Search the Site

Episode Transcript

What follows is Stephen Dubner’s conversation with Facebook founder and C.E.O. Mark Zuckerberg, recorded for the Freakonomics Radio series “The Secret Life of a C.E.O.” It was recorded last summer, long before we learned that 50 million Facebook users’ data had been weaponized by political operatives. Facebook has been the subject of intense scrutiny for years now; that’s what happens when you’ve gone from a college dorm startup to a social network with some 2 billion global users. We spoke with Zuckerberg in Chicago in a trailer, outside an event space where he’d just addressed a few hundred very enthusiastic people who serve as group administrators for Facebook user groups. He introduced new software tools that would help them manage their groups.

You may have heard recently that Facebook has been seriously questioning its mission — that it’s trying, or at least it’s saying that it’s trying, to encourage more meaningful social bonds and less partisanship and discord. That movement was essentially launched on this day, at this talk Zuckerberg gave, in Chicago.

Mark ZUCKERBERG: You know, every day I wake up and I just think to myself, “I don’t have much time here on Earth” or “how can I make the greatest positive impact that I can,” and I know that this is a question that a lot of you ask yourselves too — and it’s not always an easy question to answer. Now, the thing that I think we all need to do right now is work to bring people closer together. And I think that this is actually so important that we’re going to change Facebook’s whole mission, as a company, in order to focus on this.

And now here is our conversation, afterward, in that trailer. A trailer whose air conditioning we had to cut for a quiet audio recording. So picture yourself there: an un-air-conditioned trailer, in Chicago, in the summer, with Mark Zuckerberg. You there with us? Great… here we go:

Stephen DUBNER: Hey! Hey, how’s it going? Really nice to meet you.

ZUCKERBERG: You, too. Good to meet you.

DUBNER: That was really good. You like it? Because it’s a whole set of things that you didn’t use to ever do.

ZUCKERBERG: Yeah, I think it’s important for me to get better at communicating.

DUBNER: Oh my God, the people were so excited when you talked about the tools. Did you expect that level of excitement from them?

ZUCKERBERG: Probably not that level. But I think people are always most excited about the really concrete things that you’re doing. You can talk about the lofty vision and mission and strategy. I actually find you’re lucky if you can get people very excited about that, even if they agree with what direction you are going, because it’s just more abstract. But when you get into a very concrete work that you’re doing, then that’s what I think really energizes people.

DUBNER: I have to say — I’ll be honest with you: I don’t really use Facebook.

ZUCKERBERG: That’s okay.

DUBNER: I mean we use it — Freakonomics, we use it. But I’ll be brutally honest: I don’t want more friends. Or, that sounds bad, too —

ZUCKERBERG: But do you want to stay connected with the friends you already have?

DUBNER: No — well yes, but —

ZUCKERBERG: You don’t?!

DUBNER: Well, honestly, it’s tricky. Because when you’re trying to get a lot of stuff done — I mean, I don’t mean to sound like a total misanthrope. All I’m saying is, I don’t use it a lot. But that made me want to join every group in there, and then I realized that I’d have to quit my job and do nothing but find rare birds and go fishing and support these military families.

ZUCKERBERG: Well, the point isn’t to join every group, but I think it’s really important that people have one or two or three communities in their lives that they really care about. And I think when you have the absence of that, that can lead to real social issues — individually and then for society overall. So, of course these groups, in order to be meaningful, need to span online and offline. There are certain things that you just can’t do online. But the thing that I think is so meaningful and interesting and what we’ve seen is that they do! The people do plan events to come together in person, and the support really does expand out into the physical world.

And a lot of these groups are just things that wouldn’t have been possible physically. I mean we talked about groups for people suffering from rare diseases and almost by definition if you have a rare disease, there probably isn’t someone in your area who has that. So this might be possible for the first time in human history to be able to come together and share your experience of having that condition. I think that’s very powerful. But I think everyone needs to be a part of a few communities that are meaningful to them in order to have that support structure.

