Jeffrey Meyerson: Move Fast: How Facebook Builds Software

Transcript

[0:00:36] JB: Love it or hate it, Facebook has done incredible things. Although a lot of narrative stories have been told about the company, little has been said about its strategies until now. Jeff Meyerson interviewed a collection of former and current Facebook engineers and with additional research, turned those conversations into his new book, Move Fast: How Facebook builds software. On Author Hour today, he shares how the company navigated some of its earliest challenges, how it’s able to maintain a creative and innovative culture, and why there’s so much to be learned from Facebook’s playbook even if we don’t agree with 100 percent of it. Hi Author Hour listeners. I’m here today with Jeff Meyerson, author of Move Fast: How Facebook Builds Software. Jeff, thank you so much for being with us today.

[0:01:34] Jeffrey Meyerson: Thank you for having me on Jane.

[0:01:36] JB: Okay, first of all, why Facebook? You call this book a case study. Why Facebook?

[0:01:43] Jeffrey Meyerson: Facebook is one of the definitive products in our lives and as a company it’s one of the definitive companies of our generation and yet, there has not been substantial coverage of how Facebook actually builds their products as opposed to companies like Amazon or Google that have had more books written about their processes. Facebook has more been in the subject of literary scrutiny as opposed to reverence or curiosity. I like the idea of exploring Facebook from point of view of somebody who is interested in building products.

[0:02:22] JB: Okay, this book has kind of a unique origin story. How did it come to be? Can you tell us a bit about who you are?

[0:02:30] Jeffrey Meyerson: Yeah, I started a podcast called Software Engineering Daily six years ago. I cover a lot of engineering subjects and two of my listeners were Facebook engineers named Pete Hunt and Nick Shrock. They developed some of the foundational technologies that have made Facebook really popular in the open-source community and I was having a conversation with them and they said, “Hey, you should do some shows on Facebook engineering. We think there’s actually a lot of secrets within Facebook, like open secrets, not things that they were keeping confidential.” Basically, practices and engineering methodologies that could be uncovered if you were to dive deep and to interview some people. I said, “Okay, cool. I’ll interview some people, we’ll do some shows.” I did five or 10 shows on Facebook engineers and then we were talking about it more and we’re like, “Actually, it seems like there’s some real significant material here and I think we can actually go even deeper than just 10 podcast episodes. I think we can do an entire book.” Actually, I think this was Pete’s idea. In any case, all three of us just sort of said, “Yeah, there’s definitely a book here. Let’s put a book together.”

[0:03:44] JB: Wow. It does go much deeper than what was in the podcast. How did you approach all that additional research?

[0:03:54] Jeffrey Meyerson: Well, first I should say that the team at Scribe, the company that put us in touch are phenomenal at helping shepherd a first time author through the process. Now, I’m not being paid for this promotion but Scribe has been really, really helpful in shepherding me through this process but my process would basically, I had these transcripts of interviews and I was just delving into the transcripts and looking for interesting quotes and organizing the quotes and thinking through what are the themes that the quotes highlight and constructing a book from those quotes.

[0:04:32] JB: Let’s get into it. The title comes from a kind of quote that was part of Facebook’s ethos when they first started “move fast and break things” and that has become a pretty provocative phrase, can you tell us a little bit more about it?

[0:04:48] Jeffrey Meyerson: Yeah, the reason I titled the book that is because it is misunderstood to a hilarious degree. Move fast and break things, the idea behind that phrase is essentially that we’re building in a product space where there is so much potential that we need to explore it as quickly as possible and we’re not exploring it quickly because we’re happy to throw caution to the wind and yeah, maybe like a little bit reckless, we’re doing that because we are in a massive greenfield space and we need to capitalize on this rare opportunity. The question of what you do in that modality is a very novel engineering question because it’s rare that you're in a situation where you have massive greenfield opportunity, massive user adoption, massive financing resources and it really puts you strategically in an interesting position. If strategically you’re going to take the poise of we need to move fast then that’s going to have all of these downstream effects of how you're going to manage the company from a point of product strategy, financial management, strategic management, onboarding your search for products that are going to make money and so move fast is really about this modality.

[0:06:22] JB: Okay, but then they did update the phrase, is that right?

