Uber CEO Dara Khosrowshahi on Decision-Making

Hi everyone. Welcome to Grey Matter, the podcast from Greylock, where we share stories from company builders and business leaders. I'm Heather Mack, head of editorial at Greylock. Today, Greylock General partner Reid Hoffman talks with Uber CEO Dara Khosh Rashahi. Like most companies, Uber has gone through significant ups and downs throughout its lifetime, whether business is booming or if the company is grappling with declining revenue because of major global events outside anyone's control, how the leadership team responds dictates how the company will get to the next phase. Khosrowshahi calls these moments wartime and peacetime and says each requires a different mindset. During peacetime, he's typically more team oriented and open to hearing everyone's opinion. During wartime, coach Rajahi is speed oriented and focused on decision making. However, the signals that dictate whether a company is in wartime or peacetime can sometimes be hard to read. Khosrowshahi and Hoffman discuss how to assess various situations and how to act accordingly, particularly in confusing economic environments like we're living in today. This interview was recorded in late 2022 as part of Greylock's Iconversation series. You can watch the video of the discussion on our YouTube channel and you can read a transcript on the content section of our website greylock.com blog. Both are linked in the show notes and if you aren't already a subscriber to Gray Matter, you can sign up wherever you get your podcasts.

Obviously, we're in a troubled market and troubled times. When you have capital markets work like this, when you have potentially fundraising difficulties, you have disbelief in the market reflecting in multiples other kinds of things. What skills of leadership, what ways of thinking about this as leaders most come to mind for you and what are you doing and how would you talk to leaders here about how to think about how to navigate these difficult times?

Well, we're in the middle of it and so I don't pretend to know everything about it, but a couple of things that come to mind, which is first of all, as a leader, there's a human bias to process good news very quickly and to process bad news very slowly. The minute you see signal on good news, it's like, good news. What do we do here and there? Like you see signal on bad news, it's like, yeah, you know, let's see what happens tomorrow. And then tomorrow's the same. You're like, well, maybe things will change next week. You know, gotta be patient, gotta be patient. So I think one is just recognize in tough markets to some extent, like the leader has to be able to stare down a black hole and understand what's happening in process very quickly and then be decisive. So there are different styles of leadership during, you know, call it wartime and peacetime. And for me, during peacetime, I'm much more team oriented. I want to hear everyone's opinion, et cetera. I think we're definitely during wartime. During wartime, you just have to be much more decisive and action oriented. It was a little lesson for me. I remember this is not the first tough time for Uber. Right. Like, we've.

I think we're going to get back to that.

Yeah. Like, tough times, it's just kind of what we do. It's like a Thursday. And. And the much tougher time, frankly, for us than. Than where we are now was post pandemic. Our mobility business, which was our cash cow, lost 85% of its volume overnight. Within the context that we were losing two and a half billion dollars anyway. So I had some friends who were like, profitable businesses that lost a bunch of their volume. We were deeply unprofitable and lost a bunch of volume. And I knew we had to make some tough calls, and I really wanted my team to be bought in. So we would get together on ZOOM over and over, talk about what we have to do here and there. And one of the exchanges that I had with my cfo, Nelson Che, that still sticks with me is, you know, we're talking in circles, circles all over. Nelson's like, dar, just tell us what you want to do. Like, we'll do it. You don't need to ask our opinion. And all this, like, tell us, you know, it's during the tough times, your team needs leadership. You've got to assume that mantle. Once Nelson told me that, I was like, all right, let's go. And I took input, but I started leading. And this is not something that you do all the time because you want your team involved. You gotta be top down, and you've gotta solve for speed and you've gotta solve for decisiveness. Because even if you're decisive and you make 20%, you know, if you make a decision that's 20% off, it's better than being indecisive and not doing anything. So there's this push and pull, which is during easier times, I'm much more consultative. During tough times, I'm much more hardcore. And my team kind of knows that now. You know, they understand once in a while, I'm just gonna come, we're just gonna go, and we're solving for speed. The second part that I would say is, and this isn't just during tough times, is it's the job of the leader, to some extent, to go against the grain. And it's very easy for a leader during good times. You become the cheerleader, everything's fine. Like, the good times are when you've got to be the asshole. The good times are when everyone feels great about what they're doing. They lose discipline, they think they're terrific. That's when you've got to come down really hard on your team, and it feels nonsensical. Doesn't make sense. Like, they're doing great, they're amazing. That is not the time for you to be a cheerleader at all. And on the other side, it's during the tough times when you've got to start being the cheerleader and you've got to make assessment of your team, your situation, et cetera. So if you have a team member who's not a wartime person, move them out. That's a separate decision. But once you assess that your team is the right team there around you, that's when you really have to lift them up. And it's really hard because, like, at the end of that, of every day, you're like, oh, my God, this is so terrible. But then when you wake up that morning and you get together with your team, it feels weird a little bit, because you don't want to come off as, like, too optimistic and not recognizing there's this. There's this duality of, like, understanding the reality and be like, we can do this. Come on, we can do this. Let's go. And it's tricky because your team has to understand that you're, like, staring at that black hole, but then you're like, we got this. And then really lift up that team, because believe me, if you feel like crap, they're feeling like crap. They need you to lift them up.

