Invest Like the Best with Patrick O'Shaughnessy - Jason Droege - Building Uber Eats - [Invest Like the Best, EP.300]

Episode Date: October 25, 2022

My guest today is Jason Droege, a venture partner at Benchmark. Jason’s had a long entrepreneurial career, which most recently culminated in building and leading Uber Eats. He joined Uber in 2014 with a blank piece of paper to grow the business beyond ride sharing. Within six years, he found product market fit with food delivery, refined the service, and scaled Uber Eats to a global $20 billion GMV run rate. Our conversation pulls out the most important lessons learned during that period and how Jason now employs them in his role at Benchmark. Please enjoy this great conversation with Jason Droege.   For the full show notes, transcript, and links to mentioned content, check out the episode page here.   -----   This episode is brought to you by Tegus. Tegus streamlines the investment research process so you can get up to speed and find answers to critical questions on companies faster and more efficiently. The Tegus platform surfaces the hard-to-get qualitative insights, gives instant access to critical public financial data through BamSEC, and helps you set up customized expert calls. It’s all done on a single, modern SaaS platform that offers 360-degree insight into any public or private company. As a listener, you can take Tegus for a free test drive by visiting tegus.co/patrick. And until 2023 every Tegus license comes with complimentary access to BamSec by Tegus.   -----   Today's episode is brought to you by Brex. Brex is the integrated financial platform trusted by the world's most innovative entrepreneurs and fastest-growing companies. With Brex, you can move money fast for instant impact with high-limit corporate cards, payments, venture debt, and spend management software all in one place. Ready to accelerate your business? Learn more at brex.com/best.   -----   Invest Like the Best is a property of Colossus, LLC. For more episodes of Invest Like the Best, visit joincolossus.com/episodes.    Past guests include Tobi Lutke, Kevin Systrom, Mike Krieger, John Collison, Kat Cole, Marc Andreessen, Matthew Ball, Bill Gurley, Anu Hariharan, Ben Thompson, and many more.   Stay up to date on all our podcasts by signing up to Colossus Weekly, our quick dive every Sunday highlighting the top business and investing concepts from our podcasts and the best of what we read that week. Sign up here.   Follow us on Twitter: @patrick_oshag | @JoinColossus   Show Notes [00:02:52] - [First question] - What it was like at a high level building Uber Eats [00:07:38] - How he would structure entrepreneurial incentives on a platform like Uber for a new leader or team attempting to build on top of it [00:10:17] - What he learned about selecting competitive frontiers and mistakes made while building Uber Eats [00:15:17] - Things that Uber Eats got most right that he’s proud of  [00:18:16] - Constructive mistakes that taught him a lot from his time with Uber Eats [00:20:36] - What made India such a competitive environment  [00:26:13] - What improved the most in his playbook for launching in a new city [00:27:14] - Defining what best means in this competitive sector   [00:29:01] - Dealing with suppliers in different categories and finding an ideal balance [00:32:09] - When monogamy between the buyer and supplier matters and when it doesn’t in a marketplace  [00:36:12] - Defining what founder market fit is and being “fingertippy” [00:37:29] - His views on the relationships between leaders of businesses and their cultures [00:40:26] - Why Uber believed in him more than he did  [00:41:40] - What he learned about marketing to suppliers specifically  [00:45:18] - Differing views he has on the concept of failure   [00:47:31] - Thoughts about ideas versus execution and the relative importance of the two [00:49:10] - Effectively measuring opportunity cost and using it in decision making   [00:58:56] - The most interesting things he’s learned from his time as a partner at Benchmark [01:00:15] - The kindest thing anyone has ever done for him

This episode is brought to you by Teegas. Over the years of our partnership with Teegas, they have evolved from a pure expert network into a full company intelligence platform. I've been so impressed by the platform that my firm, positive sum, recently made an investment in Teegis. We did so because we feel that Teegis will be the gold standard platform for investing research for decades to come. Teague streamlines the investment research process so you can get up to speed and find answers

to critical questions on companies faster and more efficiently. The Tecis platform surfaces the hard-to-get qualitative issues. insights, gives instant access to critical public financial data through BAM SEC, and helps you set up customized expert calls. It's all done on a single modern SaaS platform that offers 360-degree insight into any public or private company. As a listener, you can take TIGIS for a free test drive by visiting TIGIS.co slash Patrick. And until 2023, every TIGS license comes with complementary access to BAM SEC by TECIS, which makes it easy to search and analyze public company filings and transcripts. Today's episode is sponsored by Brex, the integrated financial platform

trusted by the world's most innovative entrepreneurs and fastest growing companies. With Brex, you can move money fast for instant impact with high limit corporate cards, payments, venture debt, and spend management software all in one place. Ready to accelerate your business, learn more at brex.com slash best. That's BREX.com slash best. Hello and welcome, everyone. I'm Patrick O'Shaughnessy, and this is Invest like the Best. This show is an open-ended exploration of markets, ideas, stories, and strategies that will help you better invest both your time and your money. Invest Like the Best is part of the Colossus family of podcasts, and you can access all our podcasts, including edited transcripts, show notes, and other resources to keep learning at join colossus.com.

Patrick O'Shaughnessy is the CEO and founding partner of Positive Sum and the CEO of O'Shaunasy Asset Management. All opinions expressed by Patrick and podcast guests are solely their own opinions and do not reflect the opinion of positive sum or Oshonese asset management. This podcast is for informational purposes only and should not be relied upon as a basis for investment decisions. Clients of positive sum or Oshonnessy asset management may maintain positions in the securities discussed in this podcast. My guest today is Jason Drogey, a venture partner at Benchmark. Jason's had a long entrepreneurial career, which most recently culminated. in building and leading Uber Eats. He joined Uber in 2014 with a blank piece of paper to grow the business beyond ride sharing. Within six years, he found product market fit with food delivery,

refined the service, and scaled Uber Eats to a global $20 billion GMV run rate. Our conversation pulls out the most important lessons learned during that period and how Jason now employs them in his role at Benchmark. Please enjoy this great conversation with Jason Drogey. Jason, you've been such a builder your whole career and obviously some things people be really familiar with probably Uber Eats being a recent one that you spent a lot of time building and leading. And I thought we could start there because it's such an interesting episode of a new product inside of a big existing company that takes advantage of some of the pre-existing rails of that company and has worked to some extent. We can talk about new breets versus DoorDash.

I use Uber Eats every day. It's a thing people know and use, and it was born out of something else. I think most big companies hope for second acts or subsequent products or expansions like Uber Eats. I never really talked to someone that's been the leader of one of those successful things before. I've tried to do this with AWS and a few others. They're kind of rare. So maybe you could just describe what that was like at a high level to be the leader of a thing which was clearly distinct, but which was born out of a preexisting platform because it just seems like a really interesting type of innovation that everyone says they want. but very few people get.