DUBNER: So I don’t mean to rain on that parade. I’d love to talk for just a minute about the net effect of Facebook, or social media, or social networking on, let’s say, happiness, to pick a word, or satisfaction. I think there’s a lot of good conversation going on that G.D.P. is a ridiculously bad measure of well-being — it’s way too narrow, it favors all kinds of silly stuff — and that people are talking about either gross national happiness, or some slightly more sophisticated version of that. But it’s really hard to measure it, one; and two, know what causes what. So with Facebook, per se, the upsides that you especially talk about in there, today, are potentially massive, right? Helping communities that wouldn’t be able to find each other, or even identify each other. But I also do worry about the potential downside, which is that, I have teenagers, and there’s always the notion that what people show of themselves online is often them at their best, most buoyant, happy selves. And if I see that and it’s a Friday night, and I’m not with them, I’m thinking, “Man, what’s wrong with me?” And I’m curious if you think about the net effect, costs and benefits on humankind, which is a big, impossible question, but I’m guessing you thought about it.

ZUCKERBERG: That’s very important to us. So the way that I think about this is that technology amplifies human capacity, right. So there are good parts of people, and there are bad parts of people. I believe that on balance people are good, and that therefore amplifying that has positive effects. There’s content portrays people at their happiest, and then there’s content — like a lot of sensational news in some cases — that portrays things as much worse than they are. And I just think that the reality is that both have positives and negatives. There are people who like to point to the negatives of either, because they’re trying to make a point. I don’t think that that’s right, right? There’s a lot of research that shows that the more connected we are overall, the happier we are, and the healthier we are because of that. So it can’t be that talking to your friends when they’re happy is bumming you out, and then reading news that’s down is bumming you out even more, right?

But being an engaged citizen also is not always fun. And a lot of the part of the mission that we’ve been working on is making the world more open and connected. So the connected part is about being connected to more people, and there is generally a lot of research that shows that that is positive for people. But I think that being open is also very important for society. But it can be challenging. Confronting truths or perspectives that don’t fit with ours don’t necessarily make our lives easier in the near term. But I think it’s a healthy thing and an important thing for us to do both as individuals and as society in order to move forward.

DUBNER: So following on that, how do you bring those two notions together which is the fact that we love to connect with people that we either have something in common with or maybe a shared mission, with, as you noted, it’s really important to at least think about or understand a little bit about what people who don’t think like you, why they think like that, and how you can get along. So this is the silo issue, and it exists everywhere and it probably always has and always will. I’m just really curious, it’s interesting: a lot of the groups in there are tribes, and it’s a word that people use kind of positively, and also a lot of them, let’s be honest, they’re kind of — you want a water?

ZUCKERBERG: No, it’s cool.

DUBNER: You sure?

ZUCKERBERG: It’s just very warm in here. And now that you’ve turned the air conditioning off…

DUBNER: Well, yeah, that’s our fault…

ZUCKERBERG: Let’s go with it.

DUBNER: All right. And a lot of the groups in there were about activities. So of course everybody’s allowed to have hobbies that are totally orthogonal to everybody else. But, when it comes to political or social or gender or other affiliations, how do you think about weakening the silos?

ZUCKERBERG: So we’ve both read and just tried to understand as best we can a lot of research around how to promote positive discourse. So there are a few things that I think are pretty interesting that most people may not know. So the first is that if you want to have a debate where people engage productively, the first and most important thing is to first connect with that person over something that you have in common. Right, so if you just go into an internet comment thread and you start debating gun control, that’s probably not going to be super productive. I mean, it can be in some cases. But it’s easy to dehumanize the other people, think about them as not human, not empathize with them. So a lot of what I think social networks can do well, and these communities, are first you connect over something that you have in common. So you recognize that the other person is a person. And I think communities in that way act as a jumping off point.

DUBNER: Do you try to orchestrate that? I’ve read about your efforts to do that and the research that that’s based on.

ZUCKERBERG: Yeah, so we’re trying to work on this, but I think building communities is one of the ways that you can. So a group might come together because they like fishing. But then they go connect over other things, and they debate other things, and they find that, “Hey, we agree on other things; we disagree on them; but now we can have productive and empathetic discussions, because we’re all people, and we recognize our common humanity.”

In terms of just encouraging good positive discourse overall, there are some best practices. For example, one of the ideas that people suggest all the time is, “Well, why don’t you just show people the opposite perspective?” Right, so you have an article that comes up in your news feed about, take gun control, the topic we were just talking about. And a lot of people are like, “Well, why don’t you just show an article from the other perspective?” Well, it turns out if you don’t do this well, if you just show the other perspective, it actually just entrenches people’s belief in their original opinion. There’s a lot of confirmation bias that labels the other opinion as “other.” So you start to tune it out.