[0:06:27] Jeffrey Meyerson: They did, they updated it from “Move fast and break things” to “Move fast with stable infrastructure.” I think that shift epitomizes the recognition that Facebook had reached product market fit and profitability. I think they were profitable when they changed to move fast with stable infrastructure. Whether or not they were profitable, they had a large and growing cash cow. I think they were saying, “Okay, look, we’ve got this thing, it’s actually working quite well. Let’s transition to a strategy that’s a little more sustainable because we know we’re going to be around for the long haul, we’re going to continue to explore stuff, we’re going to continue to explore aggressively” but we now have this foundational product that is kind of like a dial tone for a lot of people and we don’t really want to shake up that dial tone at all. We want to explore in a way that maintains that dial instability.

[0:07:18] JB: You’ve broken the book up into three sections, product, culture and technology. You say and you just mentioned the word strategy a second ago. You write that this is a book about strategy. How did you come to determine that that product, culture and technology would be your main sections?

[0:07:41] Jeffrey Meyerson: that was a process of pulling back from the lower-level details of the book and trying to see it from a higher level and trying to pull it together in a way that would make sense. What I found was that you could organize a product strategy in terms of product, culture and technology because I think product effectively comes fir ew products that are adjacent to that first product and once you have those first two things as a foundation, then you have engineering and the engineering is informed by the culture and the product. I was able to lay out the guts of the book underneath that rubric.

[0:08:44] JB: Okay, starting with product, the book begins in 2011, it’s almost an oral history of sorts, spanning the early odds, would you say that’s right?

[0:08:56] Jeffrey Meyerson: Yeah, definitely.

[0:08:59] JB: The first struggle you tackle is when Facebook was switching to become a mobile company after iPhones had come out. What can we learn from the way the company maneuvered that platform shift?

[0:09:15] Jeffrey Meyerson: I think that the lesson there is, when your cash cow is threatened, you need to immediately start looking for a new way to make money in a sustainable fashion. The longer you deny that your current cash cow is going away, the harder it’s going to be to discover a valuable new alternative. I’ve heard Jeff Bezos say that you never want to be in a position where your company has to throw a Hail Mary and especially as your – once you get to a later stage of a company. You don’t want to be backed into a corner. It’s not like Facebook was going broke when their desktop display advertising business was sort of threatened by the ship to mobile but the writing was on the wall, that the world was shifting to mobile. There was this Facebook F8 event, I can’t remember what year it was but it’s in the book, where basically, they’re presenting all these products at F8 and then they realize, halfway through the F8 conference that we’re not really doing that much on mobile and it was this recognition that the company needed to shift lots of resources into mobile and everybody knows that Facebook has gone through the shifting of read in the software industry that has started Facebook a little bit with a passing glance. If you understand just how much work it took to shift Facebook to mobile and how nascent the mobile tooling was, how hard it was to make that shift. By the way, this shift was kind of hard to compel, especially before there was a real mobile advertising business that was doing well because people didn’t really know if mobile advertising would work, sounds crazy but people just didn’t know. During this period of time where Facebook was looking for its next cash cow, it was evaluating things like gaming, it was evaluating things like hosted infrastructure, that’s why it acquired parts and eventually, they just realized, actually, “Okay, this mobile advertising unit is going to be our salvation.”

[0:11:33] JB: I realize when I asked that question, I was a little bit leading in away. I phrased it as what can we learn from and I should ask you, is that the goal? How much of this is oral history and how much of this it is case study with the intention to be emulatable?

[0:11:49] Jeffrey Meyerson: The case study and the story telling aspect of it serve to drive a through line of entertainment, I wanted the book to be entertaining and I think it tries to accomplish that by exploring anecdotes and characters within the Facebook story that have not been really highlighted in the past. By framing the book around that entertaining storyline, it creates ample opportunity to do these many case studies, these many lessons, these many takeaways and whether or not you disagree with those or whether or not you find it useful, I hope you find the storyline throughout the book entertaining.

[0:12:29] JB: Okay, A little bit of both?