This was my second one. So I had some lessons from the first one, which we can talk about too, which was Axon. Maybe it would be useful to start with how I ended up in a position in the first place, and that could be constructive as to like how you structure these things if you're looking to do so. I had known Travis for a long time.

I've been an entrepreneur most of my life. Travis and the management team at the time believed that there was going to be more happening at the local logistics level than just rights. And this was in maybe late 2013 when I really started the conversation with them. And the idea was, okay, we don't know what that is. Ride sharing is going through the roof right now. We know we're going to need some lead time to figure that out in Uber is known to be an ambitious company. And this is an example of that. And so my job was essentially to come in and figure out what other businesses there could be. And so a key part of that was the CEO had buy-in, almost blindly is the wrong term because it's a little bit negative. But it was like there is something here. There was conviction at the top. There was other businesses to be started. here and there was conviction that this is a person that I believe in to find it, and I'm willing

to die on that hill. And I think that's a common theme. And that's counter to the incentives of a lot of companies that are out there. I used to say that you have to risk career-ending failure, to find career-defining success. There are just some people who are willing to do that, and they want the ride, they want the excitement, they want the adventure, they want the challenge. and the idea that come into a company where it was like the company. Every pitch deck in 2012, 2013 was, we're the Uber of X. The idea that the difficulty level was, how do you find another business as big as this, or to grow at this level was really exciting to me.

And it was really exciting to the initial people that joined it. So I think it starts with you got to have buying it from the tops of the person who's making all the mistakes early on, which was me because I made a bunch. It feels a sense of safety. Uber did a really good job at that at the time. I started day one. It was just me on the team. I was a one-man team for the first six or seven months.

There's some analogs even with benchmark here where it's like you can get a lot done when it's your perspective. You decide how every minute of your day there's not an analyst team. There's not research reports. There's none of this. We rejected this whole idea that any of that would be helpful. It was literally me walking around the city for a few months, developing some ideas. And then fielding ideas from our operations team too, which is a really, a big part of Uber success. We had this entrepreneurial culture that was pushing ideas to

the headquarters, you want to call it that, so we could kind of filter through all the ideas. And one of the ideas we tested ended up being from one of our teams, which is Uber Rush. Essentially, we just tried a few things. I wanted to get into market as quickly as possible because this was new for me. I was selling enterprise software to governments before and void before that and a bunch of other things. Getting into market quickly to just sort of get my bearings, the license that I had to do that was really. really powerful to speed up the whole process. If we had launched food delivery as it exists today, the minute that I joined, and while each is enormously successful, 50 billion dollars in your GMV,

etc., I think you would have seen another level of maybe market share gains because we would have been able to get there even faster. But there was an experimentation phase that I had. A lot of incentives are, how do I get my annual bonus? How do I get my quarterly review? How do I get all those internal things really muck it up. How many companies have the stock or the value or the opportunity to give someone who is used to doing their own thing this license to do that? I think it's possible when I joined what is now Axon, then Taser. They were an old company that managed to do this. So there are a lot of lessons learned there. Can you talk about the incentives, meaning the personal incentives for the people on, let's say in this case, you and the Uber Eats team, it seems like

all big companies want entrepreneurial-like outcomes but are unwilling to give entrepreneurial-like incentives. Founders tend to do so well because they have such insane upside exposure by equity in the business. And maybe sometimes you could incentivize someone like you through lots of equity in Uber that's still somewhat disconnected from the thing you're actually working on. If you were doing this, if you were running an Uber-like platform and wanted to incentivize entrepreneurial outcomes like this, like what you do with Eats, how would you structure the incentive for the leader of that group and that team? I don't know that you can put it all on paper.

I'll start there. You have to pick the right person and the right personal orientation and then provide them enough incentive so that it's a risk hedge that might fit with some part of their life. So for me, I was starting a family and I had a bunch of personal reasons I wanted to get back to the Bay Area, but I wanted to maintain that edge. So I had this personal situation where I didn't want to start something from scratch, but I wanted to have that entrepreneurial edge. And then the Uber stock was a rocket ship.

And so from an incentive standpoint, hey, that's awesome too. And there's other incentives along the way that I'll just trust will work out. I don't think that people need to be incented with the exact amount of money, even like a risk-adjusted amount of money to get this same incentive. I think it starts with the personality. And then you have to say, well, what would be like an outsized economic situation for this person that they can aspire to. But trying to tie it, what I've seen is a lot of mistakes here, where it's like,

well, if we get to this revenue level, then you get this. If we get to this geographic expansion level, you get this. And the problem with those things is that it doesn't give the leader of the business enough flexibility to do the right things for the business. There almost needs to be an unwritten contract between the CEO, frankly. I don't think it can be anybody else than the CEO and the person doing the job. I'm going to take care of you if this works. I can't say that you're going to make a billion dollars because you're not.

Or maybe you are in the right situation, but you're going to make more money than you need. If there's that trust that exists, I mean, the trust should precede all of these conversations anyway, then I think you can set it up correctly. There's no way to, like, have a blueprint. There's no structure. Cut and paste this anywhere you go. I think that the person doing the high-risk work just needs to know that the CEO is going to defend them at the board level when they burn $30 million, which I did, and didn't have much

to show for it before we find the success. There's something so interesting about the job to be done for what Uber Eats does, get food from somewhere and bring it to me. It's so straightforward. It just doesn't seem controversial, and people want to do it, obviously, as evidence by the size of the GMV in the market today, USN global. But as you were building the thing, I'd love to hear what you learned about and maybe what mistakes you made, selecting, I'll use your partner, Eric Vistria's term, competitive frontiers,

picking which dimensions of the product are going to matter under the umbrella of, give me food and bring it to my house. There's lots of little things under the hood there that dictate winners and losers. What did you learn about finding what the competitive frontiers were versus competitive services as you built a breeds? And what mistakes did you make along the way? I think there's a really interesting discussion here that's not direct to your question, but around cultural biases and how they can see into decision-making without you realizing it. And so I'll touch on that real quickly before I go direct into it. Uber was a company that was all about speed.

Everything was speed. If we can deliver you a car in five minutes, we can deliver you in anything. We can deliver you anything in five minutes. This was a mantra and this idea. And so speed was important because we saw that for ride-shed. The fast you got a car, the more likely you were to order the next car. And reliability matters and all of that as well. We started with speed being the thing that really mattered.

You know, the first version of Uber Eats was we put food in a car. You hail the car and you eat whatever was in the car. That's interesting. I didn't know that. Yeah, it was called Uber Fresh. The very first day we lost Uber Fresh, I think it was tender greens and the Bay City's deli down in Los Angeles. We said, here's three meals you can order for lunch.

She'll be there in five minutes. And these were restaurants that people like. It was hard to get food. There were long lines there. So we're solving a convenience problem. But you had to have food that would travel well because the food is made like an hour and a half before. And so we really focused on speed.