And on the pandemic now, having gone through it and picked up some of the kind of structured principles, the way that you do as a leader is if you were given, like, a magical phone that you call yourself the day where the pandemic bomb dropped, what would you tell yourself to do differently?

Definitely go faster. And, I mean, this is something I tell myself all the time as it relates to people in terms of decision making. We move quickly, but we could have moved a bit quicker. And then it goes to the biases that I talked about. If the right thing to do is to, let's say, cut 25% of your staff, and we cut 25% of our staff. It was terrible. You'll convince yourself that the right number is 15. You'll convince yourself that, well, I'll do 10, see what happens, et cetera. And you may not have the luxury to see what happens. And so I would say move faster and don't mollify the tough calls. Ultimately, when we made the cuts, we did go in deeply. It took me too long to get there. It took way too many restless nights, way too many sleepless nights. And, and you know, while that the end product, I think we handled it well. We just took too long.

So let's switch to it when you were doing more careful. You know, I had kind of presumed that the way that you one would do a kind of a cultural rejuvenation was one would ride in and say, there's a new sheriff in town, get on board with a new program and just do that. And one of the things that I got added to my toolkit, which is now permanently emblazoned, there was no, you write in and you say, actually, in fact, you're already great. I'm just helping you rediscover it in a variety of techniques. For example, let's do a company wide survey of what our values are and then we'll get all those values and I'll do some editorial to redo it. But I'm doing it as a reflection of what you guys are saying versus you guys have to now come join my camp as ways of doing that. So say a little bit about that kind of reforming, kind of what you did for folks here. And then that unlike the like wartime pandemic, which is, oh my God, we have to make all these hard decisions right now, that was a more deliberately graceful process. Say a little bit about that.

