a16z Podcast - a16z Podcast: Improv'ing Leadership

Episode Date: March 23, 2018

Leadership is not just about management, but about passion, a bit of humor, and resilience. General partner Peter Levine and Dick Costolo (entrepreneur, former CEO of Twitter, and erstwhile comedian) share their thoughts on the topic in this episode of the a16z Podcast -- based on a conversation recorded as part of the BreakLine program (hosted at Andreessen Horowitz) preparing military veterans transitioning into tech careers. Among other things, Costolo shares what running Twitter was like pre-IPO and after, as well as what it’s like to suddenly find yourself thrust onto the world stage; the role of improv and imagery in leadership; and the difference between preventing mistakes from happening... and correcting them as quickly as you can when they do happen.

Hi and welcome to the A16Z podcast. In this episode, Dick Costello, former CEO of Twitter and A16Z general partner Peter Levine, talk about all things leadership, how to do what you love, why a background and comedy improv might be more useful than you think, and the importance of resilience. The conversation was recorded as part of the breakline tech program for military veterans hosted at Andrewson Horowitz. Last time we met, you were still CEO at Twitter, and we haven't met since. So what have you been up to? And what was the departure like? When I told the company, I was leaving, and one of the employees asked me what I was going to do next. I was like, first thing I'm going to do is sleep in until 10 a.m. So I didn't do anything for six months except travel around. I went to Cuba, went to Africa, sort of all over the place.

And then I'd have a strange background in comedy, and I'd always wanted to go into comedy. So I contacted the HBO folks who are doing the TV show Silicon Valley and said, hey, I want to come work on Silicon Valley. And the CEO of HBO, Richard Pepler, said, well, we need a technical consultant, but we can't, like, pay you to do it. And I said, I'll just do it for free. He was like, oh, okay, great. That works for us. Anyway, so I went and worked on season three of Silicon Valley for that year. So I was in the writer's room the whole summer, which was awesome.

and it's just funny to see the perspective that the people in Hollywood have about like Silicon Valley and the way they view it differently than the people here view it. You did improv early on in your career and then you did tech and how are those two, are they related? Like how did one overlap and the other? You're the only person in Silicon Valley who kind of has that background from my understanding.

Yeah, no, I think that's true. So the back... You need to tell some joke or something. What happened was, I was getting computer science, University of Michigan and at the time it wasn't in the engineering department it was in the sort of humanities department back when I got it because I'm old as fuck so I decided I had to take a class in my senior year that allowed me to have time to work on my operating systems class because my operating systems class was super hard and it involved tons of coding and I had to

spend tons of time on that so I thought well I'll take an acting class because that'll be like super easy and I won't really have to do anything and I'll just go up and do scenes and stuff and can spend most of my time on my operating systems class. And I had a ball doing that my first term senior year, so I decided I was going to take another acting class and start doing like stand-up at the student union on like open mic night and stuff. And that went really well and was super fun. So I decided when I graduated not to take any of the programming job offers I had and instead try to go to Second City in Chicago and do improv comedy and get into Saturday Live from there because that was sort of the stepping stone to SNL for everybody at the time and still is in fact.

I remember I got there my first day and you're doing this Second City Training Center and Steve Correll and I and Rachel Dratch were all in the same group. So he and I and Rachel know each other from like, God, 30 years ago now. Anyway, long story short, you know, I eventually, several years later got my audition for SNL and didn't even make it to callbacks. It just got sort of nuked in the first round. And then audition for Mad TV, which was on Fox at the time and didn't get that. I was like, all right, got to go get a programming job and we went back into tech from there. And amazingly, years and years later, the improv background and doing improv in front of hundreds and thousands of people without having

a script totally serves you well being a leader. One, first thing you learn in improv is that listening is the most important thing you can do on stage because you've got to be aware of what else is happening in the scene and what else other people are doing. Otherwise, if you go out there with some preconceived idea about where the scene should go and it's already going in another direction, it's not going to make any sense. I mean, the people who are the best listeners turn out to be the best improvisers. Well, 80% of what you do as a manager and a tech company in Silicon Valley is gather feedback and listen to your team and the people who go into the room thinking that they already know what the answer is, and making decisions that create misery for the team