And it was crazy. You'd hit the button, walked out, and there was a sandwich that you'd otherwise normally would have had to wait an hour and a half to get because it's like a such a pretty popular sandwich. That was a cultural bias that it had to be fast. And we launched that product and scaled it to a few cities before we topped out. And it wasn't failure along the way for lunchtime. The churn curves look good. The union economics were okay.

We were working on them. Drivers were doing five, six deliveries per hour, which is insane efficiency. And then we just topped out. And we kept asking our customers. What would make you use this more? And they'd say, hey, if you had more restaurants, better restaurants. And we're like, no, no, no, no. But how much faster does it need to be?

And you're like, no, no, no, no. I just want you to put these five restaurants on. After a couple months of this and me going way too deep on the analysis of the business, I was just like, oh, this is so dumb. I don't just listen to our customers. And so we did. And then along the selection route, as soon as we launched a product that had 100 restaurants or 50 restaurants whenever you open the app,

business just exploded. Within the first few hours in late 2015, we launched Eats in Toronto. In the first three hours, we did as much sales as we did in the previous four days with the other products. So it was very obvious we had gotten the product wrong with the market, right? Everything starts with selection. It's sort of obvious to anyone who's been in retail. You need the best restaurants. And you saw, like in early days, Caviar went to the very high end, and they had the very best restaurants. And those restaurants then would drive demand for the network, and then they would roll in other restaurants that would fill in the gap, increased conversion across the app, because you had more restaurants, and they stopped at the

high end and eventually sold the Doordash. DoorDash and Postmates would list restaurants they didn't have contracts with. They would go into a market and almost strip mine in the market. What are the best restaurants? Let's put every single one on, and whatever ones get hits, then they would then call up those restaurants and say, hey, got all those demand. How'd you like to keep that demand? Here's a contract. It all starts with that, and then everything flows from that. Nothing matters if you don't have the best restaurants. All the metrics break. I think that was lesson number one. I think there's a few lessons within that, which is why I mentioned the cultural biases comment, which is we are not building rides plus food. We are building an e-commerce marketplace.

That actually means we need to really challenge our beliefs. What are the beliefs that? that we go into every decision with, and we lost so much time and a little bit of money, not really on the grand scheme of things, but so much time, more importantly, because of our belief system. We started asking ourselves around big decisions. Why are we doing this? What do we believe in the first place? And that can feel very ivory tower. Let's as a management team talk about our beliefs. What you believe determines what you think. What you think determines what you prioritize and what you prioritize determines what you do, and what you do to determine whether or not you're successful. So if you aren't evaluating what biases do we have as an organization,

what biases do we have as a business, what are our personal viewpoints, what are we so convicted on that it's just plain wrong. We were totally wrong that speed was more important than selection we thought it was. I'm curious in both sides of the spectrum here. What do you think were the things that Uber Eats got the most right as a product organization and the things that it got the most wrong, setting aside speed. I understand that the emphasis on speed is probably the biggest. But maybe starting with the good, what are the things maybe all the way up through how the app operates

today that you think were done the most well? You would advocate that other e-commerce marketplace builders adopt or whatever. Like, what are the things you're most proud of? We had a product organization, and this might sound a little odd, but at Uber, everything had to be global. There was global ambitions for everything. We had global teams, operations teams. And our operations teams were kind of part of the product team,

if you think of it that way, because you need the restaurants, you need the local apps. Like, that is part of the product experience for the apps in the marketplace and how that worked. We had an insane focus on reliability, which came from our focus on speed, which we were by far the most reliable marketplace when we rolled out for the first time of any of them that we were aware of. If you think about why is that important? Well, when you're moving at a speed where in 2016, we set a goal for ourselves to launch 100 market. which is insane to launch that many cities, but we said we're up to the challenge. What our product team had done previously is they had thought through, how do we thread

the needle on all the needs for every country that we plan on launching into? What is the most important thing that we can build that will have the most value across 25 countries? And then they four strength with that go-to-market in mind, and they were a very business and marketplace-minded product team, which was exceptionally useful. and product teams get a lot of flat because they can't move as fast as the business. It's just the nature of product versus business changes. The number of failures that we avoided that were very likely failures that would cause massive delays, they did an exceptional job thinking through

first. Everything from having new taxes in Europe or even like Spain versus the UK to do you have to take cash payments in Latin America? All of these things, at a big company, you have to comply with certain accounting standards because we weren't a startup that could like fly by the seat of our necessarily at this stage. How do you just get out of the business's way? That team did a really good job at that. The other thing that we did really well was we had a very strong data science team. The amount of data science drivenness of the UI allowed us to see the business in a highly,

highly data-driven way globally and almost instantly. I didn't have to interpret a lot. You didn't have to wonder. Yeah, you just knew. Yeah, I was like, what's going on the Chile? You would hear from the operations team, great perspective. You need that, often correct on many fronts. Okay, let's get into the nitty-gritty.

What's actually going on with conversion? Why are we putting these restaurants first? How does the algorithm work? Why? And prove to me that that's the right decision. There was a lot of focus on conversion just generally as an organization. It's really interesting how in some ways the speed thing was maybe a hindrance to start. What you're describing is the deployment of pre-existing scale that existed and muscles that existed at Uber,

A startup couldn't have done those things nearly as well, had the data science function, had the speed of business operations and things like that. It's the cool thing to think about back to the original question of what are the existing scale benefits in the big business? How do we apply those to the new thing without letting our other biases seep in? It's a really cool set of lessons. What about on the negative sides? Setting aside again speed as the original priority and selection as the right variable. Anything else that you think is instructive that you really got wrong, picking the wrong competitive front? frontier just going down the wrong direction?

We knew Uber was a get out over your skis type of a business, and that's the kind of operator that I tend to be, is just let's go for it and just optimistically believe that we're going to figure it out. And I think where that created complexity down the road, the product that was going to work in Europe, for example, was not the product that was going to work everywhere. I knew that we needed to fast follow with a lot of internal support across things. that I'm not sure. We certainly weren't able to execute on when I was there.

Maybe they've made some progress now. So take India. I'd do a whole thing on the India competitive environment. They're the fiercest competitors that we faced for sure. The product that needed to exist for that business to exist is not the U.S. product. If you look at Zamado and they do all kinds of things that have nothing to do with food. Zamado's got this in-restron dining loyalty program that allows, them to subsidize things on the delivery side. You had this super app strategy, which if you think

about the support you need internally across our division, organizationally, it was about 80% of the functions were my team. I got to make the decisions, budgeting everything, and the rest for matrix in. And I think organizational design is also really important to your initial question about how to you set these things out. You start touching things like payments and you start touching things like, well, should Ridesby in the Eats app, now you really start to touch a business that at the time was at a different stage. That is a very complicated thing to bet on. And I think in some cases, we were successful and in other cases, we weren't. Say a little bit more about what made India such an interesting competitive environment.

What did you learn there? Sounds interesting. I have this whole thing about getting a market first when things are working. And boy, with marketplaces and network-effects businesses, first really matters. And there's a lot of discussion about does first matter, you know, first mover advantage, is a really an advantage. And everybody has counter examples, my space, Facebook, blah, blah, blah. If you are at the business that will be the business that you cut and copy at scale, you have to be there first. It really, really, really helps.