Yeah, absolutely. So when I came in, one of the, and one of the significant reasons why I came in was because the company needed a cultural reset and the world was looking for it. I think leadership was looking for. The board was, was looking for it as well. And so we had to move very, very quickly in terms of that reset. And I will tell you, like the old culture of Uber, I don't consider it a bad culture. Just some of the cultural values started getting weaponized, you know. So I'll give you one example. There's a cultural value which is be toe stepping, right? The idea behind that was challenge others. Don't be afraid to challenge us. That's a great kind of culture to have, but it got weaponized into it's okay. To be an asshole. Right. And so it was really important to me because there was incredible talent at Uber, met a couple of them here, you know, engineering talent, et cetera. And I wanted to respect that, partially because it was like part of my job as being CEO and I want to keep that talent. At the same time, I knew that we had to make a cultural reset very quickly. So we went out there and we crowdsourced from Uber employees what they thought the best norms, cultural values should be. And we took some of the older values that we thought were great values, big bull bets. And then we also combined them with some newer values that we wanted to carry the day. You know, the one that we talk about a lot is like, we do the right thing, period. And so for me, it was a. It was a symbol of, hey, we're going to take the good of the old. And there's a lot of greatness that made Uber what it was. And I'm here to pivot, not say that everything that you did was terrible. It's actually interesting. I'm now like, we actually very recently reset those values as well. And we went through an exercise. First of all, I think a lot of companies are. People view as a. Values are forever. They never change. I personally think that's bs like companies should be constantly adapting. Right? You shouldn't be changing your mission. Mission is super slow. Maybe once every 10 years or so, values and norms slow. Strategy, medium. Tactics, pretty fast. And so we actually revisited our cultural norms and values. And this time I was more top down. I work with my team, but I felt like I had a right to be top down because I'd gotten to know the company and I wasn't like some stranger saying, I'm going to make a bunch of stuff up not knowing anything about the company. And actually this time it really did work out, because it did. It was a reflection of some extent who we were. But really the new company that we.

Are going forward and I presume do the right thing, period, is still on the.

That one was the one that stayed.

Yeah, no, exactly. What did you find? Like, one of the things we had at the.

By the way, I'm going to interrupt you for a second.

Oh, yeah, of course.

What's the job of a board during tough times?

Well, it depends a little bit on company CEO health and position, but I think that too often boards get to adding to the tough times versus helping.

That's what I'm saying. Like, it's a job of the CEO to go against sometimes. Exactly.

Yes.

And I think from a board standpoint, the CEO knows he or she's a piece of crap when things are badly. Like the board doesn't need to remind you.

Yes.

Right.

And so I think that it's very similar to what you were saying about the CEO job, which is, look, when it's obvious of shit we're having difficulty navigating this particular facet of business might be, you know, a go to market strategy, might be a competitive thing, might be a capital raise, might be a delay. And like, like one of the most irritating things I find happening in a board meeting is like, oh, look, we slipped this product by a quarter. And then everyone spends an hour going, well, slipping a product release of a product by a quarter is a really terrible thing. It's like, yeah, we know it's a terrible thing. We don't need an hour to talk about it. Right. And if you don't know it's a terrible thing, or the CEO doesn't know it's a terrible thing, that's a different problem.

Yeah.

Right.

And by the way, the next time they'll commit to a much slower time frame so that they don't have to tell you that they've slipped.

Yes. Right. And so, so I think that the key thing is, are you making the focus decision? Are you making the hard calls? Are you not getting ameliorated, you know, into kind of softer things? Like, one of the general things is it isn't just that people like to be liked, but, but there's also a question of, there's a lot of competing ideas. I have this mode where I try to teach other board members of. It's a very simple thing of green light, yellow light, red light, green light is they're the co, they make the decisions. You're just talking to them. That should be the state you're in. Red light is they might not know it yet, but they're not the co. You know, there is some transition that is coming.

They will know.

Yes, right. The most often mistake that, that board members who are inexperienced is they stay in yellow light too long. They go, there's a question. And then they start doing like micromanaging. Right. And so when, when I get to a yellow light circumstance, I pull the board members aside and I explain this theory to them and I say yellow light is a limited amount of time with a limited measurement. And it goes one of the two directions.

That's right. If you make the assessment that they are, then you got to lift them up.

Yes. And figure that out. So One of the interesting things that I saw you and Eric talking about at the Masters of Scale Summit, where Eric asserted that a new CEO in a turnaround circumstance had to trade 80% of their staff. And you said, well, actually it ended up being 50% for me. And I think that's a good number. Say a little bit about what that changeover of staff looks like, how to manage it with grace and care for the organization. What are the kinds of principles when you're doing this kind of change in direction, whether it's a rejuvenation of the company, whether it's a pandemic, like, how do you approach that?

