This is the latest article published today at http://themacro.com . In this interview, Y Combinator President Sam Altman gave a detailed interview with Facebook founder and CEO Mark Zuckerberg. In addition to a lot of informative content, there are many interesting points mentioned in this interview. For example, how did Facebook experience the run-off of the entire management team, such as how the production of the world-class product, News Feed, etc. Wait.
Sam: Welcome to the "Create the Future" talk show. Our guest today is Mark Zuckerberg. Mark, you have one of the most influential companies in the history of the world, so we are very excited about your arrival.
Mark: What makes me talk? Haha.
| Facebook's early
Sam: Why not talk about it from the earliest days of Facebook? Tell us about what it was like to start building Facebook.
Mark: OK. For me personally, I have always felt that the most interesting thing in the world is how people and humans operate. When I was in college, I was reading psychology and computers. Psychology taught me that there are many areas in the brain designed and serviced specifically to understand people, understand language, communicate with each other, and understand facial expressions. And when I first started watching the Internet in 2004, I discovered that you can find almost anything on the Internet, such as news, movies, music, materials, etc., but for the most important part of people, “the other people and their What is happening is not there.
And I think this is because other information is publicly accessible to search engines, and in order to know what happens to people around you, you need to build a new tool that allows people to update themselves. For example, in order to select a good class, I built a tool called “Course Matchâ€, in which you can enter what kind of course you have chosen. You can also see who has selected this class or they have The evaluation and so on.
And to my surprise, people are willing to spend hours to click on these courses to see this information. This is only the information of people and courses. Isn't it interesting that people are interested in these things? After all, these things are just boring text messages and there is nothing interesting about them. But it made me understand that people have a great desire to know people and things around them. Of course, before I really did Facebook, I did about a dozen other similar little things in Harvard, and these things were eventually integrated into Facebook.
Sam: Did you think of Facebook as a company when you first started?
Mark: When I was doing the first version of Facebook because it was something that I and some of my friends wanted, it was a tool that allowed us to connect with the people around us, and I never thought that would be a company. I was very impressed with the night when the first version was released: I went out to eat pizza with a few friends who are still working on Facebook. We talked about the possibility that someone in the future might be doing a Facebook-like community for the entire world. It will be a great company. But obviously, we didn't think that would be our own.
Sam: There was no thought at all.
Mark: Yes, no. We think that's simply not possible. We didn't want to be a company. We just did something that felt useless to the school.
Sam: So now, looking back, do you think there is anything that makes Facebook different from the other dozen things you do? Become it like this now?
Mark: I think there is only one difference, that is, we have always insisted on keeping Facebook running. Like Course Match and some other course tools mentioned above, they achieved their mission and we will not continue to manage it. And Facebook is like this... People love it too much and use it continuously. I remember that in a matter of weeks, two-thirds of all Harvard students were on Facebook, and people from other schools like MIT began to email us to let us open Facebook in their schools, and we really follow What they said continued.
We didn't really want to be a company when we started. We just kept doing what people needed. This led us to keep going into new schools and eventually to go beyond the school. After a certain degree, when we hired a large number of people, we decided to be a company with the goal of connecting the world. But this is not what we thought when we started.
Sam: Looking back at the beginning, from the early days of Facebook, did you have any experience and advice to share with others who wanted to start a business?
Mark: Yes. I always think that you should start with the problem you want to solve, rather than start with something you decide to start. Moreover, the best companies are those who are determined to bring about some social influences and changes, not those who want to make big money or because they have hands to open a company, even if the social influence is only a small point. .
So I have always felt that this is a somewhat contrary idea to Silicon Valley. In Silicon Valley, people always decide to start a business before deciding what to do. This seems to me to be extremely untrue. For those who really create a business, they know that entrepreneurship itself is a very difficult thing, and what really allows you to stick to it is to believe what you are doing and to know that what you are doing is creating value. And this is the origin of the great company that I think.
| The toughest time for Facebook
Sam: I want to talk about the trough, because I think people never really realized how bad it would be. So can you share the most difficult period in Facebook’s history?