aren't based on the right information or full information or maybe based on one stakeholder's point of view and they don't end up being very good managers or leaders. The second thing was it was very easy to do things like interviews. I remember on first CNBC interview, the person was like, hey, this is going to be live. Is that okay with you? And I was like, I've been booed on stage at 11 p.m. at the Adelaide Opera House by 2,000 drunk Australians. Like, I'm going to be fine having a camera here while we talk for five minutes about advertising. You know, all these things that you do that seem like they're totally random end up helping you as a leader. I teach at the GSB here, and often my students come to me and say, hey, like, what should I

do? After I graduate, what should I do with my career? And the simplest answer I give them is do what you love as opposed to what's expected of you because many people come out of whatever environment there's all these external pressures you come out of a school you come out of a program whatever and people expect you to do certain things and we tend to bend in those directions because of these external pressures and so you're famous for doing what you love and not what's expected and how do you get over that what's your advice there i couldn't agree with you more my commencement speech at University of Michigan in 2013, I basically told them the exact same thing you just said. I said, listen, you all got here by meeting and exceeding expectations.

You know, you get into the University of Michigan and you graduate from it by meeting and exceeding expectations. The problem is now there are no longer expectations. And if you do what you think you're expected to do or supposed to do, instead of doing what you love, you know, when things go wrong, as they inevitably do, you'll be standing there frozen on the stage of your life wondering, well, now what am I supposed to do? And instead, if you do what you love and things go wrong, as they inevitably will, you become resilient. You're like, all right, well, this isn't the acting stuff. I just, all right, well, that didn't work. I'll go do another audition. Okay, that didn't work. I'll go do this. And then, you know, bounced around and did that stuff for seven, eight years and didn't make any money,

but it was awesome. I was having a ball performing. And every time I took a big risk in my life that wasn't the right thing to go to next, it totally paid off for me and worked out long term. And, you know, the other problem with doing what you're expected to do is, you know, people always tell students, you have to go make an impact in the world. And I always thought, like, I don't know what to do with that. I don't know what I'm supposed to do. And if I think back on the things at Twitter or the things that I did in my life that ended up making an impact or having a big impact, I didn't even know they were making an impact while we were doing them. I remember one time Medvedev, the president of Russia, was at Twitter, and we were like scrambling around trying to deal with him being there that day, and, you know, there were all these security dogs in the building, and the building landlord didn't want dogs in the building, and the CIA was like, you're going to have freaking dogs in the building.

And here, you're running around, and, like, it's a nightmare, and people are, like, trying to get an autograph from him. I'm like, he's not going to give you an autograph. He's the president of Russia. anyway the next day in the front of the new york times was like you know medvedev greets obama on twitter and there's the exchange greetings on the platform and i was like oh that was like a big day in the world but at the time i was like shit stop asking for an autograph so i just remember that as one of those great examples of you can't think about that stuff in advance and try to plan it out it's kind of like what happens when you look over your shoulder as like oh that

with those things we did made this big impact. So you just have to do the things you want to do. If you're not doing what you love, you're not really going to be great at it. And there's going to be somebody else who's doing that job that you think you're expected to go do, who loves that job, who will crush you, right? Because they really want to go do it. And so the answer is actually so simple, go do what you love. No, you're right. I remember I saw, I hadn't seen Steve for like 25 years. And he gave a talk here at the Lucille Pact. or Children's Hospital, and I brought this Chicago Tribune review from 1986 of our show at Second City, this picture of him and I and the other six people in the group.

And I went up to him afterwards and showed him the picture, and he patted me on the back. You know, I was the CEO of Twitter. It's like 2012. And he patted me on the back and said, I'm sorry it didn't work out for you. Consolation Prize, right? Consolation Prize. Maybe let's switch to tech. You know, maybe when you started and I started around the same time in tech, there was no

internet at the time. It's just coming out, right? Like news groups, now it's everywhere. So, you know, if you were to sort of look ahead and say, what consumer trends might be out there? What are you saying? What are you here? I think it can be overstated or over-dramatized,

but the machine learning stuff is just so important. I mean, even when I left Twitter, it was really starting to impact our ability to deliver the right kinds of things to people. It's powering half of Twitter moments, which is instead of having to craft your own timeline, which takes a lot of time and effort and going and finding all the accounts to follow.