Yeah, so I don't think that there's. It's too easy to be formulaic about things that aren't formulaic. And a company's an organism. It's like a human body. Right. So you can't make grand pronouncements. For me, it was pretty practical, which was I had a lot of respect for Uber as an operating entity, whereas from an operating entity, from technical standpoint, from a product standpoint, I thought that the company had real strengths. High bar for talent, high demand for throughput. Where Travis, who was a founder, didn't invest, was in the framework around the company. Cfo, general counsel, a bunch of the. Some would consider it BSE kind of framing controls, et cetera, governance, the board, that part of the company was quite immature. So I had a hypothesis coming in. I'd spoken to a bunch of folks, I'd spoken to board members, I'd spoken to a bunch of folks who had worked with Travis, had great things to say about him and a bunch of folks who didn't. So I will get a complete view. And so when I came in, I had a plan. So my hypothesis was this was actually a company that was pretty nails in terms of operations, product data, science. And of course, we still can get better. But the framing of the company, the corporate staff, et cetera, had to be turned over. So that's, you know, I moved very quickly there, those areas. I brought in Nelson Che, new cfo, I brought in new chief legal officers, new chief people officer, et cetera. But a lot of the leadership in OPS stayed with us and on the technical side, brought in a new team, I think up level them as well. And I think, to me, the lesson is go in with a hypothesis on where you think strengths and weaknesses are and what you're trying to personally achieve. And I say, so far, so good. Two years in, it still felt a little bit like Daro's new people and then the old guard. And we were one team. Didn't feel like one team. Pandemic brought us all together and it's like one team now. It's whether you were there before or after, etc. Everyone's sitting in their chairs comfortably. There's a lot of trust. These kinds of really difficult circumstances can either bring a team together or tear it apart for us. Fortunately was a former and the team feels really good now.

Well, with existential threat. If you survive. If you survive.

Yeah.

Last question before we open to the audience. Uber is one of those companies that has a unique lens into what's happening in the economy, what's happening in society, what's happening. What are the surprising lessons kind of now post pandemic coming out of it? What's growing, what's shrinking, what economy is working or not working frequently Disasters, a recession. Question.

Yeah, of course this is going to seem a little boring, which is people always look coming out of pandemics. They look for forever changes. It's very rare when things get changed forever. Everything is reverting to the. Everything's reverting back to the norm. There's this hypothesis in the pandemic, oh, it's going to change things forever. And you know, we're talking about Shopify. Shopify's growth has now come to a norm, which is a really good norm, but it's, it's reverting. We're seeing our rideshare volumes, you know, explode now versus really low, low levels. And so we've been desperately looking for patterns, work patterns, you know, work day commute versus weekend versus party nights, etc. Everything. It's like people are people, they go out. All the patterns are back. The one pattern that is not back is cities opening up. And the US was really trailing the world, I would say six months ago, the world was opening up much faster than the US US is caught up to the world, which is very interesting. But the US is a tale of two coasts, which is the west coast is opening up unbelievably more slowly than basically everywhere else. The bottom three out of our hundred cities, Portland, Seattle, San Francisco are still like 60% of pre pandemic highs in terms of trip volume. And like New York, Miami, Atlanta, you know, Houston, anything in Texas, all of those are, you know, back and happening. So it's pretty remarkable how the coasts have bounced back so differently in one country and we don't see that in other countries. Most countries are actually quite similar in terms of the UK comes back France comes back, Germany comes back in an even way. But the US is. It's just the tale of two coasts. Hopefully that changes.

Indeed. So I actually have a stack of questions, but part of what we try to do here is opening us up to the room. Oh, there's a hand here for sure.

You know, you came in, obviously, under tricky circumstances, to say the least. And you probably finally got through the moment where you're gelling some of the cultural norms, you're moving on, and then the pandemic hits.

Yeah.

And it's got to be a little bit like, fuck. Like, can I just catch a break here? How do you, as the CEO, how do you steel yourself for that moment where, you know, you can't wallow in.