Mark: OK. I think the hardest part for me is that when Yahoo was going to spend a lot of money to buy the company, it was a watershed. Before that, we just thought about what was the right thing and completed it. We opened a lot of schools, opened high schools, and went out of school. We did a lot of photo functions because we felt that this was what we needed to do to help people express themselves and to know more about people and things around us. . But when Yahoo wanted billions of dollars to buy the company...
Sam: How long has it been since the establishment of the company?
Mark: When we were just a few years ago, we had only 10 million users at the time, so it wasn't obvious how much we could be more successful. At that time, we really considered the future of the company. "Wow, what are we doing here are so powerful and meaningful?" We thought. This also caused some interesting conversations between the company and the shareholders. In the end, Dustin and I decided "No, we think we can get out of school and connect more people with more than 10 million users. We can do more. OK, and it really does this."
So we took this as our goal, but it also brought a lot of pressure because a lot of people think that we should sell the company. For many people who have joined startup companies, I feel that I did not do very well in communication. We just appear in the company every day and continue to do what we think is right. So for many early joiners, they are not on the side of us. For them, joining a startup company for a few years and then selling it for $1 billion is a home run. I understand this point, but I also think that if we do not communicate well, what we need to do has caused such tremendous pressure to oppose it. So, the most painful thing is not to reject Yahoo's offer. It was during that time that there were a lot of people leaving the company. They didn't believe what we were doing. If you look at our management then...
Sam: Didn't all the management go at the time?
Mark: In the year after rejecting Offer, all the management went away.
Sam: Did you regret that decision during that time? Have you ever thought that maybe we should sell the company?
Mark: I'm particularly fortunate that what I believe eventually achieves it, and it's achieved very quickly. I think that thing was in the summer of 2006, and we will publish "News Feed" again in the next month. After 10 years, this is one of the most used products in the world. After that, we open up registration for all people in the world, which allows users to grow rapidly. So, within months of rejecting Offer, I think it has already proved that our original decision was correct.
But I think that since then, there have been a lot more difficult decisions to be made. You have to bet on the direction of the company in the future, or you can bet on hundreds of millions of dollars to spend on some things, and maybe 5 to 10 years later on you. Can we know if the original decision was correct? These will be more difficult than the original decision.
Sam: This is what I want to talk about next, but before that I would like to ask, have you ever wanted to sell Facebook since then?
Mark: No.
| How to decide what to do
Sam: Well, so I want to discuss a problem that is very common to the product or the entrepreneur. That is how to decide what to do, how to know when to place a bet on the company, and how to do something completely new? One of the best things I think Facebook has done is to find out what to do and be able to continue to innovate like this, which in my opinion is the hardest thing to do in a company. So what is your suggestion?
Mark: I think the core is to build a company that focuses on learning fast. The company itself is a learning organization, and the decisions you make can make him learn faster or slower. In many cases, setting up a company is like trying to follow a lot of scientific research methods. Try to make plenty of hypothesis inferences. If your experimental design is constructed well enough, you will learn what to do next. I think this is a very important philosophical method.
So within the company we will make a lot of decisions, and each one is fully authorized for independent engineers. What we invest in is the huge experimental structure. Every moment, there is not a single version of Facebook in the world. There may be hundreds or thousands of versions in the world because engineers have the right to try different ideas and let maybe 10,000 or 100,000 users go. Try, and then they will get a conclusion, knowing whether the results of their experiment are good or bad, no matter if this experiment is to make the content in the News Feed better or UI changes or other new features, they can use their own version of the data results To compare with a basic version, the comparison data here may be internal data such as sharing amount, revenue, and activity.
By continually experimenting multiple times and authorizing each person, you can imagine that the progress is not much faster than if every decision was approved. So I think it is critical to build a learning culture and continue to work hard. The longer the time, the greater the help.
Sam: What about bigger decisions? For example, if you make a big acquisition or launch a News Feed product, how do you make this decision?