India, when we were launching and we were launching all these markets, I was like, oh my God, India, the basket size is $3. It's a war zone for our ride-sharing business. The competitors there are fierce. Swigian is a model are just very, very fierce competitors. These stats are just absolute beast when it comes to competing. And not that our team wasn't, we were. But I delayed about a year to launch it because I'm like, I can't launch 45 countries in 24 months.

It's just too much. We launched about 10 months later than I wanted to. And by that point, if you look at where the funding situation was, when the team wanted to launch it, I think Swiggy had raised about $14 million. By the time that we decided to launch it, I think they were just closing on another round. I don't know what the number, called 50 to 100. But the point is, is they were pulling in Maspers, who is like infinite pockets. And so now, okay, we're not just competing with Swiggy, we're competing with Maspers bank account.

And then Zimato, I forget to do their big backer was, but they pulled them a big backer. Even if you out-execute, you can't out-spent. And so it became this massive discounting game, which is just a brutal place to get into for any business. And we were coming from behind. Fortunately, we exited the business as the motto. I mean, like the number of things that were going on every single day, discounts were happening on the smallest cohort basis across these competitors scraping for one basis point of market share in Heiderbod. Discounts for 30 minutes here are just crazy levels of optimization.

That India team was just lights out from a competitor. standpoint, we just launched a little late and the network effects or whatever you want to call it, we got behind it. The idea of discounting brings to mind the question of unit economics and what it was like to build a business where I think the end state unit economics were uncertain at the beginning. You just have to go in not knowing where things are going to end up. And so much of the success of the business, both eats and then the parent business, is dependent on somewhat end-state-ish equilibrium in union economics on a pari basis.

a per delivery basis, a per hour basis, whatever. What was that like? Just talk me through the experience of that being a variable that mattered, but was obviously morphing and shifting as you built. I think price and philosophy for startups is just the most fascinating topic. It's so complicated and everybody has strong opinions. Charge the most, charge the lease. You want to see if there's market demand.

How we did it here was when we launched in Toronto, I said charge an amount that is in excess of what we think we need to make. make the business work and see if the market will accept it. We charge 30% of restaurants. We charge the customer $5.00. And drivers, we knew with hourly efficiencies, we could get that cost under control over time because we'd seen that game before with rides. And it works. I mean, 30% is a big take. Restaurants were so, no pun intended, hungry for demand. And they were interested in trying it and they wanted to serve more customers. They're like, okay, fine, let me see if this works, because there is incrementality in Zendarm.

delivery, and there's a lot of discussion that's bait about how good is delivery for restaurants ultimately a lot of them adopted it. Most people don't know this, but the business was by mid-2017, it was barely losing money. And that was 19 months into launch. And we were at a few billion dollars of $2 billion of run rate GMB, maybe a little bit more at that point. I didn't have a concern about the economy. This will figure itself out the market is there, the willingness to pay is there. We had taken a few markets really early on and we had gated them. In Toronto, for example, we launched with free delivery, make sure restaurants get a ton of demand in the first couple weeks. We made a good impression. But then within three weeks,

we said throw the $5 delivery fee out. Okay, what happened to retention? Let's try to get driver costs to basically market rate. What happened to retention? It looked good. It looked good enough where I had confidence. I had enough buffer. Incrementality plays a big role on this for restaurants. And my assumption all along was that there's some restaurants where the incrementality would be really high, and they would be able to afford pretty high take rate. 30% was too high, but I think it's settled around 25 or 24 or something like that. And then there's restaurants where the incrementality would be really low, meaning it cannibalizes in-store sales. They would have a much bigger problem with the fee. I didn't know which restaurant

was which, and nobody on my team did, and neither did the restaurant. So the market kind of just needed to play out, but I figured that they would differentiate between their in-store prices and their delivery prices over time, even though we discouraged that. We didn't want to hurt the marketplace or for them to jump to that too quickly. They need to run a profitable business. We did understand that. So I just figured that the United Economics would sort themselves out as long as the customer demands stay as strong as it was. And it did. And the pricing strategy changed a lot. We didn't start out with service fees or tipping, and then we were pushed there over time because consumers in the U.S. are not elastic to service fees are tipping.

There was a lot to learn on the pricing side, but we had conviction pretty early on that this was a good business. What did you learn about, given that this was very region-dependent? You'd be launching in a specific city like a Toronto or something about doing that really well. From Toronto to the very last one you did, what improved the most in your ability to launch and Playbook for launching a city for a product? Probably picking the best restaurant to launch with.

putting in the effort, because it's a very high sales effort to get restaurants who don't need demand onto your platform. And we didn't have the functionality for a very long time to list that restaurant without their permission, which was sort of a scrappy competitive advantage of the upstarts who didn't have the same policy and perception issues that like we did. Whenever we would launch, we would launch with better restaurants. That would then create a more virtuous, their flywheel. If you got the best restaurants first, you got the most interested customers first. You also got the early adopters first who tend to be less price sensitive, unit economics look a lot better, and then everything just starts spinning. And what does

best mean? And this is a question about e-commerce marketplaces in general. Best could mean a lot of different things. Does it mean food that travels the best? Doesn't mean brand names that are most well-known. What did you learn about what actually best meant? I think it's the most supply constrained. I think if I were to summarize it, I would say that. Because, you know, you know, You would put the best known brands on. Sometimes it'd be good, sometimes it'd be bad. If you were in Paris, for example, if you were a consumer in Paris, you knew the restaurant that everybody went to for the best whatever lunch that they were serving.

We never found a consistent way to see that in the data on the internet. There was no, an algorithm. You just actually had to be like, no, no, that place is hot. But having people on the ground really, really mattered. A lot of restaurants where it was like, this is a big name. if they're a restaurant that is at scale, that's obviously correlated with a big name. But if they're at scale, then they're not supply constraint. It's not hard to get it because they have five or six different locations.

You can walk to it, whatever. So I think supply constraint plus popularity, I think, is a really important point. That's one piece. And then the other piece, which is a little bit random, and I wouldn't say that we ever figured out exactly how to identify these is after hours places was sort of an insight that we had. Every restaurant just by nature of the market closed at 10 p.m. Because that's when foot traffic closed or 9 p.m. for dinner, there is a market for food at 11 p.m. And so what happened was is that after hours, we'd see these hot dog and pizza places that people had never even really heard of just go vertical in demand.

And so we started to realize that, well, it was mostly around the single location or two location really hot places that you could only really know if being in a city. There was also this day thing that was underserved. So under supply or underserved, I think, is the framing output. To ask these questions, I'm cycling through your partner, Bill Gurley's little models for what makes for a great marketplace. There's a couple that come to mind that lead to questions about what you've learned and also kind of your opinion on good models for this marketplace idea. The first of them is fragmentation of the supplier base, where my mind goes immediately as McDonald's or Chipotle or having a customer like that on the supply side on the platform that brings a lot of its own demand, but has way more supply. power, if you will, than a corner restaurant that doesn't have any of the advantages of a Chopol-A? What did you learn about dealing with suppliers in those two different categories?

