Startup School: An Interview With Mark Zuckerberg


Facebook founder and CEO Mark Zuckerberg has taken the stage at Startup School, where Y Combinator’s Jessica Livingston is interviewing him. I’m liveblogging the interview below.

Mark Zuckerberg: “I love being here. These are like, my people.”

Q: I want to go way back, before Facebook. What did you learn from those experiences?
A: I mostly built stuff that I liked. When I got to college I started messing around with other programs. There’s this story —  I was making Facebook the week before finals, and there was a class where you had to learn all these pieces of art. I was supposed to be studying, but instead I was building Facebook. A few days before the exam I was screwed. I took all the images, and made a website, where you could add notes to each image, and it was a ‘study tool’ where everyone else filled in all the notes that I needed to pass the class. After that the professor said it had the best grades of any final he’d ever given. This was my first social hack. With Facebook, I wanted to make something that would make Harvard (and more open that) more open.

Q: How were the first users using Facebook?
A: Looking people up, but it was so simple. There were no messages. You could look at profiles, poke people. Everything else built over time. People say launch early and iterate, Facebook is clearly a good example of that. YC has a shirt that says “do something people want” and I think that’s a great way of looking at it.

Q: Tell us all the dumb things you did.
A: Where do you want to start? *pause* What kind of stuff do you want to start with?
We weren’t even set up as a company at first. I started it with different friends at Harvard who were really smart, and they didn’t have the same levels of commitment. I moved to Silicon Valley, lots of folks didn’t want to move out. A lot of the early founder group was fractured. I didn’t want to be involved with setting up the business at all. We had this guy Eduardo . Instead of setting up standard company, we set up as Florida LLC. I don’t know all the things wrong with that, but lawyers out here said that was number one to unwind. In the beginning we weren’t trying to make it as big as possible. We wanted to provide value. Instead of launching schools that would be most receptive, we did least receptive. We launched at Stanford, Columbia, Yale where each of them had their own community already. When we launched Facebook at those schools and it took off, we realized it could be worth putting our time into it. My friends are people who like building cool stuff. We always have this joke about people who want to just start companies without making something valuable. There’s a lot of that in Silicon Valley. We wanted this to be valuable.

Q: But that’s a problem for startups — it’s not valuable til people start using it (chicken/egg problem).
A: Facebook is inherently viral. There are lots of sites that include a contact importer, and for lots of them it doesn’t really make sense. For Facebook it fits so well. It wasn’t until a few years in that we started building some tools that made it easier to import friends to the site. That was a huge thing that spiked growth. Before that organic spreading on campuses. One amazing thing, we launched at schools with most people asking for it. We didn’t have enough money for all the schools, we had servers for $85 a month, kept getting more as we needed them. At Dartmouth, half the student population signed up in one night.

Q: Who were your first investors. How did you pitch them?
A: I’m not good at pitching anything. I never eally pitched. There’s Eduardo who did some. Moved out here Peter Thiel was our first investor. Sean Parker helped set us up with first outsourced accounting firm, introduced us to Peter. There were already hundreds of thousands of users. It was clear that if we executed we would continue to do well. I was 19, I doubt I was at all impressive. For the first two years the only advice he would give me was “don’t mess it up”. Gets back to the “build something people want” motto. Early on we were clueless, all of us came from having users and growing at a sustainable rate. When we were first meeting with Peter, we didn’t have Facebook.com. We were TheFacebook. That’s a winner.

Q: What would you do now different?
A: I’d get the right domain name. The moral is that we could get the domain. We ended up “tens of thousands” for the domain. I think we used Register.com which was also a mistake. The core people at the company a lot is still in tact. Management team has improved over time. Was hard to get people good at managing businesses early on. Now we have Cheryl, but early on there’s no way we could have gotten her. There’s nothing wrong with making these mistakes and not getting it right at each step along the way.

