through. Another version of this is that
[0:19:53] This is something an old board member of mine used to say. Sales fix everything. So if you're growing, if the company is winning, if there's momentum, the other problems in the company can be terrible. And sales will sort of whitewash over everything else. So this is what we tell startups when things aren't working. Like, go sell the product. And if it's not good enough to sell, make it better. But don't start worrying about all these other problems because you're not going to fix them.
[0:20:19] If the company's not winning, you're going to continue to have internal political battles. You're going to continue to have pissed off people that aren't working that hard. This is the way. This focus on execution is what will solve all of the other problems. There's one unique risk with execution, which is something that I want to mention. I don't think this is much of a problem or not as much of a problem for New Zealand startups, but it is something that I think a lot of startups in Silicon Valley struggle with.
[0:20:48] Which is it's easy to focus, it's easy to let spending get out of control at a startup. So the obvious answer is don't spend more money than you have and don't have a plan that requires you to raise more money. It's okay to temporarily choose to let the company become unprofitable to invest in growth. But you always want to have profitability within grasp. And this is what we tell our startups. And the way people get into big execution trouble is when they don't have profitability within grasp.
[0:21:18] So it's fine to choose to be unprofitable. It's fine to invest in growth. It's fine to say, you know what? It's going to take us a while to build these rocket engines. And so we're going to not make money until we have that done. But we know exactly how long it's going to take. And here's our plan to raise money. And here's how we're eventually going to make money. That's fine. That's a good strategy. The mistake there would be to say, you know what? It's going to take a while to build these rockets. And we've never done a calculation about how much it's really going to cost.
[0:21:47] And which customers we're going to get and when they're going to start paying us. And that, in that second case, you don't have profitability within Grasp. You have no way to become profitable. Sometimes it's even better than this, which is it's a startup that's spending money on user acquisition, but they can stop spending that money anytime they want, and that will flip them over to be profitable. So if you have profitability within Grasp, then you're just making an execution trade-off, and you can...
[0:22:14] Flip that anytime you want and stay alive without being dependent on a miracle outside of your control. This is something that one of the Airbnb founders drew on a business card, which he then gave to some stranger, which showed it to me. And I thought it was so good that I put a picture of it here. But someone asked him to explain what Y Combinator is and what we do. And so this is how he drew it to answer it. And so there's two gears. You have one that's a product, one that's the market. You're trying to get them to mesh.
[0:22:43] And you're trying to close the gap. And the way you do that is to go meet people. This is almost impossible to do sitting in an office or a co-working space. The only way to close that gap on the time frame that a startup needs to do is to get uncomfortably close to your users, to go spend time talking to your users in person every day and figuring out how you can make their lives better and getting them to be honest with you about how much they like your product or don't. And then the last thing.
[0:23:12] Because this was the thing that surprised him the most. When Airbnb launched, which is now this phenomenal success, they had what they call the 1,000 days of sorrow. Where for 1,000 days, things weren't working. And they were in the flat part of that graph. And that first little spike is when they got some press. They got a few users, but then it fell back to nothing. Then it went down to zero. Then it hung along in this long, long, long period called the trough of sorrow. And then it started to grow, started to grow, started to grow, and finally got really big.
[0:23:42] But that took like five years. And the trick about startups is they take a very long time to start working. When they do start, exponential growth is a magical force and it's great. But don't let yourself get bummed out and don't expect overnight success because startups really are this very long slog. Execution just takes a long time and you have to just keep at it. All right, I am out of time. Thank you all very much for listening to me and I hope to meet all of you later today. Thank you.