And in an ideal marketplace, what do you think is the right balance of those two categories? There's no easy answer. It really is a wander. We had all these theories. Agility, I think, was the most important because we were constantly learning. You start with a big incumbent as your primary marketplace participant. That is a bad idea. It is very, very, very hard. I think Instacard is an example of doing that successfully. Just amazing that they pulled it off and credit to them for doing it, but there's not a lot of those examples. And that's why the emphasis on what are the single or two location maybe can't reach their customers, but have great food is so important because they have a high willingness to pay because

they have a high income mentality because they have so few locations. And they're fragmented. Restaurants are naturally fragmented. Even McDonald's, which is enormous and has 1% of the World population visit one every single day, which still blows my mind. It still only has a single-digit percentage of the restaurant market in the U.S. It's certainly in the world. It's probably even smaller. I'm not actually sure what the exact number is. Most of the market is actually fragmented.

The value does accrue to the marketplace when you're aggravating fragmentation. I think the lesson is it's back and forth. You have to prove the concept of delivery with the smaller restaurants. Once you prove it, then you basically go create FOMO with the larger, restaurants. The McDonald's, for their credit, actually saw this coming and they came to us. We didn't go to them. And then once you have the larger restaurants on, yes, the margin structure of those restaurants tends to be, I wouldn't say less. It is less on paper. But because they bring so much demand to the platform, you have to look at the spread on your marketplace

fee between the larger providers and the smaller providers is customer acquisition cost. That's really what it is. If you have a rackery and you have a provider that you're giving a discount to, the only reason why you would give them a discount is if the marketplace is getting value from that supplier. And that value tends to come from new customer acquisition or maybe greater liquidity for your driver base or somehow greater conversion. There has to be some tangible, measurable benefit. And we saw that with the larger chains. And then once the larger chains come in and there's a lot more demand, then you go back to the independence who might be later adopters and say, hey, we can now generate you a lot more demand. I know you were a skeptic before.

So on and so forth as you go through the bell curve of adoption. McDonald's was unusually. so and maybe rare for being the largest restaurant chain in the world was a very early adopter of delivery. They were pushing on this from early 2016. We launched them in mid-2017, which is less than two years from the start of the business. But that's, I would say, more of an anomaly. What about the principle of monogamy that Bill talks about where one of the risks of marketplaces, like you wouldn't want a marketplace for babysitters because you find a good one and then you just leak off the platform, you just work with that babysitter over and over again? Somehow that seems different here because I might order McDonald's a lot and have a semi-monogamous relationship

with McDonald's, but I still do it through the Uber Eats app. This is actually a real thing. My kids want McDonald's all the time. We get it on Uber Eats. What did you learn about when monogamy between the buyer and the supplier matters and when it doesn't in a marketplace? It doesn't take much. So just take Bill's babysitting example. You only need one babysitter and you use that person forever. Maybe you have a backup one, but then that's it. In the case of food, you have your five restaurants that you were from. I don't know if that's the exact number still, but it's something like that.

But everyone has a different five. One, I'm not going to install five apps. I don't even remember the name of the late-night place that I go to because I just know that it's a place that I get like a Chicago dog or something. You know, at 11, I'm at my friends. You need just a little of stepping out. Yeah, I think so. Even Instacard, and I don't know the data,

but our own personal behavior is, We shop from two, sometimes three grocery stores, which maybe that's the key. It's just enough. If you reverse your hat from operator trying to build a single thing to invest or interested in other e-commerce marketplaces, are there other attributes of a marketplace that you would pay special attention to that we haven't talked about given what you learned building one? I think the orientation of the founder is really important. And I'm not sure that I entered Uber with this, maybe just by nature of my personality, this was easier to learn. But they have more financial and liquidity stock market type dynamics where you really want someone who understands

that the job of a marketplace is not just connecting supply and demand, but understanding where the margin structure to be grabbed is to potentially either take his profit or give to the other side of the marketplace to stimulate growth. There's this relative leverage idea, and maybe not the best articulated as that, but it's how much can I charge a restaurant before they are making the profit that they need? I need to find that number and know that number, and I need to do that on the consumer side, and I need to do that on the driver's side. And that doesn't mean that your job is to be maximally extractive.

That is not what I'm saying. What I'm saying is to have an understanding of, hey, when I charge customers a little bit more, I can charge restaurants a little bit less. And when I charge restaurants a little bit less, more restaurants join. Understanding what is the right balance of those things, I think I would look for in a person like that mindset. If you take the idea that you can put supply and demand all in one place, okay, that's fine. That can work for a while. But eventually you've got to get down to the nitty gritty.

And I think marketplaces at their course are the most successful ones do get in the nitty gritty. I won't force you to give him a compliment, but I can do it. Like, Tony Shoe seems like this personality where he is available at the detailed level in seconds, not in hours. He understands the detail and other marketplace founders, yourself included, have that finger feel for the edge of what's going on in the business. And that seems like a really good lesson. That's different than amazing storyteller, vision builder type founder.

This is a business. And clearly, Dorosh is not a good job. You can't say by the numbers that they haven't and that they did a good job going where we weren't. If I'm going around the world, at one point, we were competing with 10 or 12 different companies and 10 or 12 different founders' CEOs, which is its own interesting exercise from a career standpoint. I can say that among that basket of CEOs, some of them are fingertipy, and those were the ones that were the hardest to compete with, and some of them weren't, and those are the ones that we typically beat.

How confident are you in extending that fingertip idea beyond marketplaces? Do you think it starts to break down in other business models, or do you think that's just always the case that leaders that have the finger feel for the edges of the business and the details of the business are just better leaders? I think what you're talking about is what does founder market fit open to be challenged on this? How could you not be fingertip with the most important thing in the business and build something massively successful. If you're building a consumer social product, do you have to be

fingertipy with the economics to be successful? Probably not, but you have to be fingertipy with how consumers are using it, why they're using it, how are they spreading it? That part of it. And for marketplaces, your CFO is a very important stakeholder if you're building a marketplace, in my opinion. If you're building a SaaS business, which I was fortunate enough to do before all this. You have to be fingerprintedically almost with the speed of execution. Who's on the team and the sales organization? How do you make that engine go? Do you understand how your go-to-market organization is being optimized? I would say that that's probably the most important thing from my experience, if that's how to business. But that doesn't necessarily mean that I understand my margin structure

down to penny if I run a SaaS company. I've got some leadership and culture questions. You touched on a lot of these issues. Actually, something Tony said. when I did this format with him, was so interesting to me, which was around what should a culture look like? And his answer was, a culture should be 80% just the personality of the founder expanded,

and then 20% clean up around the edges, which probably runs counter to very generic cultures or values you see on a wall or something. What are your views on the relationship between leaders of businesses and their cultures, good, bad, and ugly? I believe that you can see the personality of

the CEO in the salespeople, in the frontline engineers. The way I think about it is, if you care about your people and they know that they're being cared about, and as a CEO or a business leader, I'm talking about your management team, it is very hard for them to not reflect that and then care about their people. And whatever that means to them. In some cases, it means care about their career. In some cases, it means care about the personal life. In some case, it means care about the balance or financial.

but knowing your people, that for sure will cascade down. And you can see that I used to take phone calls. I used to sign up as a restaurant and take phone calls from our sales team just to see how we were pitching it. And I would hear things from inside the organization being translated through to what they would say, how they would say it, how they would negotiate, how hard would they push or they wouldn't push? And it would make me ask the question, is that an effect of the organization?