What you really want to do is not just present another viewpoint, but you want to give people a range of viewpoints. Because people are smart. And when they have the full picture of what is going on, they can make a good rational assessment for themselves about where they want to be on the spectrum, and what they believe. But it’s really important to not just tell people, “Hey, here’s the other viewpoint, you should look at this.” What people need is the whole picture. So I think that good journalism does that; it doesn’t just try to show one side of a story, but it tries to give the full picture. And when that doesn’t happen, then we can help play a role of at least trying to show a number of different pieces of media that might in sum give the whole picture.

DUBNER: So how much do you care about, or maybe love, social science research, the kind of insights that give rise to these kind of possibilities. So I’ve tried to figure it out, and I’ve read a bit. I know you studied psych for a while, and I know that you hire a lot of—

ZUCKERBERG: I wasn’t at Harvard for very long time, but I was technically a psychology major.

DUBNER: And I don’t know if you know Sudhir Venkatesh, a sociologist you hired? He’s a good buddy of mine.

ZUCKERBERG: Yeah! He wrote Gang Leader for a Day. It’s a great book, I read it.

DUBNER: Anyway, because Sudhir was in … Yeah, anyway. So Sudhir’s amazing, and I love the notion that someone who thinks the way he thought as an academic and as a writer and a scholar — his insights are being applied to something like this, which is accessible to everybody. So talk about how you seek out those kind of things, whether it’s the people, per se, or the research, and how you make it actionable. Because I think I’d like to argue that social science research is having its golden era; I hope it lasts forever. But people didn’t really apply this kind of thinking very much in firms and governments 20 or 30 years ago. It’s really happening, so I want to hear about how you make that happen.

ZUCKERBERG: Well understanding how people are using our services — both in terms of what they want, so we can provide services that meet people’s needs, and understanding what’s good — are really important domains that we want to work on. So we take data analytics and data science very seriously as a company; I think it’s one of the core strategic things that we’ve done well that other companies are seeking to emulate now. But especially because of the context of what we do as a social system, it’s especially important I think to understand.

DUBNER: Who are some heroes of yours from that realm — like Bowling Alone and Bob Putnam? I’m just curious like what you’ve read or thought over the years.

ZUCKERBERG: Putnam’s work shares a lot of the themes that I was just talking about today. He wrote some of the seminal work on community membership and did some of the longest ranging studies on that. It’s an interesting question of where you draw the line between what is social science and what is economics. But I think recently Raj Chetty‘s work is incredibly interesting opportunity and mobility. We’re doing some interesting work together, and we try to team up with folks who are doing interesting work.

There’s interesting research now that shows that the average American has fewer than three close friends who they would turn to in a crisis. So one of the questions that I asked inside our company, and I started a team to work on this is, “Well, can we build some products or services that make it so that the average person has one more close friend?” So not just helping them connect to more people who they know, but, if you could do that, then that seems like a very meaningful change that you could make in the world — one thing that I’m doing this year is traveling around and trying to just see how people are thinking about communities and their work, and —

DUBNER: You’re going to all 50 states, I understand?

ZUCKERBERG: I’m going to 30; I’m going to the ones I haven’t been to yet. It’s interesting so far. But one of the things that I’ve found is that there’s this myth that I think a lot of people have; that if other people in other places just had better information, then they’d make better decisions. And I’ve generally found that that is not true. You know we all lack some information. Of course we can all make better decisions if we had perfect information. But for the most part, a bigger influence is actually who you know, who your friends are, who your family are, and how they help you filter the information that you have.

I can give a few examples of just how … this is a really poignant example. When I was in Ohio, I sat down with a group of heroin addicts, and one of the things that was really interesting is when you’re going through recovery, the first thing you have to do is detox, of course. But then after that, the next thing you have to do is basically get new friends. And it turns out if you remain friends with anyone who you were using with before, then you are very likely to end up back using heroin and endangering your life. So it turns out it’s not that these people don’t know that it’s bad. Or that they don’t want to end up addicted to it. But it just is that the people who you’re friends with — having those close friends, the three or four folks in your life are just so important.