Mark: First of all, I think if you do things well enough, you don't need to make crazy decisions that look big. So you may need to evolve a bit more with the community. Take News Feed as an example. Although this is a big change in itself, this change has been observed constantly for several years. During this period we have been observing user behavior. So, like at the beginning, Facebook did not have a product like the News Feed that can see other people sharing information.
At the beginning we had only the user page, and then we found that the most common behavior of users was to click around. They will point to different user pages, even as many as a few hundred, in order to see what new changes and updates have been made to his friends' profiles. We learned from the fact that people are more than just interested in finding a person or knowing a person, but are also interested in the person’s daily status update. So at the beginning we made a product that can show which of your friends have updated the page again, so at least the user can know which page the user entered. Then, the first version of News Feed is very simple, basically it is to put the user updated content in order on your home page.
So I think if everything goes well, you can know what to do next by using quantitative data and combining qualitative feedback. Then you can use intuition to decide what method to use to solve the problem, and based on these assumptions to experiment, get more data and feedback, and constantly revised to know how to do it.
We spent a lot of money to buy Oculus. I think if we do it well enough to do this internally, then perhaps we don't need to make a decision to buy Oculus, but the problem is that we are not doing it well. Moreover, the Oculus team is the most talented team in the field so far, so the decision to acquire is justified. But in short, I think that one of your job responsibilities as a CEO should be not to put the company in a position where you have to make mad decisions to survive. Of course, this is almost inevitable. You cannot have an advantage in all things. So, it's better to be brave enough to make big decisions than to be proud and not to admit that you could do better in the past.
| How growth is done
Sam: In terms of growth, I heard you say a few years ago that the best thing you think about Facebook is to invent the concept of a growth team. This is what I've heard from many founders now. Do you think this is still what you admire? How was growth done in the early days of Facebook?
Mark: The ability to make the company's products grow quickly is what I think Facebook did best. The traditional way to grow and market is to establish a communication or marketing team and then buy advertising. I think sometimes this is necessary, especially when you want to send out some messages. However, if you really want the product to grow, I think the best way is in the product itself.
So the growth team we did not have any tricks that other companies can't learn. They just care more about the data and build the underlying data structure, so you can do a variety of different experiments and really try to do the whole community growth, and this is what I think is the most important for network companies. thing.
Sam: How much do you think the growth team contributes to the overall growth rate of Facebook?
Mark: On the whole I think it's very big. I don't have a specific number in my head, but even if I look at only some of the specific features they do, such as "people you may know," you know how big this meaning is.
| How to do Facebook recruitment
Sam: Another great thing that I think Facebook is doing is recruiting, and I've been telling the founders that this is something you must do well. So how do you do the recruitment, what do you see when you recruit?
Mark: If you think about it, you know that when I started the company I was only 19 years old, so I can't think that experience is a very important thing, otherwise I would doubt myself. Therefore, we invest in those who we think have talents, even if they have no experience. They may be fresh graduates, or they may be like CFOs listed on Facebook. His main focus is on product development in a genetic engineering company. Before that, he did not have the experience to lead the company on the market. So, it is to focus on truly talented people.
Sam: So if you don't have a certain experience, how do you judge whether someone else is talented?
Mark: Often you can judge through what he has done before. Everyone has done something more or less. Even if you are only 19 years old, you may have done amateur projects or other interesting things. I think the most important thing is, don't feel that only someone with experience can do something well.
Another good thing that I think we do is give everyone a lot of opportunities, so the company is not only the 19-year-old I started the company, there are many of my classmates or the same people who dropped out. What I am most proud of is that we have 12 different divisions. All but one of them are not managers when they enter the company. They all grew up from below.
Sam: That's too much.
Mark: The only exception is David Marcus, who was a CFO of a $15 billion listed company, so I'm glad he can join and let him take care of a division. But apart from him, all the other people did not report to me when they first joined the team. Some of them were engineers, some were data analysts, and some were product managers. They all grew up to this stage. And I think people can see that you are creating opportunities for them, which allows the best people to integrate into the company and let the other best people want to join because they think "Oh! I also want that opportunity ."
| The next 20 years
Sam: What is the most exciting thing for you in the next 20 years? What do you think will be the biggest change? How will Facebook change and how will the world change?