Is that something with this employee? I don't think that there's anything that replaces caring about your people. And that's how I look about culture. I never really look at like, I want my culture to be X. I think that you can look at this in reverse, which is under what circumstance does, I'll call it a customer-facing employee, frontline employee, the person writing to code, or whoever's doing the actual work, do a better job if they don't feel protected, that they don't have air cover,

they don't feel known or cared about. And it can sound a little bit fluffy, but that's what, I got from the organization when I started Eats, which was, we believe in you. There's this thing at Uber in general, and we tried to really extend it to Eats. Uber believed in you more than you believed in you. And that was a very powerful and maybe under-discussed part of the organization. We had all these very high-performing people, and our job was to unlock their potential. As far as how cultures spray across the organization, now that I'm answering the question,

I'm thinking about it a little bit more, is when you're running a global business, You have to let it bray a bit more than you think. The world is not a monoculture, and your French teams and your Mexican teams and your Australian teams are going to have different cultural norms in terms of how they do business, how they think about the business, how they think about building culture. Maybe that's why I hesitate on the idea that I as founder am the culture. I don't think you can build a global decentralized business that way. I think you have to let a lot of it be in market. This idea you mentioned is really interesting that Uber believed. in you more than you believed in you. It sounds really nice. But just explain why is that so powerful?

Because a lot of people just self-limit. So we've grown up, we're taught that there's a way to win in school. We've taught that there's a way to win in your career. There's so many smart, capable, high-energy people out there who've never really been trained on taking risk. Then they will hedge their own risks when they are overestimating risk. In business, when you have opportunity cost is your biggest cost, which is also a really important topic, I think, is you don't want people to be constraining themselves when they're going to market and doing their job, taking risks, which have a small financial cost, but have a big ultimate equity cost. If Eats had started a year later, it probably wouldn't be $50 billion in GMV a year.

It would probably be 20. If we had worked and expanded slower and as expensive as sometimes markets were, we needed people to feel that they could make their own decisions. You can push decisions down to the edge and there's not a penalty. We believe in your decision making. And we're willing to accept the portfolio of pluses and minuses for your decision making, knowing that we cannot centralize that any better. When you think back to the relationship that you had with both the supplier and the demand side, how different was the marketing function to those two different places? Or asked differently, what did you learn about marketing to supply?

specifically to start, and then I'll ask the same question about marketing to users. There's a high conflation of marketing and sales on the supplier side for sure. Uber was such a rocket ship on the ride side. But at the time, there wasn't a well-developed marketing function. It's hard for us to open up new markets just because things exploded out the gate. When we were talking to restaurants, from a sales standpoint, getting them used to the idea, the pitch was actually pretty straightforward, which was 10% of restaurants. restaurants do delivery. They see customers you wouldn't otherwise see. If you think about your

business, you are very subject to the people who can drive to you and people who can walk to you. I had this chart that I used to present internally where it was all the restaurants I ate at when I would walk from Uber's offices to my house in San Francisco before and after eats. And my demand pattern totally changed because of delivery. And so we sort of talked about access to new customers, but access to new meal times. If I eat somewhere in the CBD, I can get it at home now. So it wasn't that complicated. More demand is a really simple message.

On the consumer side, this was a behavior that they were already doing before, which I might be alone in this viewpoint. One of the reasons why this business exploded and one of the reasons why we had so much conviction is if you look at takeaway, takeaway as an activity, not as a company, That was a global thing. And so people were already going and getting food themselves, not spending time at their family, having to do a trip with their kids in the car, whatever inconvenience. Before us, we merely came along and said, hey, for a small feed, how about we do that for you?

The marketing, once you have the right restaurants that people were having that behavior with, just went off by itself. It's an incredibly interesting thing to think about preexisting behaviors that technology makes better versus a completely new behavior like Airbnb. We weren't staying in random people's houses before we were staying in hotels, but we were going places and staying somewhere and checking in and going through those same motions. Do you think that that's an interesting way to find new business ideas? Just look for stuff that people do en masse already, that technology hasn't yet affected.

I think a lot of the reason why we feel like, oh, duh, of course that's a business in consumer land, is because it's something that we're already doing just with dramatically more value or dramatically less friction. I think that's the simple thing. It's very hard for customer to change their behavior, and there's a lot of talk about this ad nauseum. I think it's just been proved over and over again. Where the rub is, is on, okay, is this a new behavior? How do I retrofit the story that this is a existing behavior done better? Because you can talk your way into any idea being something consumers are already doing. There's a lot of selection bias. Were people sharing videos before YouTube? Not really. Were people expressing themselves on the internet? Some of these things

you do have to check yourself, honk, I'm not being too focused on that. I think there's some obvious ones. People were definitely sharing photos before Facebook and MySpace. People were definitely expressing themselves on the internet via GeoCities and other forums. People were definitely making international phone calls before Skype, which was their wedge, right? Likely it started in Europe and wouldn't have worked if they started in the U.S. So I think that there are these things. You just have to check yourself on being overly simplistic. When you and I first connected, we talked a bit about somewhat different view you have

on just the concept of failure. I'd love you to share that IDX is a little bit different. Silicon Valley has this thing where if you're not failing, you're not trying hard enough. And I don't know if it was my upbringing or love, but this has never resonated with me. And I think failure can be okay, but there's a sense of celebration. If you follow the past, the failure, there's only a few paths. There was an opportunity, and you didn't execute on it. Okay, that's not good.