Another example that has really stuck with me is when I visited a juvenile detention center. And one of the facts there that’s just mind-blowing is if you go to a juvenile detention center … and some of the kids are there because they committed what you’d call a crime, they stole something or hurt someone. But some of them were there just because they misbehaved in class a little bit. And going to a juvenile detention center dramatically increases your chance of becoming a criminal once you get out, because what you’re essentially doing in that center is building a social network that reinforces itself negatively. All the examples that you’re getting are other people who either have criminal behavior or are misbehaving, so kids who who might have just been okay, a little not behaving as well as they should have in class, are getting all the wrong lessons and friends.

So, making it so that we can have a positive social network I think is actually one of the most important things that we can do for growing opportunity in society. That’s certainly what Raj Chetty’s work is. So that’s definitely a big thing that we study and think about how we can improve.

DUBNER: Let me ask you something in his work that he found that was surprising and interesting: do you know about the moving to opportunity research that was done years ago? And all the first round of scholars that looked at it, they said it didn’t work, and he came back with a colleague and found out that actually it did work if the kids were younger when they moved, because by the time you’re 15 or 16, your patterns are pretty set. You think school sucks or you don’t. But if kids were 9 and under, I think it really worked. So to that end—

ZUCKERBERG: Actually he found a linear correlation. So if you moved when you were 9 you got half the impact of moving when you’re born, and if you moved when you’re 18 that’s kind of the end of the impact.

DUBNER: Right. So given that and given that the communities you’re building are presumably mostly for adults, I gather, what do you think about that?

ZUCKERBERG: Well, people over the age of 13 can use Facebook.

DUBNER: Okay, this is a corollary to a question I often think: people talk about early education — and I’m really interested in the project you’re interested in, because I’ve known people who’ve done that. I just think that’s a smart way to think about it: use technology to customize because people learn differently. People have different abilities and so on. But even when you talk about good early education, the kids who end up doing worst in this country and most other countries are already doing bad by the time they’re 1 or 2; they’re born into circumstances that are just really, really hard to surmount. So, without putting all the world’s problems on you and Facebook —

ZUCKERBERG: This is what Priscilla, my wife, focuses on. She’s running a school which is focused on the intersection between health and early childhood education. She’s a doctor and she wanted to help kids. And then through her pediatrics program realized that education and health are so intertwined, and that you need to start educating the parents, from the time that they’re pregnant, about what the right behaviors are, and then you basically want the kids in the school or in a program, or at least to have good habits being built from birth, and have them involved in that as quickly as possible.

When we think about education, we often think about concepts like math or reading. But very early on, when you’re learning how to walk, health is completely intertwined with education, and then of course as you go up through your education — it’s hard to learn math if you can’t sleep at home, or there are different issues. So there are all these interesting trade-offs — you could do a whole podcast with her talking about — she has students who come in and who are in an environment at home where they can’t sleep as well. So she has to make this trade-off — they have a nap time during the day, and if a kid is sleeping and just sleeps for four hours instead of the hour and a half, does she wake them up to do math, or does she let them sleep? And I think a lot of the time what ends up being the case is that the health is the precursor to education, so you let them sleep. But they’re totally intertwined in the way that you’re talking about.

DUBNER: So Facebook is obviously not a government. You don’t have an army as far as I know. Do you? Right?

ZUCKERBERG: No, we do not.

DUBNER: But in some ways, it’s become a nation-state in the way that we used to think about nation-states. Except that it doesn’t provide those services, and it doesn’t use monopoly of force and so on. But what I mean is you probably have —

ZUCKERBERG: Well, it’s a community.

DUBNER: Okay. But it’s a global community, organized by interests, activities, and it’s voluntary. So to me it’s — right, nation state is an exaggeration. But what I’m getting at is this: governments throughout history and especially now, try their best, I would argue, to help their people. And they often don’t do a very good job, because the structure of government turns out to be pretty suboptimal and the incentives kind of weird. In a way, Facebook, it strikes me, has more leverage over how people actually organize and live their lives, right? The choices they’re able to make, the information they’re able to get hold of.