Q: What did you do differently at Facebook?
A: We didn’t realize what we did differently because we didn’t have context. I look at some companies that call themselves technology companies, and there are lots of managers who aren’t technical. So we were always committed to maintaining technical people in the company. One of our major marketing roles is an engineer. That kind of tech culture is important to have in the DNA of the company. Google is a tech company that I really admire. And as we’ve grown it’s clear how we’re different. I look at Google and think they have a strong academic culture. Elegant solutions to complex problems. We pride ourselves on strong hacker culture, building things quickly for lots of people. We have small team — ratio of engineers to users is by far more than any other company (1 mil+ per engineer).

Q: How do you keep this environment?
A: We do try to attract people, but our goal isn’t necessarily to keep people forever. Some companies are really good at training people. A lot of people for a long time went to IBM because it was great to learn sales. We want Facebook to be one of the best places people can go to learn how to build stuff. If you want to build a company, nothing better than jumping in and trying to build one. But Facebook is also great for entrepreneurs/hackers. If people want to come for a few years and move on and build something great, that’s something we’re proud of. Steve Chen when he started working on YouTube was working on Facebook. They left, did something cool. I’m not encouraging people working at Facebook to leave. We’re not pretending that we’re building a company that hackers would want to stay at forever.

Q: I ask lots of successful founders did you know it would be as big as it is? Tell me a couple moments when you went, woah, this could be huge.
A: I guess it just kept on growing, so our expectations did too. The interesting thing about the Facebook story. A lot of founders say they have some huge vision. A lot of people I hung out with at Harvard was this trend for more and more information becoming available, this was leading to increasing openness. Didn’t think we’d have a chance to influence it. So I did this little project, which over time grew. A few years in, we were sitting around and said this maybe could impact this. So I think probably around the time people kept using Facebook after they graduated from college. We started to realize it was something that was pretty universal.

Q: Did you ever have a scaling issue?
A: Yeah. We limited growth in the beginning because we didn’t have money. As we got more money we’d rent more servers for $85 a month. When we first took money we decided the biggest thing we could do was accelerating user growth. When economy when south we decided to get cash-flow positive. We got around lots of problems early on but limited growth on purpose. Some of the stuff like photos was pretty crazy. With photos there was a 2 month period where people weren’t sleeping much in the company. That was kind of crazy.

Q: What have you learned about running a big company?
A: Not clear that I’ve learned much (joking). Tech companies when they’re small can move so much faster. It’s been reinforced for us how much we need to keep moving really fast even as we get big. Otherwise inertia slows thing down. Last time at Startup School. People tell you “you can’t do this, there’s all this stuff you have to know”. My message last time was that no, don’t let those people tell you that. I think it was taken out of context that I don’t value experience at all. I think a lot of you are here because you want to take your shot at something and build something.


open to questions

Q: We know so much about people’s history because of Facebook. I want to know how you’ll make this available to people 2000 years from now. It’s just bits. How will anyone know what we were up to when it’s all digital.
A: I like to pride myself on thinking pretty long term, but not that long term. So you win. Marc Andreessen told me a long time ago when he was Netscape Microsoft had longer time horizon which helped them win. Back to your question. We feel we’re part of this movement to become more open. I think a lot of companies have this sense of purpose, but that might not be forever. Microsoft just released Windows 7. How much innovation is happening with operating systems? There was a huge period of time where the OS innovation was most important. Now I think pushing more opening is going to be most important. I think the answer to your question, if tech stays strong people will be building comapnies to take this into a long time in the future.

Q: One of the stories I heard about you was when you made changes. You were reported as saying “the most disruptive companies do what they need to do. What was behind that rationale”
A: One saying we say a lot is the biggest risk you can take is to take no risk. In evolving world if you don’t change you will lose. Change is really disruptive for people especially when it’s a web service, people aren’t opting in to the change. On a website you don’t want to keep forking your code. You want to push people to use one version. When I was talking before about making the company move quickly. I’ve heard values are worthless unless they’re controversial. We’re willing to give up a lot to make sure we can move quickly. We want to have one code base (we did Facebook Lite, which was a bunch of Y Combinator folks who did that).

Q: Can you talk what it’s like in order to speak publicly so people don’t pounce on you? Seems like you’ve changed since 2006.
A: Have I not said enough offense stuff? Is this less interesting?
Q: Yeah… probably less interesting (oooooh from the crowd)
A: Maybe we need more controversial questions.