There wasn't an opportunity, but you thought there was. Okay. The opportunity is there, but maybe the science isn't there yet and it'll be available in the future. I feel like that's the most permissible. Yeah, I get that. But as we've moved away from science dependent, things or artist engineering problem or unsolved engineering problems into more businessy, marketplaces and consumer apps, the tech's there for 95% of these ideas. I think it becomes

an excuse within an organization. And it becomes an infectious meme, especially with too much capital. Because what happens is, is you go out and raise some crazy amount of money in your A or B. And then you say, okay, let's take 20% of that. And we got to be betting. on the future. And then none of it works out. And it doesn't work out because, hey, if we're not failing, then, you know, we're not trying hard enough. So if the implication is, is that we're trying hard because we're failing, but maybe not. I just think it's an easy excuse. And if you start with the, there isn't a good reason for failure, but failure does happen and it shouldn't be overly penalized. And it should be highly examined. That would be more my approach that

differs from others. There's a related idea or thought, which is around ideas versus execution. And I think where I'm getting at here is that people may be undervalue ideas. A good idea everyone talks about embedded in the failure as good is try a lot of ideas. The idea itself doesn't necessarily need to be great. You just need to try enough of them and something will work and just keep going and iterate or whatever. I think ideas maybe are more important than we give them credit for. And I'm curious how you think about ideas versus execution and the relative importance of the two. I'm a big believer that ideas are very important and execution is a whole lot of it. I've heard this said a lot.

Ideas aren't sacred. So tell everybody your idea and get all this feedback and that'll inform your execution, which is the really hard part. And I can't think of a reason why I would socialize a really, really good idea with anybody about my customers. Why would I just go around telling venture capitalists and competitors and incumbent say, I got this really great idea? Because one of them might try it. And if it is a really great idea, I want as much time as possible to execute on that idea. The only group I think you should really socialize your good ideas with are your customers.

Look, if ideas don't matter and you just run through a smattering of ideas as quickly as possible, I'm not sure if you went back through the history of ideas, business software or consumer, that you would find that certainly there's exceptions, that you would find that the founders just tried something for three weeks and like, ah, it didn't work, let's move on. You tend to start with maybe an idea, and the specificity of the idea is really important. There is an abstract idea. People should be connected together on the internet.

That might be the idea, or maybe people like sharing foot. That might be the idea. If you're convicted on that, you need to try a number of things around that for a while. there's the big idea and then there's the idea of how to implement the idea, and then there's the actual implementation of the big idea and the sort of the sub-idea. And I think there's a lot of nuance that gets lost when people talk about this idea of ideas versus execution. You said something earlier.

I forgot to follow up on, which is around measuring opportunity costs, especially in a fast-growing company. How do you do that? Everyone talks about opportunity cost. It's probably the biggest cost, blah, blah, blah. It just seems like, well, what's the next highest return thing we could be doing? What's the outcome from that? So what's our best alternative?

But how do you do that? It just seems like a really hard thing to actually do in practice and use in real decision making. How did you do it? I can tell you how we did it at Uber. We looked at the penetration in early zip codes and we said, what if the entire globe had the same penetration as these zip codes? The number is striking.

We came up with an early estimate. I make it sound official, but this is more of a general conversation with the team. This is maybe three months after launch. we basically assumed that we would get to about $20 billion of GMV before this thing started to tap out. And this was three months into launch. We'll give ourselves credit for estimating, at least in a close range. Without COVID, it would probably be more like 30. And so we just said, if it's really that big, that's $3 billion of net revenue, what's the cost of getting somewhere second versus getting there first? We launched in Toronto for a very good reason, which was

he was very under penetrated from a food delivery standpoint, but economically, we had on amazing team there, of course, too. Economically, it was very right. It had the dynamics of a good market. We had a team we knew that we would execute, so we took those variables off the table. And so when they penetrated, we said, that's what a fresh market looks like. And then we launched a very competitive market, like in New York. This is what a very competitive market looks like.

So when you then map out the world, you can say, like, whoa, most of the world was Greenfield. At the time, we launched Eats in Miami, and I might be getting this off by a little bit, I think it's right. There were 19,000 food deliveries per. a week through an aggregator, not dominoes and all of that, but through a third-party aggregator. This is late 2015. Now I think it's got like a milly. So that's how we did it. We just extrapolated out the micro to the macro. We had a rough idea of the unit economics so we could swag the opportunity

accounts. What's it like being inside of a consumer business like this that people have so many opinions about? Because I'm more used to the B2B. There's not a lot of talk about some enterprise piece of software other than amongst its customer base. But it was something like this with amazing growth and lots of great stories about the business and controversial and iconic leader and Travis, what was it like hearing stuff about the business that I'm sure was off completely or by a little bit and not letting that distract you from the task at hand. I'm just always so interested, now that I'm on the inside of several companies as a board member, it's just so obviously different inside than it is outside, but it seems hard to manage for a company like Uber.

So what are lessons to do in there? Particularly hard in 2017 when the entire company went through a culture crisis and didn't have a CEO for four or five months. Not to over-emphasize it because I'm not really a touchy-feely guy in general, but like this is where the caring sort of comes in. How is my team going to feel when this world comes and how do I precondition them and how do I feel strategic about this and how do I implement communication to them that helps them with this, knowing that it's going to be a problem?

And so there's a couple things that I've always said that I've always said that I've pulled from my history. The one is people say stuff. Just say stuff. Experts say stuff. Pundits say stuff. He gets put in articles and the world and blogs. It's perfectly correct. I used to just say that in the organization. I'm like, look, guys, people say things. Just because they said it doesn't mean it's true. There's plenty of the stuff that you hear that's just wrong. And then the second part of it is, and this is particularly important when you're going through all the change that the company went through to execute like that team did. It was just that team meaning the Eats team. I say it almost in a third party way because it felt that way. It was like they were executing

independent of what I was doing. It is like the end is never the end. And we are so good, especially smart people and motivated people, we are so good at coming up with very good reasons why this has to be the end. And we do it as a way to escape the pain that we're feeling at the moment because you just want to let go. When you're working out, you want to stop working out. It hurts. When everyone around you is telling you this thing's going in the toilet and the unit economics are totally broken because the funding environment is totally soured the business and restaurants don't like you because of your fees and drivers are going to rebel.

All these things, it is very rational to say it's the end. We don't have a CEO or it's transitioning and a lot of people did. The churned at Uber went way up whenever the hardest times hit, like the employee churn. People left and people came. but that's how you manage it. You just try to call it before it happens and give people some tools to the extent that I can do that,

but that was my job as a leader, give people some tools to just realize like, hey, the world will keep marching. The second thing, and I'll labor on this for a little bit more, because I have a couple mechanisms I do myself, which I would share with the organization. When I joined Uber,

and this was what preconditioning I did for myself for stress, was I put on a piece of paper. We had just had a sign a month before, and I'm like, this is going to be hard. How do I balance? all this. I wrote my like order of priorities that I would never break. I break them once in a while, but it was basically health at the top, then family, then career, then money, then things. If you go through that, if I'm not healthy, I can't take care of my family. My family's not happy. I can't

take care of my career and so on and so forth. And the main one I would switch is health and family. So many times I was at the breaking point myself and I would tell the team this and I said, look, I have to just go exercise. I have to spend time with my kids. I can't do this. I can't do this for that long in order to be long on this journey. And then the last thing I'll say is compartmentalization, breaking down the problem is something that's also really, really important. Dealing with external stressors, like I think you initially mentioned in your question, dealing with your own self-doubt, which is really what those external stressors cause. I had this thing where I wake up in the morning in a bit of a panic. You're competing with everybody in the world. So much shit going on. Yeah.

