Sam Altman (@sama) runs Y Combinator. He was cofounder and CEO of Loopt, which was funded by Y Combinator in 2005 and acquired by Green Dot in 2012. At Green Dot, he was the CTO and is now on the board of directors. Sam also founded Hydrazine Capital. He studied computer science at Stanford, and while there worked in the AI lab.
Perhaps the biggest innovation in venture investing of the past decade is Y Combinator and the early-stage revolution it engendered. Since 2005, YC has funded over a thousand startups, including AirBnB, Dropbox, Gusto, Instacart, Reddit, Stripe, Zenefits, and others.
Since Sam Altman took over as President of YC in 2014, YC has launched a growth-stage fund, expanded the type of companies YC invests in, and established a nonprofit research lab. In parallel, Sam has coached and mentored many of Silicon Valley’s smartest high-growth CEOs, drawing on both his experiences as founder and CEO of Loopt (funded by YC in 2005, and later acquired by Green Dot), as well as his time working with many rapidly growing startups as an investor.
Sam and I discussed an area he has thought deeply about: the role of the CEO, and the hurdles that most often trip up leaders at high-growth startups.
Elad Gil:You mentioned that staying focused as a CEO—how to distinguish between things that seem important but aren’t and things that don’t seem important but are—is top of mind for you these days. What do you see as the role of the CEO, and what are some common mistakes you’re noticing?
The role of the CEO is basically to figure out and decide what the company should do and then make sure it does that. Many CEOs try to outsource those things. Sometimes they want to hire a VP of product or hire a COO and make him or her do everything. But really the CEO has to drive the company’s overall direction. There are a few other things that only the CEO can do, or that the CEO at least has to be heavily involved in, like recruiting and evangelizing the company to new hires, major customers, investors, whatever. And there are some jobs where people only want to talk to the CEO; fundraising is a great example.
But really the only universal job description of CEO is making sure the company wins. And so deciding what the company is going to do and making sure the company gets that done—that’s the most critical part of the job.
Elad: And that statement, if you unpack it, contains a lot of the sub-pieces you hear a lot: Make sure you don’t run out of money, make sure that you’re allocating resources to the right spots, make sure you’re all moving in the right direction.
Sam: The hard part is that most people want to just do the first part, which is figure out what the company should do. In practice, time-wise, I think the job is 5% that and 95% making sure that it happens. And the annoying thing to many CEOs is that the way you make it happen is incredibly repetitive. It’s a lot of the same conversation again and again with employees or press or customers. You just have to relentlessly say, “This is what we’re doing, this is why, and this is how we’re going to do it.” And that part—the communication and the evangelizing of the company vision and goals—is time-wise by far the biggest part of the job.
Elad: As companies scale, I’ve seen people run into the fact that they end up with a lot of overhead. The bigger your company gets, the more time you end up spending on process: Figuring out your sales comp plan and how it should be structured, getting involved with certain aspects of customer support and exception handling, and all that stuff. A lot of people start to lose track of that bigger picture, of where the company should be heading. Are there key approaches to avoiding that trap, where CEOs end up on all the tactical stuff and forget to pull back out?
Sam: Everyone wants an answer like, “Well, you should not do any of the tactical stuff,” but actually a lot of it is really critical. The hard part—and this takes most first-time CEOs a while to figure out—is determining which is which. What is the tactical stuff that seems like a waste of time but is important, and what seems important but is a waste of time?
For example, I do think figuring out compensation structures is really important and something the CEO should spend time on. And it’s something that most CEOs don’t. You’re building what the company measures and what salespeople get paid for, and that’s one of those counterintuitive areas that I do think is really important.
“You have to get really good at saying no.”
– Sam Altman
The trick to being effective at this is that you have to get really good at saying no and just not doing things. There are a lot of things that are urgent but not important. The hard part of being a good CEO is that you have to be willing to let some things fall apart. You don’t have enough time to do everything well. And in practice, what that means is that there are some urgent things that you just don’t do. Getting comfortable with that takes a long time. It’s hard.
Elad: What are some examples of things you think are often urgent but don’t necessarily merit time if you’re scaling like crazy and just trying to keep up with everything else?