That you got to lead the team. You can. And, you know, I'd say on this one, I'm pretty lucky, if you want to call it that, which is. My family were immigrants. We lost everything when we came to the States. It crushed my dad. He, you know, he built his whole career. And it's a little twisted, which is like, I'm super competitive. I work really hard. I look like. I just. I love all this, right. But one, losing everything and then our rebuilding it. Cause we were lucky enough to come to the States. Losing everything and always having my family around me and then losing everything and seeing it crush my dad has created this weird circumstance for me, which is like, worrying. It doesn't help. Like, getting stressed out. Like, there's nothing constructive about getting stressed out. Either you're gonna fail or not fail. You're gonna make good decision tomorrow. Bad decision. Like, why the hell stress about it? So my wife calls me a robot, so I can't give you, like, constructive advice on that. I'm just able to like, hey, this really sucks. Okay, what do we do? You know, And I'll do my best. I'm always gonna have my family. And that just lets me. I kind of don't need to steal myself. I'm lucky that way, I guess.

The thing I would add is the CEO job is a very lonely job.

Yeah.

Right. And so part of the reason why we do events like this and other kinds of things is find some people that you can talk to about it because we're stronger together.

One piece of advice that has really. Or one mechanism that really helps is I have always had a really close relationship with my chairman. So when I ran Expedia's Barry Diller, he's a real business mentor of mine at Uber, It's Ron Sugar, and I Have a one on one with him every week and it's an hour every week. I do not miss that one on one. Like we will have it every single week. Sometimes it takes 10 minutes. Usually we fill an hour. That helps me. You lose that board dynamic. He starts helping me as a person and then when it gets to the board meeting, he knows so much.

Where's the next question?

If you have to like, think about Dara before Uber and after Uber, how has Uber transformed your leadership style?

Oh, boy. I don't know if it's transformed my leadership style. Like, I guess Uber's a much more public leadership job. Like at Expedia. Expedia is really cool. 13 years, got to be a part of building a great company. And my favorite part of Expedia was like succeeding as a team. Bunch of people made a bet on me. It really worked out. We're like great friends, but publicly no one gave a shit other than my investors. It wasn't in the public ether, etc. One of the things that I love about Uber is the impact that we have. It's unbelievable. It's a really important product. So for me, getting comfortable in this seat, in the public seat has been a real learning for me. I'm kind of looking forward to being out of it, just not yet. I still have some work to do. I'm not an extrovert. Like, I go home and I lock myself up in a closet and I cry. And that helps me get ready for tomorrow.

As long as you're not worrying about it.

Yeah, exactly. So I think that's been the bigger transformation, which is, and you know, the advice that I would give folks here is if you're going to be on the stage, it's a performance, so perform. You know, it's not like it's not the same as sitting at a table and talking to folks. There's Barry Diller, who was my mentor. You know, one of the things he said is, Dara, when you're on stage, it's not your job to inform, it's your job to entertain. Just don't be boring. And so I think all of you, hopefully you're gonna be on stage. Like, you're gonna be on stage in front of your employees, et cetera. And when you're on stage, don't treat it casually. And by the way, I know, like, there's some CEOs who are like, you know, I don't like doing this part of my job, but there's a lot of parts of your job that you Know you don't like. Right. There's really cool stuff and not. But these are moments for you to communicate, take them seriously. Even if it's an all hands, if it's in front of 50 people, if it's in front of something like this, don't take these moments casually. So I think that's something that I've learned over a period of time.

Hey Dara, thanks for taking the time tonight.

I have a question.

I'm curious, are there any other kind of countercyclical or contrarian things in retrospect that would be useful to do in different market cycles?