Yeah, so much. Everybody's telling you, you know, where you're doing a good job? no one's telling you you are, where you're doing a bad job, everybody thinks you're the worst. So I would literally just go through this checklist in the morning. I woke up, how is the business today? And what did it change from yesterday to today? From a customer standpoint, not from an internal standpoint, not much. Business is still growing? Is my team still in place?

Is anyone at risk in place? Okay, no, they seem happy. If the team's in place, do I still have funding because I'm burning because I'm trying to grow and I'm trying to compete? Okay, I do. I have six more months. for worry about going back then. Okay, don't need to worry about those things. Is it growing? Are we improving on unit economics? Are people happy? Do I see people fighting? You can break down as a, I mean, I wasn't formally a CEO, but I was head of this business unit and kind of

CEO of the business. If you break down the problems, then you finally get down to the thing that you actually need to worry about. And when you do that, then you can take all the other stuff and put it in a closet and just focus on that and feel comfortable with that because you went to this checklist. I love all of that. It's also interesting. that you've got health above family, which I think when you might see that list for the first time, like, wait a minute, shouldn't family be at the, but then you realize the reasoning behind it. I'm a huge fan of methods like that to keep yourself on pace. I love it. As you're now outside of that business and working with benchmark, looking at companies, I know still working with companies,

but also have your investor hat on, how would you describe the landscape that you were in on the ground floor now from a higher viewpoint today? You could interpret that as ride sharing, local delivery, however broad you want to think about the umbrella. How would you just describe like the state of the market, the state of the businesses in it? You think it's a growing market? Do you think they're good businesses? What is your more analytical, less operational take on the entire space now? They are good businesses. So start there. They were built during a time when you could solve problems with money. If you have enough money for enough time, it just becomes the culture of how you build companies. The structure of a lot of these companies is not what the

structure should be. Here's a thought exercise. If you were to restart any of these businesses, forget over for a second, DoorDash, Ropi, delivery, whoever. What would the work structure and cost structure look like if you were to just restart it from today? Or if that's too painful because who knows where you would have got for a revenue standpoint, what would you have done if you would start that in the capital environment in the 2000s rather than the 2010s? I think you'd have a lot lower had count. I think you would have grown a lot slower, but you'd have a lot more runway. I think a lot of the growth was pulled forward by the capital. The turbulence created by hypergrowth needs to be paid down, and that is a very hard thing to do for these CEOs that are

dealing with now. And I have a lot of sympathy and empathy for how hard that must be to go from, hey, the cost structure is kind of not where it needs to be. The space in general, I think it's misunderstood because of this. I've seen the unit economics. The rise of business is now showing at Uber that they can generate quite a bit of cash. I think there's a lot of people who believe this for a long time, but there's a lot of people who are like, hey, it's been a long time. Where's all the cash flow? And you have to pay attention to the full cycle. If you're going to grow 600% a year or whatever, then 300%, then 250, whatever the numbers are, that hyper progress is going to come with hyper pain at some point. There's an old.

of adage. Progress is pain. That's because change hurts because people don't like change. Progress is pain. The hyper progress is hyperpane. It's just a function of when you take it. And I think those businesses are taking it right now. What have you most learned from your time with the partners at Benchmark as a venture partner there about investing? What are the most interesting ways you viewed the world have changed or have your curiosity that maybe you wouldn't have been focused on prior to working with them in this recent period? As you probably know, Each partner is different, and so you learn different things from them. The opportunities that Benchmark gets excited about are probably the opportunities that they would

join themselves. And they do join in a very substantial way. I shouldn't say that they don't do it. But they would join as an operator in a fully committed way. That's the conviction you feel, or at least I felt being there. I think that's different than, hey, we need shots on goal or, hey, we got to put a lot of bets out there because you never know and power law and all this other stuff. You go into an investor mindset thinking my conviction can be a little bit less on an idea because I'm in a place

like more bets. And I suppose that's a style. And I think what I've learned is my original instincts of, hey, if I wouldn't join it, why would I do this? That feels like my interpretation of the filter, but they do so few deals that there's probably some version of that in the organization. It's such a powerful idea that's so easy to ignore as an investor with incentives to put capital out, incentives to make lots of investments from. I mean, frankly, the whole industry is set up so that your job is to make investments, not to not make investments. And yet, I think what you're describing is almost always the answer is you're not going to make an investment. So it's a really interesting incentive question and hard standard to uphold,

which I think is what makes it interesting. Jason, this has been so interesting, so much fun. There's nothing more interesting to me than learning from someone that's built something tangible and hard fought and competitive, which obviously Uber Eats and your prior stops are all those things. I think you know my traditional closing question. What's the kindest thing that anyone's ever done for you? This is going to sound weird, but it's the first thing that popped into my head. The $10,000 seat check that we got for scour.net, which was our first company, it was from two of the dads of two of the five co-founders of this business. we started at EBCLA, they were like, yeah, we view this as part of your education. And for us, we were buying a server and data centers and things like that.

And we needed it to keep this service live. And the service was growing. And it just felt like, oh, my God, someone believed in us and it's really just for our own benefits. That's pretty amazing that that happens in the world. My own dad did that for me with a business that I started. My dad did not grow up in an age where you've got on startups. tried to dissuade me from that.

But he did the same thing for my next business. It wasn't a lot, but it was a vote of, hey, I worked hard for this, and I believe that you can return this to me someday. And even if you can't, I view it as part of your journey. Those two moments, I think, were just like, wow. You mentioned your dad there. And earlier you mentioned your upbringing, which in a way that made it sound like clearly was influential. How did you describe your upbringing and what it baked into your personality? I think financial austerity of a middle class family that acted lower middle class.

Money was always very central to my upbringing. Lack of it was a cause of a lot of stress, or at least that's how I placed it. I think that's probably why I picked those as the moments. For all of the saving a dollar here or worrying about money there or maybe not treating someone right because of money, the idea that you could just give it as part of someone's career. journey. I think that's why I'm connecting those two things together. And also, I grew up in a household that was very, or really two households. My parents were divorced. They were very risk-averse, which parents can be to kids, which makes sense now that I have my own kids. You don't want them to

fail. But it really motivated me to prove that I didn't have to be that way, that the world was not as riskier as scary, as people said. I'm going to prove that this is a journey. I didn't have to go be an account because it's a safe thing. I didn't have to be an engineer because there's always a job. I could go do something else. Jason's going to be hard to count the number of gems from this conversation when we're adding it. Thank you so much for your time. Yeah, thank you. It's awesome to be here. Appreciate it. If you enjoy this episode, check out join colossus.com. There you'll find every episode of this podcast complete with transcripts, show notes, and resources to keep learning. You can also sign up for our newsletter, Colossus Weekly, where we condense episodes to the big ideas, quotations,

and more, as well as share the best content we find on the internet every week.