[0:12:32] Jeffrey Meyerson: There are some exciting stories for sure, you also cover Facebook’s response to Google plus? Yeah, absolutely and you know, that story has been told a lot so I wanted to tell it in a way that seemed a little bit fresh and I thought that basically the best Google Plus storyteller is actually this guy who is kind of a controversial figure right now. This guy, Antonio Garcia Martinez but he was at Facebook during the time of Google Plus and he’s quite a good storyteller and then from the engineering point of view, you have Keith Adams, who also told me about his experience during Google Plus. It’s kind of like this Death Star moment where Google appears with Google Plus and it’s basically a product that looks like it is designed to kill Facebook, that’s exactly what it looks like. It looks like basically a copy of Facebook except it’s kind of got this fresh Google tinge to it and I remember being in college when Google Plus came out and I kind of used Google Plus for a while. I was like, “Maybe I don’t need to use Facebook, maybe that’s fine.” You know, back at Facebook headquarters, it was war time like the engineering strategy completely shifted to, “We need to defend against Google Plus. We need to iterate so quickly, we need to build products and morale so quickly that we just completely crush Google Plus” and they were successful.

[0:14:10] JB: I’m just thinking maybe I want to ask you a little bit about the second section of the book now, culture. What is Facebook doing with its culture that it’s unique or emulatable?

[0:14:24] Jeffrey Meyerson: Facebook does a few different things with culture. I think there is sort of the before you get into Facebook and the after you get into Facebook phase. You know, the hiring processes is extremely astringent and before you get hired by Facebook, the hiring processes is kind of adversarial just like a lot of Silicon Valley places. They want to make it hard for you, for you to get into Facebook. They want to make it very, very challenging because you know, navigating engineering challenges and solving technical problems is quite difficult but once you get into Facebook, Facebook does a lot to make you successful. The way that they do sort of cultural indoctrination and I say that in the most friendly terms, there is nothing wrong with their indoctrination strategy but it convinces people of the Facebook way. They take you through this bootcamp process where essentially you get introduced to all the ways that Facebook builds software. You get introduced to the Facebook tools and services and internal engineering practices. You’re going to build some software with the assistance of people that have worked at Facebook for a long time, so it is a very comfortable safe environment and it is this great leveling field because whether you have a PHD or you’re a brand new graduate from a computer science school or you have been in the industry for ten years, if you are coming into Facebook you’re going to go through bootcamp and you’re going to solve stupid little bugs and so you have this great equalizer that’s at the beginning of the Facebook engineering journey and then in order to find a job that will be a good fit for you within Facebook, they have this thing called headcount. Headcount is an allocation of engineering resources that every team gets, like I might be on the ads team and you know, during my quarterly planning, I may have three people allocated to me in headcount and then when bootcamp happens, these people get hired into bootcamp and they don’t know what they’re going to work on yet and they are floating around in bootcamp and somebody with headcount can basically go into – it’s a little more complicated than this but somebody with headcount could basically go into bootcamp and say, “I want to take this person from this pool and see if they’d be interested in working on my team.” It becomes this kind of sales process or negotiation process to the extent that I understand it. It’s a negotiation process between these teams that have headcount and these pools of engineers that are at bootcamp and when I heard about this, this is the first time that I have heard about that process and I thought that was a really cool process. I’m not sure if it’s entirely true, if I have it entirely accurate because it sound almost utopian. As an engineer who has worked at companies where I was assigned work I didn’t really enjoy, the idea of having this bootcamp where you essentially get paired with work that you are going to enjoy was kind of utopian to me, so I’m not sure as to what extent is true but this was I thought a defining feature of Facebook’s culture.

[0:17:37] JB: That’s so interesting. I mean returning to the beginning of your response that it’s really hard to get in but then once you’re there, you’re kind of taken care of, I feel like that’s what I’ve always heard about Harvard University that it’s like impossible to get into but once you’re there, it is actually pretty easy and of course, Zuckerberg went to Harvard and now of course my mind is wondering about the – and then –

[0:18:09] Jeffrey Meyerson: Well, I mean that’s kind of like any super exclusive club, right?

[0:18:14] JB: Yeah and then this whole utopian process you’ve just described also sounds a little bit like college like you get in and you get to try different things and pair with what fits best.

[0:18:26] Jeffrey Meyerson: Absolutely. I think what a lot of people who go into the software industry miss is the days when they were working all day on something that they were passionate about and something where they could explore ideas that they were interested in and a lot of people wind up in jobs in the software business where they’re not really that happy with what they’re working on and I think Facebook does as good a job as any big corporation I’ve seen with pairing engineers with interesting work.

[0:18:58] JB: That’s important, certainly, yeah. When you were doing all of these interviews with former Facebook engineers, what did you learn that you found particularly either surprising or exciting that you didn’t know before?