Moments just kind of uses human curation and machine learning to deliver the right kind of stuff to you. All that is going to be fascinating and important and have really profound implications for lots and lots and lots of domains. I mean, healthcare, we're going to look back 10 years from now and go, oh my God, we let humans who had only seen four cases of this decide what to do next instead of running it through, you know?

the symptoms and evaluating against terabytes of historical data that was stupid you know how did you guys how did people live you know we'll think about it that way and i think that's going to be real like wild and amazing and have lots of cool implications and ethical implications you know the alpha go machine beat the world go champion lise adol move 37 in the second game alpha go made this move that all the expert commentators were like oh it screwed up it made a mistake look it made this stupid, random move that's obviously a mistake. And, you know, surprise, surprise, it turned out to be this, like, genius, beautiful move and won the game.

And I remember thinking, kind of like the throw to the Eagles quarterback this weekend. I kind of like the Eagles quarterback. But, you know, it's one of those, there are going to be lots of cases of in specific domains of, it did this thing a human would never do, and there will be lots of machine learning examples of, well, the machine thinks we should do X and the people think we should do Y, but the machines will prove to be right more often than not. So I just think that's going to be fascinating and lots of interesting implications there for tech. Cool.

Now you're not at Twitter. You know, you obviously learned a tremendous amount there. And in my role here, I always try to figure out, like, what makes a great CEO? As a board member, you look at CEOs, everyone is different. Yeah, extroverts, introverts, technical, not technical, inside out, upside down, this, that, and the other, some are great. Some are not great. Some are great. Lots are not great.

Lots are not great. Lots are not great. What in your mind makes like the great CEOs? Man, I can talk about this for hours, so I'll try to summarize. One, they're super resilient. Look, as I've said already a couple times, things inevitably go wrong. Like, they just do. Some things happen, you know, you make stupid decisions.

You're like, God, why did I, you know, in hindsight you're like, oh, that was such a stupid decision. Great CEOs are resilient. I always told the team and the Pixar president at Catmills. President, Ed Catmill says this as well in his great book Creativity, Inc. It's not your job as a leader to prevent mistakes from happening. It's your job as leader to correct mistakes as quickly as you can when they happen. If you try to lead by preventing mistakes from happening, nobody's going to frigging take any chances and the company will slow down to a crawl because everyone's going to go, I don't want to get in trouble for, you know, something might go wrong if we do that. And everyone's going to start going around and asking for permission and then nothing gets done. So it always tells my managers, like, don't try to prevent mistakes from happening, correct mistakes when they happen. And I always tried to make sure that I admitted to the team when I screwed something up. Like, hey, the last time we did this reorg, it took too long. When reorgs take a long time, people start jockeying for position and being really political. I totally screwed

that up. It cost a bunch of brain damage for everybody. Sorry. So people knew, like, it was okay to make mistakes. I'll tell you a great resilience story, a specific one. So we're doing our IPO for Twitter in November 2013. And it took us like seven years to get to $16 a share. And then it took us two weeks to get to $26 a share, and then a day to get to $47 a share. The company didn't get five times better in the last two weeks than it did in the first seven years. So just remember, we now have a stock that will go up and down. Never had that before. So, you know, and you try to, like, prepare the team for that and just get them to be resilient to it,

but you have to keep sort of reminding them of that stuff as bad news comes in. Or tried to always tell them, like, you know, we're not as good as we are when people think everything is going great, and we're not going to be as bad as things seem when people start writing bad articles about us. What was it like to lead Twitter, let's say through movements like the Arab Spring or the natural disaster, the tsunami in Japan?

What is it like when you're there? And there's news, right? It all comes to you guys. So I'm having my Monday morning staff meeting one day and Kelly, my chief of staff walks in and she's like whispering, phone call for you. I'm like, I'm in the middle of my Monday morning staff meeting. And she goes,

it's the White House, you know, and I was like, what do they call me for? What did I do? And so there was this weird, you know, time period where a bunch of us out here had to start to realize, wow, we're like in the thick of these international discourse and political conversations. And Egypt, you know, during the Arab Spring, we sort of started in Tunisia with people organizing protests on Twitter. And they were able to organize protests on Twitter because they were writing under pseudonyms and not, you know, their actual names. So people weren't able to come over to their house and arrest them. And then, so spread across North Africa and then Egypt shut us down and we work with Google to be able to allow people to use Twitter over the, you know, phone networks and through sort of various proxy servers they set up, et cetera. That was amazing to go from being like a, you know, computer science guy and leading a tech company to involved in these conversations in the White House. And when the White House call, I'm just curious, like, what do they... Oh my God, I remember. So it was Valerie Jarrett and she's like, look, the president wants to have a conversation about a bunch of stuff related to, you know, the way.