Sam: I saw one kind of crazy example recently, a founder from a company that YC is an investor in that is not doing particularly well. I was talking to him about how things aren’t going that well, and he said that one of the mistakes he made is that he has 74 investors. But he was really proud, because for those 74 investors, he responded to every annual audit request. They told him that he was one of the only CEOs that responded right away. He was really proud of this. And I said, “Look, this is a crazy thing. Your company is on fire. You got one not-important tactical thing done while the company is failing, and you feel great about that. All your investors, no matter what they tell you, will be far happier with a big return than you responding to their annual audit request.” And he kind of got it in that moment. But it was this thing that people told him was really important, that he had to do. He felt good that he was doing that, even though he wasn’t doing all these other important things—like, you know, getting users and getting revenue.
The hard part of getting users and getting revenue is that it means spending your time building stuff and talking to users—and the implication is you should do nothing else. That’s not entirely true, of course, because there are things—we just mentioned employee comp structure—that are really important. But there are all these other things that seem important, like responding to investors’ audit requests, that you can just not do.
Elad: It sounds like you just answered the question, but I’ll re-ask: What are the things that you think people should stay focused on as a company scales?
Sam: If the CEO disconnects from the product, that’s usually bad. And that is something that you see happen to varying degrees as a company scales. There are a lot of CEOs that say, “Well, I want to just go think about strategy all the time. I’m really tired of managing people.” Because managing people is hard. And so a lot of CEOs have tried a structure where they bring on a COO (that’s a good idea) and then stop going to the executive team meeting (that’s a very bad idea).
Elad: You mentioned getting disconnected from product. How much do you think that depends on the founding team? How do you think about the broader leadership team versus the CEO as the full driver or primary owner of things?
Sam: I do think there are many examples of teams that have several strong founders, where you see these roles get divided up. And I actually think that can work really well. It’s important to know who’s doing what and have some amount of clarity. But I think it can work really well to have multiple people sharing the responsibilities.
Elad: How do you think the role of the CEO, and what the CEO should focus on, shifts in a downturn? Or alternatively if the company is just not doing well?
Sam: In general, the other thing that CEOs should not lose sight of is that they’re building businesses. And at some point they have to deliver returns and make money. People always say that the CEO’s job is not to run out of money, and what they usually mean by that is fundraising. But the other way you can do that is to make money. And so I think if the company is not doing well or if the environment is really hard, it becomes much more urgent that the CEO—who should never lose sight of the financial performance of the company and the cash flow of the company—is really looking at that all day.
“People always say that the CEO’s job is not to run out of money, and what they usually mean by that is fundraising. But the other way you can do that is to make money.”
– Sam Altman
Elad: When it comes to board management and communication to boards, what is the most important role of the CEO?
Sam: I think the most important thing is that boards hate surprises. And boards hate feeling like you’re trying to hide bad news. You want to over-communicate with boards for sure. Certainly if you have bad news, you want to get that to them ahead of the board meeting
As a general operating principle, I also don’t think it works super well— most of the time, there are occasions when it does—to go into a board meeting and say, “Hey, I’m really struggling with this idea. What should we do?” Not only because boards don’t like that. (And in general they don’t; they want confident leadership.) But because in a board meeting, you usually have all these weird dynamics of different VCs trying to impress each other. You’re actually just much better off to have individual conversations ahead of time when you really want open-ended brainstorming. You’re likely to get better results.
Elad: Are there any other CEO distractions that you see a lot? One thing I’ve noticed, for example, is that a lot of founders equate press with success. So you see these CEOs chasing press, when that’s really not the most important thing they should be doing in many cases. Unless it’s a product like Twitter where the press is really the customer acquisition mechanism very early on.
Sam: Yeah, I think it’s almost always a huge mistake. Twitter is one crazy example—as far as I can tell, Twitter’s best users are still journalists. If your customers are the press, then yeah, go after the press. But most of the time press feels great and delivers nothing. There are exceptions; that’s a little bit of an overstatement. And it’s usually somewhat easy to get press if you’re doing something interesting. But I think most founders—actually, I’m pretty confident in saying almost all founders—overweight the importance of press. So you should do it. It definitely can be helpful, and there’s clear value in it. But you see far more startups make the mistake of orienting their entire company around press than making the much smaller mistake of not focusing enough on press.