And so I’m curious how you think about that. I know you were an accidental C.E.O. and an accidental social entrepreneur, but it strikes me that you’re working really hard to take this massive accidental enterprise really seriously and optimize it for the most number of people. So I guess my question is big and lumpy and impossible, but I really just want to know what that feels like, because, look, I don’t know how to read what you’re trying to accomplish here. You’re this incredibly smart and accomplished young guy with this incredibly big and impressive company. And maybe you’re just trying to make it bigger and better and that’s it. But it doesn’t read that way to me. When I read your letter to the global community —

ZUCKERBERG: Yeah, well, I never started this to build a company.

DUBNER: Yeah. But I wouldn’t have — knowing what I know about you, if I looked at you 10 years ago — I also wouldn’t have thought that you were necessarily in it to help fix society or help make society better. Maybe I’m just wrong.

ZUCKERBERG: Well, I think at each point you try to do the best you can with the position that you’re in. So 10 years ago, I was just trying to help connect people at colleges and a few schools. And that was a basic need, where I looked around at the internet, and there were services for a lot of things that you wanted. You could find music; you could find news; you could find information, but you couldn’t find and connect with the people that you cared about, which, as people, is actually the most important thing. So that seemed like a pretty big hole that needed to get filled, and maybe it’s more functional and more basic, but that was the thing that needed to happen.

Now you know we look out at the world and we say, “Okay, we’ve been focused on making the world more open and connected.” And I always thought that that would be enough to solve a lot of problems by itself. And for some it has, but the world is today more divided than I would have expected for the level of openness and connection that we have today. So now I just believe that we have a responsibility to also work on that. So you can paraphrase what we’re working on now is, “open, connected, and together.” So that is basically the idea that we’re talking about, when we say bringing the world closer together.

Here’s another way to think about it. There are lots of different issues and things that help bind people together and make us stronger as a whole than the sum of our parts. A huge part of that is the economy and our jobs and all that. And Facebook is a big player there, but we’re a relatively small part of the overall world economy. But when it comes to helping people build communities, I’m actually not sure that there are many other institutions in the world that stand for building communities and have the tools to be able to empower people at as large scale to do that. So that just strikes me as something that’s, “Okay, if that’s a unique opportunity that we have, then we also have a responsibility to go do that.” And that’s a little different than where we were 10 years ago, when there were many social networks that were bigger than us. We were just at schools and all that.

DUBNER: The tools that you talked about today; obviously you’re giving some user data to the users. What does that represent in the path of Facebook sharing its data generally. And I realize that what you’re giving to the users is useful. I love that when you announced them one lady actually said, “Statistics. Woo!” I’d never actually heard people cheer for that.

ZUCKERBERG: I missed that. But that makes me smile.

DUBNER: Yeah it was awesome. And obviously you’re not sharing, you know, income—

ZUCKERBERG: It’s aggregated. It’s insights into how people are using groups, right, so basic demographics —

DUBNER: And who wouldn’t want that, right? If you’re the admin.

ZUCKERBERG: Well you need to present it in a way that’s actually useful.

DUBNER: Yeah, but I’m sure there are people who want you to share much more data about your users. Yes?

ZUCKERBERG: Well I think one of the interesting challenges that you find running a company or a community at scale is there are people who want things that are completely conflicting. So there are certain people who want us to share more information, and then there are a lot of people who really don’t. For some of these social decisions that we have to make, I find that the right place to be is when you’re getting yelled at from both sides equally. And you try to just make the best decision that you can on this.

But, both parts have good arguments. Of course, privacy is extremely important, and people engage and share their context and feel free to connect because they know that their privacy is going to be protected. On the other hand, if you’re trying to enable people to build communities, giving them some insights into how people engage in their communities in an anonymized way that that isn’t sharing anything about the individuals and the communities, can help them do their job, and help bring more people together, and help people’s lives as well.

So you try to just do the best that you can and know that there’s not always a simple and optimal solution. And another dynamic that’s interesting is that sometimes the balance of what people want shifts over time and that enables opportunities to do more in one direction or the other, that wouldn’t have made sense before.

DUBNER: There’s also, as you’ve pointed out, people don’t know what they want. We’re really bad at predicting or —

ZUCKERBERG: Oh, I don’t know; I don’t believe that!

DUBNER: Well in the revealed preferences, in what people actually do, you can see what they want. But if you ask people what they would like, the social science research at least says that declared preferences, there turns out to usually be a really big gap between that and revealed. No?