[0:19:15] Jeffrey Meyerson: The extent to which Facebook is able to hire really, really smart people, people with amazing personalities, immense personalities. I interview a lot of engineers. I’ve interviewed probably 1,500 engineers over the last six years of doing my podcast and the interviews with Facebook engineers’ standout not just because these people are astoundingly smart. They have distinctive personalities. If you go and interview a cadre of Google engineers, I’ve interviewed lots of Google engineers, the average personality there is somewhat different than the average personality of Facebook. For whatever reason, when you apply some filtering criteria at scale, you’re going to get a distillation of that hiring criteria and I find Facebook engineers to have different kinds of personality than Google engineers. I talk about a little bit in the book but this, think about the foundational seeds of Facebook, it came from this guy hacking on a random project that he may or may not have stolen from somebody effectively, an idea that he sort of stole. I mean I don’t think he really stole it, he was very inspired by previous ideas but he was just hacking on it and kind of like making something that he thought people would be really excited to use, they’d be really intrigued by as oppose to the Google founding story, which is two graduate students working on this deep computer science problem search that has been evaluated for many years but hasn’t really been cracked. You have two very different founding environments. I think that that has shaped the kind of person that goes to work at these companies. At Facebook, you have these scrappy creative engineers that may or may not have gotten a degree in computer science and at Google, you have these people who are really well credentialed and it leads to very different philosophies. I would say that’s one of the things that really stood out.

[0:21:34] JB: That’s so interesting. Did you hear from Facebook? Is anyone going like, “Hey, stop giving away all of our best practices.”

[0:21:43] Jeffrey Meyerson: Not yet. I mean I wasn’t like super public about the book. You know, I was just interviewing a lot of the engineers were ex-engineers from Facebook. They are people that are no longer there and the other thing is like the book is actually very complimentary of Facebook. I mean I’m kind of a – I’m sort of this pro-technology company as you can find I think and so I was never really thinking of this as yellow journalism. I was just trying to find some cool engineering practices from Facebook, so I doubt I will fall into any kind of conflict with the company.

[0:22:19] JB: Speaking of that, Facebook has received a lot of criticism especially since the social dilemma came out. How did you navigate that or choose not to?

[0:22:31] Jeffrey Meyerson: Facebook is a company that is on the frontier of what is possible and anytime you’re writing that frontier, whether you’re a software entrepreneur or a musician or an artists or a filmmaker, you’re going to have critics and the more prominent you are, the more critics you’re going to have because the more you’re going to shake up the status quo, the more you’re going to alter people’s lives in ways that may make them uncomfortable. That’s just what Facebook does because of the scope of its impact. I mean you take out Instagram for an hour a day, it’s this weird product, it’s affecting your brain in weird ways, if you don’t use it responsibly it can effectively destroy your life. That’s kind of high stakes and so there’s no way you can, if you’re building a product that is that high stakes, there is no way that you can kind of walk around unscathed. You’re going to have controversy.

[0:23:34] JB: It seems like that’s not the focus of your book.

[0:23:40] Jeffrey Meyerson: Well, I mean I talk about it a little bit. I mean part of the book is and in fact, at its essence the book is meant to be a little bit irreverent in the sense that I don’t even really pay attention to the fact that Facebook is a “controversial company” because a lot of those arguments are pretty dumb and whether or not they’re dumb, it’s like they get in the way of thinking about like what is it about this company that is able to produce so much productivity and human ingenuity. Like it or not, Facebook is an incredibly creative and innovative company and what this book is saying is that there’s a lot to learn from what Facebook is doing.

[0:24:27] JB: Well, that’s seems like a perfect stopping point. Jeff, it’s been a pleasure speaking with you. Again listeners, the book is, Move Fast: How Facebook builds software. Jeff, in addition to reading the book, where can people go to learn more about you and your work?

[0:24:42] Jeffrey Meyerson: You can follow me on Twitter and checkout Software Engineering Daily.

[0:24:47] JB: Great, thank you so much.

[0:24:49] Jeffrey Meyerson: Thank you so much Jane. Thanks for hosting me.

[0:24:52] JB: Bye-bye. Thanks for joining us for this episode of The Author Hour Podcast. You can get Jeff Meyerson’s book, Move Fast: How Facebook builds software, on Amazon. You can also find a transcript of this episode as well as previous episodes on our website, authorhour.co. Make sure to subscribe to The Author Hour Podcast for more interviews and insights into life-changing books.