Yeah, it's a great question. And by the way, it's a really easy answer in theory. It's much harder in practice because if everyone else is dancing and you're not dancing, you're going to feel lonely. I think one of the regrets in terms of what I could have done better pre pandemic was I could have been tougher on cost. I should have been tougher on cost. Not could have, I should have. And so when times are good, there's this temptation to chase, chase, chase. Higher, higher, higher. You see that positive signal? Absolutely. Add five engineers to this team. Oh my God, it's growing, you know, 500%. Ten more engineers, let's go. The chase hides poor work. That chase prevents you from really driving the team to build an excellent product. Because what they'll do is they'll build a good product and you'll throw a bunch of money behind it and they'll never have to get to excellent. So I wish I'd done that. But I'll tell you, it's really hard. Cause the chase, like it just, you just want to go. And then when especially it's a peer pressure of, you know, their friends, teams are doubling size, tripling in size. Why aren't I? So there's this then. The only way that I've seen it work is there's this like, I don't mean to be like too male. Like there's this like warrior mentality. There's this hardcore mentality that you introduce and you, it becomes kind of a heroic affair for you to do with 5 engineer, something that someone else is doing with 12. Because the temptation is people tend to equate, a lot of teams tend to equate team size with what you think of them. And that I found is the hardest thing to crack, which is the manager who's the hero isn't the one who takes her team from 10 to 50. It's the one who says I've got 10? We're a bunch of ninjas and we're going to stay at 10. Don't give me another person. It's hard to do. It's very easy for me to sit here and tell you all this, but it's hard to do.

I was just wondering, Uber, even before your time, went into a lot of markets globally and then also now pulled out of some of them. In hindsight, or do you think now of you were earlier in Uber, was that a great decision or should there have been sort of more of a focus on a few markets?

I think it was a brilliant decision. I think it's one of the great gifts that Travis left me. It was harder. It took a lot of guts, took a lot of money raising to do it. We are now going through what I call a painful process. And, you know, when you go into those markets, you have to hack a bunch of the local behaviors. And then usually you can replace the hacks with technology. And, you know, humans are the hackers, so to speak, you can replace it with tech. And tech is always in the end better. Almost always in the end, better. You do it for the big market. So there's a bunch of debt that the company's taken on in terms of complexity and in terms of rolling out, you know, pricing, routing algos to every single market out there. So I won't pretend that it's not. There's a bunch of debt that we're repaying, but if you're repaying debt, it's against a great asset that we built. So I think, yeah, like Southeast Asia turned out to be too competitive. I'm really glad we got out of those markets. But the, you know, we've gotten out of 10% and we're still in 90%. So it was a huge, I think, asset. It was the, the blitzcale that worked, you know, one of the big ones, right?

Indeed.

Just curious.

I mean, speaking of Southeast Asia and some of the competitors there, I'm curious if you have any opinion on why super apps haven't worked in the United States yet.

You know, I'd ask you the same question, too. I think that one of the elements that have made super apps work in Southeast Asia and China is just broadband and bandwidth. And so it's been a necessity of that market for these super apps to work. And I think the Western markets, you know, Western markets love pure plays. And if you're trying to do 10 things definitionally, you're not going to do it as well as someone who's trying to do one or two. We're actually, I think we're trying to figure out, you know, we've got our rides app and our eats app and single driver app. And so we've had lots of debates as to how these apps come together and it's really difficult to work it out. But there's single identity, single payments, et cetera, single CRM stack that we're using to essentially build one experience with multiple apps. That's our solution. But I don't have a perfect answer for you. It's really speculation and there's a lot.

Of other speculation on that too. Well, you can all see why it is. Actually, I haven't told you this yet. Why after I saw Dara on the Master Zale Summit, I texted the Master Scale team and said, we need to do a follow up interview with Dara. There's too much good here. And so with that, let's thank Dara for spending the evening with us.

That concludes this episode of Gray Matter. You can watch a video of the discussion on our YouTube channel and you can read a transcript on the content section on our website greylock.com blog. Both are linked in the show notes. And if you aren't already a subscriber to Gray Matter, you can sign up wherever you get your podcasts. I'm Heather Mack. Thanks for listening.