ZUCKERBERG: Well, I would say that in a lot of these discussions that I have, people focus on what we as Facebook are doing. The real secret to why this works well is because we focus on giving everyone else as much power as possible.

DUBNER: What do you mean by that. Meaning just give users power to use it as they wish?

ZUCKERBERG: Yeah, give people the freedom to share as much as as you can; give people the ability to get access to as much opportunity as possible. And there’s a whole spectrum on this. First, in order to be able to use a tool like this, you need to have access to the internet. Which is something that we take for granted here in the U.S., but more than half of the world doesn’t have access to the internet. So we work on basic things like improving the business model of telecom operators. Or we’re designing solar powered planes to beam down access to the Internet, because that’s a basic thing.

Then once you have the internet, there’s the whole legal framework. We are very active in advocating in many countries to give people the freedom to share more, and express more of what they want. The U.S. is somewhat of an outlier on having constitutionally protected freedom of speech in a way that very few other countries do. Every other country has many more restrictions on what you can say than you can in the U.S. So that gets in the way of people’s freedom, and we are active on pushing on that.

And then only when you get through these basic foundational and legal frameworks, do you get into the tool. Which is, in the U.S. people can have the freedom to say what they want to anyone who they want, but that may not help you so much if you don’t have a tool that actually enables you to reach other people with your opinion. So that’s a thing that Facebook and the internet have really worked to change over the last 15 years. Now we really are in a world where anyone for the most part can write something and share it, and if it resonates, it’ll get shared widely, and it can start to change opinion broadly. But in many ways, that ability is the practical arm of free speech. That just didn’t exist before.

But there’s this whole spectrum of things that you need to do, and that, that’s the thing that we’re hugely committed to. That’s why when we rolled out the mission today, the basic idea behind it and the vision, is to bring the world closer together. But the reason why that isn’t the whole mission is because it was really important to me that the mission focused on empowering other people. So the mission actually is to give people the power to build community and bring the world closer together, because there’s no way that we’re going to do this, no matter what we do — you can ask me all the questions you want about what we’re going to do, but it’s actually going to be other people doing this, and we succeed when we empower other people.

DUBNER: Now a cynic would say, “Well, sure it’s in Facebook’s interest. The bigger they build a global community, the bigger and better the company is.” Which is not untrue, but it’s the prerogative of every company to grow as big and as profitable as they can. So let’s say that someone puts on their “I doubt the do-gooder part of you” hat. How do you respond to that?

ZUCKERBERG: I think a lot of people just can’t get out of their own way. So I think for a lot of companies and governments, they would do better by giving people more freedom, and they don’t for whatever reason. So you may be right that it is strategically the right thing to do, but that doesn’t mean that everyone is doing what they should do. So I just want to make sure from the mission of the company on down to how we execute and think about our strategy that that’s always front and center.

And a lot of times people like to think about, “Well what is the impact that we can have by improving this product?” I really want to train our organization always to think about what is the impact that we can have by giving these people more power and freedom to go do what they want. So this just goes back to your original question before around, “Do you believe that people can make good decisions for themselves?” And I deeply do. I really think that things end up better when you give—

DUBNER: I don’t mean to imply that people can’t make good decisions for themselves, although on some dimensions I would argue they can’t. Like with diet, with health particularly. I think you’ve written a little bit about this, that the leading causes of death in the rich world are all essentially self-inflicted. Or at least a lack of optimizing how you take care of yourself. And you could say that people think the tradeoff is worth it.

ZUCKERBERG: I’m not sure I’d say that.

DUBNER: That the tradeoff is worth it?

ZUCKERBERG: No, that the largest causes are self-inflicted. I mean, cancer…

DUBNER: Cancer truly T.B.D., because we really don’t know yet, about most cancers, what causes them. That’s the problem. The environmental causes and behavioral causes, I wish we knew, because then we wouldn’t have as much cancer. But cardiovascular — just take the biggest one —

ZUCKERBERG: It’s also largely a function of age.

DUBNER: It’s true. It’s true. And look —

ZUCKERBERG: I’ll debate you on this one point.

DUBNER: All right. And one reason why we have more cancer now than we should is because people are living longer, not dying as early from cardiovascular deaths, which is great. There’s always a silver lining. But the economist Gary Becker from the University of Chicago, years ago, he was the guy that started all of this in terms of turning economics into a more interesting social science. And he argued once that all deaths are suicides to some degree. Because none of us actually really optimize staying alive long, because life’s too fun and interesting and challenging for that. So I think we all make trade-offs all the time, and I think that that’s what being human is about, and it’s maybe fun. No? You’re shaking your head.