these platforms are being used and how they're being used. So a bunch of tech CEOs go to the White House, and we're sitting in the Roosevelt room, and President Obama comes in, and we all sit down, and unfortunately, he asked me the first question. So I'm like, stupidly, stupidly, all I do is answer the question,

and then he says to someone who I won't name now, so-and-so, what do you think about that? And the person goes, first of all, Mr. President, I just want to, I think I'm sure I speak for all of us, want I want to thank you for inviting us here today and I'm like fuck and so you know and so

while they're in the middle of that I go wait and everyone in the room stops and looks at me and go and I go I meant to say that anyway so good thing so there's a Brian Roberts of Comcast

he like hit me in the need he was like you're in huge trouble it must be so interesting all of a sudden you're in the middle of geopolitical stuff that you have no experience background in and you know like you're a platform for facilitating communication and like everyone expects you to have all the answers like hey what's going on in Tunisia like you have some inside information because you happen to be running this platform it's like really interesting there were all sorts of moments like that at one point so there were some things going on

if I remember correctly in Turkey it was that people were tweeting sort of blasphes things about Adeturk, the founder of the country, and in Pakistan, they were tweeting some things that were sort of, you know, against the speech restrictions in Pakistan. And we got a fax from Pakistan. And someone came into my office. The head of trust and safety came into my office was like, hey, we have a fax from the, you know, so-and-so department in Pakistan. I was like, we have a fax machine? I was like, where is it? I didn't know you could still get those, you know? And I was like, how do we know it's from Pakistan? you know. It's just so weird. And they're like, how do we, what do we, do we, do we fax them back? What do we

do? I mean, so it's like, you're trying to find people in the country to deal with about this communications issue, but you're like, you don't have anyone on the ground in Pakistan. It's just like you're, you know, you're fumbling around trying to figure out how to do that. As a result of that, was there a whole group inside a Twitter eventually that was sort of government interaction? We put together a government group. We had a team in D.C. that's trying to learn everything about how Washington works. How do you fax Pakistan when they fax you? Dealing with things like

everything from that to onboarding global leaders, right? You know, Modi and India on the platform and Prime Minister Abe and Japan. Those are, I mean, talk about nerve-wracking. The Prime Minister of Japan wants to meet you when you're here. You're like, oh my God, I've

fumbled the ball in the Roosevelt room. I'm going to go to Japan and do that correctly. I'm like, you know, you see your trying to learn all the things you have to have the protocol and the way to come into the room and who's supposed to speak first. I'm again a computer science guy. Like you're trying to just figure the stuff out as you go. Right, right, right. Cool. All right, let me open it up for questions. I've noticed that the most successful people also have a really amazing capability to tell stories. And you're known as an amazing storyteller. Thank you. You're welcome. So what is your approach

or is it all improv? No, it's not all improv. One of the things I used to like to do at Twitter is because I felt like imagery was important and powerful is I'd put an image up in front of the company and tell a story about that image and then when they would see that image like, oh yeah, that's your representation of this story. So, you know, 2014 comes around. Sure enough, as I said, the night of the IPO, the stock went to 47 and I got up in front of the company and gave that speech. By the end of the year, the stock had gone to 75. In fact, the last week of the year of 2013, while I was, you know, had taken a much needed vacation with the family at the end of the year after the IPO. The stock went up $5 a day, every day for no reason. Like it's, I don't know, it was the end of the year and people were like giving other people Twitter shares for Christmas or something. So I got back and our stock was at $75 was at $75 was a stock. Again, keep in mind, a month earlier, it was like, it's going to be a $16 a share stock. It's not at 75. Not much has changed at all. And I walked in the office and looked at Kelly and I went, we're so fucked. Like, this is going to be a horrible year, because, first of all, over holiday

break, all my employers are multiplying their number of options by 75. I now have this, in three years, I'll have this much money when all my options are vested. And I, like, you can tell people until you're blue in the face, hey, we're going to now have a stock that goes up and down. They're like, uh-huh, this time 75 is this much. Right? When it goes up to 100, it'll be this much. So I should look at these kinds of houses. So, you know, it's In 2014, you know, as the news cycle inevitably starts to get to be, oh, what's wrong? Like, why isn't it eight times better than it was last year?