Mark: We have a ten-year route plan for the three changes we want to see and focus on. "Connect" to allow everyone in the world to connect to the Internet. At present, more than half of the world’s people are not connected to the Internet. I think many people in Silicon Valley feel that the Internet is a matter of course, but it is not. And if we want to solve some of the world-class big challenges, we really need to connect everyone and let everyone have the opportunity to participate in the solution of the problem. So I think connecting everyone is a very critical thing and it will also benefit everyone in the world.
The other thing is AI. I think AI will give tremendous potential to all areas. In Facebook we apply AI to many things, such as letting people see more meaningful content or letting you connect to people you really care about. In addition, in many ways, AI is used to diagnose diseases and find more effective drugs, build self-driving cars, and so on.
I have heard a story about someone who has made a machine learning application that can automatically determine whether it is skin cancer through pictures of skin lesions and is as accurate as the best doctor in the world. So who doesn't want this thing right? Everyone can become the best doctor.
When people are afraid of the potential damage to humans caused by AI and AI, I will be a bit disappointed because I think that in the treatment of diseases or safe driving, AI can save humans and push people forward, so I I think this will be a big event in the next 10 years.
In the end, I think the one thing that will have a big impact is... You know, every 10 to 15 years, the computing platform will have a revolution, and the new platform will allow people to do things they couldn't before. Twenty years ago, most of us were using desktop computers. We used them at work because it made us more efficient, but most people did not use them for entertainment. And now we use mobile phones to help us connect with each other, which is a more humane device.
But after this there will be a new computing platform, I think it will be VR and AR. And I think this will make people more creative, and let people experience other people's feelings more immersive than video and other media. So I am also very excited about this trend.
Sam: So when you started Facebook when you were 19 years old, one of the questions we often hear at YC is "I'm 19 years old this year. I want to do anything that can make the world better. What should I do?" So you're right What are those people who are 19 years old who want to influence the world?
Mark: I always think that what entrepreneurs should do is choose what they really care about, but don't really turn it into a company before proving its feasibility. I believe that if you look at the data of the most successful companies, the vast majority come from such people, not those who decided to start a business from the beginning, because you will always encounter different bottlenecks.
Sam: I totally agree. Asked how long did Facebook really become a company?
Mark: I don't know. I remember that within six months of Facebook’s establishment, Peter Thiel made an investment and became a Delaware registered company. When we first talked to Peter about financing, Dustin and I clearly told him that we planned to return to school. When I set up Facebook, I took up the summer semester and wanted to go back to school and continue to work on Facebook in the fall. When Peter responded, he basically said "Oh, of course." I guess he understands us better than we do. .
Sam: Do you think he really believes that you won't be back to school?
Mark: He must be. Of course, you should still ask him.
Sam: Yes, I will. There will be him in this series of interviews. When I arrive, I will ask him.
Mark: Yes. I think he probably would say he knew we wouldn't go back to school or he thought he could convince us, but in the end it didn't need to be. Obviously, the workload has grown too fast, but we did not drop out of school at the beginning. We started by telling Harvard that he would have to leave school for one semester, and then say that he was going to leave school for one semester, and then he said he would take a year off. After that, we basically decided that we would not go back.
Sam: Speaking of Peter, as the last question, what is the most useful suggestion he has ever given you?
Mark: I think what Peter told me was: "In a world that is changing so fast, your biggest risk is not taking risks." I very much agree with this sentence. I think that many people, when faced with major opportunities, will think of many negative results. Although many of them are right, any choice has a good side and a bad side. If you do not make these changes, I believe that you are doomed to fail and fail. So to some extent, I believe the biggest risk is not taking risks.
Sam: This is a very good ending. Thank you.
Mark: No thanks.
Lei Feng Network (search "Lei Feng Net" public concern) Note: The author of this article Qu Kai, originally contained in 42 chapters (ID: MyFortyTwo) brings you a one-handed translation. This article was authorised by the author to be published on Lei Feng. Reprinted, please contact the authoritative and retain the source and author, and do not delete the content.