ZUCKERBERG: I disagree with that, too.

DUBNER: Yeah? Tell me why.

ZUCKERBERG: Well, I think that having a sense of purpose is the thing that brings us both happiness and health. So if you’re framing it as “doing stuff that’s fun leads you to your demise,” I think there is a lot of research that would suggest the opposite.

DUBNER: Yeah, I agree with that. Fun meaning cheeseburgers and French fries and not taking care of oneself. That’s what I mean by fun. That’s a shallow version of fun. All right. So let me just ask you: I loved the Reid Hoffman conversation in that piece.

ZUCKERBERG: Yeah, that was fun.

DUBNER: It was really interesting, and I loved how he framed it. And you were obviously really good talking with him. You said something on there that I wanted to ask you about. How many versions — or whatever the proper noun would be — of Facebook are running at any given time? And just explain that to people who use it, and what that idea represents.

ZUCKERBERG: Sure. So one of the basic strategies of our company is to learn as quickly as we can. That is more important to us than any specific strategy of, “Okay, here’s how we’re going to build the best messaging app, or here’s how we’re going to build the best news feed,” is building a company that is just agile and learns as quickly as possible from what people are telling us.

So the best way to learn is to basically try things out and get feedback. So if you just have one version of Facebook running, then that constrains how much people can react to. So we build this whole framework that allows people within the company, any engineer, to change some code, create a new branch of what Facebook is, and ship that to some number of people. Maybe 10,000, some small portion of the community in order to get good feedback from that. And there are a bunch of rules around a bunch of things that you can’t ship.

DUBNER: And how is it related? I assume that if I’m the engineer and I want to do that, I do it with someone, with conversation and approval.

ZUCKERBERG: Some, there are definitely guidelines. There are things that, if what you’re doing is sensitive to people’s information at all, then of course there are a bunch of checkpoints that you need to do before doing that. But people try out different ideas for how to suggest you better friends, or suggest you better communities, and that doesn’t need to go through a lot of process of the company; people can just try those out, and we’re trying out hundreds of different versions of things like that. And the idea is that cuts through red tape at the company. So now a given engineer, instead of having to get their manager and then their manager’s manager and then me on board with changing the app, they can just do it.

And then at the end of that test, they get all this feedback back that is both quantitative — so how their version of Facebook performed on everything that we care about: how connected do people feel; how much do they feel informed; how happier; all these different things — and then we get qualitative feedback back as well. And if their version is an improvement, then we roll that in, and then that becomes part of the trunk version of Facebook that now everyone else is measured against. So now every day, we’re running lots of different versions to see what’s best and what people respond to. But again it gets back to this strategy, which is the real company strategy is to learn as quickly as possible what we need to do in order to bring the world closer together.

DUBNER: How about a couple lightning-round fast questions. What’s one story that your family always tells about you?

ZUCKERBERG: That’s a good question. You’d have to ask them.

DUBNER: But you probably know it too. It’s like, “Oh yeah there was a time that Mark did—,” because I love you know, I love the stories you talk about with Reid. ZuckNet was awesome. I also love that the snowball fight game, where you could have a real snowball fight but it’s a lot more—

ZUCKERBERG: Yeah. I think my sisters were happy enough to play the games that I programmed growing up, because it was better than what we would do physically. So they prefer playing a snowball fight game, or some strategy game that I made, even if the graphics were terrible, and the game wasn’t that good, because I was still learning how to program; I was like 12 or 13 or 14. They’d prefer that to getting chased around the house with a Super Soaker or something like that. So there’s that.

I think my dad has a lot of fun stories about how he got me into technology. And he’s a dentist, but he was always very focused. He took a lot of pride in being the first dentist in the area who did digital X-rays instead of physically. He was just such a geek, and he loves this stuff. And he didn’t really know how to program, but he was just like, “Mark, don’t you think this is cool?” So that stuff I thought was pretty good.

DUBNER: You obviously have to make a lot of decisions all the time and there are a lot of different ways to make decisions, and deciding to not do something is often much more important in retrospect than deciding to do it. So other than deciding not to sell Facebook early on — which I’m guessing was at least a little tempting — what’s the best decision you ever made to not do something, or not pursue something?