Elad: Speaking at events, too—it seems like people will end up on these speaking junkets versus being in the office and focusing on the business or meeting with customers.
Sam: Yeah, a lot of founders fall in love with that. It’s not that hard and you get to travel and your company pays for it. And you feel special. The point that I always try to make is that it’s important to do very small amounts of that. Less is more. If you look at the most successful founders, they’re not the ones that are on the circuit.
Elad: What are the biggest challenges you see as companies scale? And how does the role of the CEO need to change accordingly?
Sam: I think the biggest change people fail to make is that at some point your job becomes more about hiring people and working with them to get what you want done than doing it yourself. The number one failure to scale that we see in CEOs is the failure to make that transition. You need to be conscious as it’s happening, and realize that at some point you’ll get far more leverage on your time if you hire people and work with them closely than try to do everything yourself. And that’s a hard transition for many CEOs to make.
Elad: How do you see people navigating that well, or what are some tips to do that effectively?
Sam: No one does it well the first time. No one is naturally good at letting go of something they care deeply about. So, I think you just have to give yourself permission to screw it up for a while. You can internalize that it’s important to delegate and give people authority, but you won’t do it well the first few times. Just continually try to get better at it.
Elad: Relatedly, I think CEOs often forget that the same thing is happening to members of their team who joined early, maybe just out of school, and who are managing for the first time. They need to learn that skill too. And so you see, at some point of scale, different team members bite the dust in terms of effectiveness. Then the organization limps on until they figure how to either coach that person or bring in somebody more experienced.
Sam: Yeah, this is actually a pretty systemic mistake that I think very few companies get right, which is how you evolve early employees. Obviously I think that’s a really important thing to do. These people have been with you for a long time. They’re probably extremely talented—if you’ve done well, you usually have very good early employees. A) I think it’s the right thing to do for the employees, and B) I think they have institutional knowledge that no one else does. It’s really worth trying to keep them.
Elad: When those early employees haven’t scaled to the point where you want them to, though, what do you do? Do you try to offer them an individual contributor role? Do you offer them more of an influencer role, where they’re, say, a CTO but don’t have direct-line reporting? Do you get a coach for them?
Sam: I don’t think there’s a one-size-fits-all answer to that. What I tell people is just that I think it is really worth more effort than you would normally spend on an employee to try to figure it out.
Elad: Do you have any advice for people who are trying to scale as CEO, but are so busy that they don’t have the time to properly hire the people that they need? Is literally the only thing they can do just take a big step back and not get certain things done?
Sam: I have come to that belief. It’s never easy advice to give or to hear, but I haven’t seen anything else work.
When you’re at these breaking points, I think what happens is most CEOs find some sort of formal or informal coach, often a board member. In my case it was one of my board members. We set up a dinner once a month. I would talk to him, and he was incredibly generous with his time. He was a former, very successful, CEO. And I basically said, “Hey, things are breaking. This is not going to go well. Will you teach me how to be a CEO?” And he did. I think most founders find someone to have a similar dynamic with.
Elad: I think a lot of people too should actively look for that, or reach out to people that will have those additional experiences and are willing to put in the time. You have to be very proactive about it and figure out what do you want from that person.
Sam: One of the hard things is that, as the landscape of investing has shifted, there are fewer and fewer investors willing to really put in the time on one company. And that is really bad in this particular case.
Elad: One way we’ve tried to manage that at Color Genomics is by finding investors who don’t have a lot of angel investments but who’ve built businesses from the ground up. There’s a number of people like that out there, but traditionally people didn’t go after them as investors. People just stick to the same pool in Silicon Valley.
Sam: Yeah, I think that’s incredible and a really, really good thing to do. Founders don’t realize how important that is until things break. And so they go for the big-name investor or the investor paying the highest price. And then a year or two later, everything is on fire and they really wish they had allocated their round differently.
This interview has been edited and condensed for clarity.