I put the image of Michelangelo's David, just the statue of David in front of the company, and I gave this, like, long talk about it. And I said, listen, during the Renaissance, all the other Renaissance artists who portrayed the David and Goliath, they represented it in sculpture as David having beheaded Goliath and standing on his head with his sword at his feet. You know, Virokio, Donatello, and these bronzes

with David standing on Goliatho. Elias had. And when Michelangelo portrayed the David, you know, he's got the sling over his shoulder and the rock in his hand, and he's like tense, his neck muscles are tensed up as he's obviously looking in the direction of Goliath. And art historians like to say that Michelangelo wanted to portray David in that moment between conscious choice and decisive action. He knows what he needs to do, and now he just needs to go do it. And I showed that image of the company, I said, look, there's going to be all these articles this year about like, you know, Dix and a shithead, you know, the company can't do anything. Why is Facebook have a billion users and you guys only have

300 million? Just remember this. Like, we know what we have to do. We've had a plan that we've needed to execute against for the last year. We all know what those priorities are. We've made our decision about what we need to do. Now we need to go do it. We're between conscious choice and decisive action. Don't get distracted by the external noise. We're not as great as everyone thinks we are today and we're not going to be as stupid as everyone thinks we are tomorrow. So I tried to do that with stories. I would do that like every so often, you know, whether it was that or another image or some other piece of art or something that was easy for people to remember. I wonder for both of you, if you could paint a picture of your circle of trusted advisors,

how you think about structuring that, and then maybe how that might translate for us who are new to the industry who need to learn and grow and have assistance with decision-making. Maybe just to, I'll answer it slightly as opposed to trusted advisor, the idea of mentorship. When I was at this company, Veritas, I came in as an engineer, and like I just, you know, started coding, right? And that was like what I did. And I formed a relationship with the CEO at the point, you know, early in my career there. I don't know actually how it happened. Well, we went out to lunch one day and he said, hey, what do you want to do when you grow up?

And I said, well, I'm not really sure. Maybe I want to run a company someday or whatever. And, like, he took notice of me. Finding a mentor in an environment is super important. at least it was for me. They can guide you. They have seniority and expertise to say, hey, why don't you go try this? And they sort of watch out for you a little bit. So they won't let you go off the guardrails too much. And like they want you to try new things and experiencing things. So I'll often suggest that to people. I mean, to answer the question, trusted advisor,

it's more about mentorship. I have lots and lots and lots of them. And I'm always interested in more. And I never have viewed those advisors as having to be people that are, like, senior to me. So Kelly was, like, one of my trusted advisors of Twitter. I always knew she would tell me the truth, you know? People would go, like, that was great, that was great, and Kelly would go, like, you kind of, like, that was kind of, we were all over the place.

I don't know if everyone understands what you were talking about. You know, because everyone else is, like, telling me what I want to hear. How was that? It was great. And Kelly would say, like, you know, like, you need to tighten that up. So I would always look for people that would tell me the truth, you know, and it didn't matter what their seniority was or whether they had done this job before or anything like that

that I felt like I could learn from. So Kelly, this woman who worked for me at Twitter named Kim Scott, who used to run Google at AdSense publisher side at Google, who was also like, Kim's not going to, you know, BS you. She's going to be like, don't do that again. That was horrible. So I would have a sort of half-hour weekly meeting with her because I knew she would just, you know, tell it like it is. And then Bill Campbell, who is, you know, on the Apple board and worked closely with Steve and Eric Schmidt at Google. And Bill was also like, really no BS. Like, how is that? Shitty. You know, like, oh, okay, well. Got it. Don't do that again. You know, that kind of stuff. So people who would tell you the truth and be forthright with you and cut through the, you know, well, this was okay. And it was always people I felt like I would tell me the truth and I could learn from those people. Okay. Thank you very much. Thanks. Thanks, everybody.