ZUCKERBERG: So you’re asking about a discrete, big decision, but I actually think the most important thing is what decisions and what process, on a day to day basis, you choose to let people have the freedom to do, and just not get involved with. So a huge part of how Facebook works is giving a large amount of freedom to our engineers, the company, and to people who use the product to make with it what they will, and trusting people to do that. So there’s this balance of how much is it going to be my ideas and my will, versus the people around us and the company. And I think having some restraint there ends up being very important.

DUBNER: Was that hard for you to get to, or?

ZUCKERBERG: I think it’s hard every day. When you’re running something, you of course have the ability to make as many of the decisions as you would like. So the real art I think is not when you know that you have someone who is a superstar, who is going to make great decisions, but deciding to let people do things that you disagree with, because on principle you know it’s just going to free up more creativity and people will feel like there’s more potential to try different things in the future that may be better, if you let them go do those things, even if you disagree with them.

DUBNER: That’s really admirable. I would think it’s hard. I think most people would have a really hard time doing that.

ZUCKERBERG: I believe a lot in giving people freedom.

DUBNER: I believe; I do. I mean I believe in the belief, but I just think that’s a hard thing to do. All right. Last question quickly: if you weren’t doing this, if this hadn’t worked out, if MySpace had become Facebook, what do you think you’d be doing?

ZUCKERBERG: That is a really interesting question. I’ve always really cared about the idea of connecting people and bringing people together. And the way that you do that is different at different points in history. Another way that I think about it is: if Facebook didn’t happen, you’d just drop me in a desert now what would I do? I believe that technology is a huge lever for improving people’s lives. It’s a great thing that an individual can sit down and construct something and share it with millions of people around the world; almost nothing else other than code and technology gives you that ability to do that. Maybe producing media on top of that technology, but the technology is the platform for that.

But today a lot more people message each other than just use social networks, which is why we’re very focused on Facebook Messenger and WhatsApp as well. So if you got started 13 years ago in a dorm, the right thing to do is to build a website for social networking. Ten years ago, or seven years ago, maybe the right thing to do is build a mobile app for social networking. Now I think one of the most important things that you do is build tools for more private communication, because people have the power to do that.

I think that line will always be shifting, and I would bet that you know at any moment that you would want to get started, you probably could. And there are always going to be new ways that people want to share and connect and feel supported, and there are always things to build. I’m just a big believer in technology, and bringing people together is two of the most important levers that we have to make progress as humanity.

DUBNER: All right. No offense — and I know it’s hot in here and you’ve got to go — but you really didn’t answer my question, which is just literally what you think, if this had gone very differently, like you couldn’t do it differently —

ZUCKERBERG: So differently that I wouldn’t want to be an engineer any more? My answer is that I would build whatever the next thing is. I still think you can care about the mission, but Facebook is not a one product company at this point. And you know there are new social network companies that get started all the time. I’m not sure exactly how I would think about it.

DUBNER: So there was no impulse to become a dentist, for instance?

ZUCKERBERG: No, that stuff makes me queasy. So I never had the whole doctor thing. And Priscilla’s got that covered for the family. But no, I believe a lot in technology. I think that there are lots of different ways to get started. Our path as Facebook I think is good proof that the line is not clear. I started it as a website for Harvard students to build a community there. There was no news feed, none of the stuff that you think of as the most important parts of what Facebook are today. So you start with something; you find a niche; and then you can grow it to serve more people in that way. That’s what I care about.

DUBNER: All right.

ZUCKERBERG: Thank you. It was really fun.

DUBNER: Yeah. Thank you very much.

ZUCKERBERG: Alright, we gotta get some A.C. in here.

Freakonomics Radio is produced by WNYC Studios and Dubner Productions. Our staff includes Alison HockenberryMerritt JacobGreg RosalskyStephanie Tam, Max Miller, Harry Huggins, and Brian Gutierrez. For this series, the sound design is by David Herman, with help from Dan Dzula. The music throughout the episode was composed by Luis Guerra. You can subscribe to Freakonomics Radio on Apple Podcasts, or wherever you get your podcasts. You can also find us on TwitterFacebook, or via email at

Read full Transcript