5: >> I'm Rex Woodbury one of the view from the top student leaders on behalf of the leadership team thank you for joining us today.

11: We're excited to be able to bring an outstanding diverse slate of speakers to Stanford this year to inspire all of us as future leaders.

20: And now it's my pleasure to welcome Dean Levin to introduce our guest today. >> Welcome to view from the top, today I'm delighted to introduce Tony Xu,

29: MBA class of 2013, the co founder and the CEO of DoorDash.

35: It's a special pleasure to have Tony given the history and close connection that he and DoorDash have with the GSB.

41: I know we've all developed an even deeper appreciation of Tony and his company over the last 12 months and I will say that in 11 family.

52: I think his reputation peaked just after my son Ben's 13th birthday in December,

58: so Ben received a bunch of gift cards from Amazon and Niki.

63: And he came to me after his birthday and he said, I want to cash these in and I said, why you love Amazon and Niki?

71: And he said, yes but I want to swap them for DoorDash gift card, so we're all looking forward to hearing from Tony and

78: today's conversation with our own Joy Huang MBA class of 2021. Thank you Tony for joining us it is really wonderful to have you at the virtual GSB.

89: >> It's always really exciting to have an alum join us is especially for a product that all of us are so familiar with.

98: >> That's good to hear [LAUGH]. >> Yeah, my guess is a lot of audience dialing in are watching this interview

104: with food they got from DoorDash right now. >> Send me feedback at any point in time >> [LAUGH]

109: >> And just tony@doordash.com >> Be careful what you wish for Tony. >> [LAUGH] >> I know people are probably really eager

116: to hear about all the lessons you learn from the GSB. So this might come as a bit of a surprise for some of the MBA students.

124: But you once said that more things growing up as a kid prepare you to start a business than anything else, could you tell us more about that?

135: >> Sure, I think it's really hard to follow maybe a curricula or some prescribed set of steps to do something that has very high variance.

146: Obviously when we started the company out of my apartment off campus, we weren't thinking about what DoorDash would look like seven, eight years later.

155: And I think that's really what starting a company really is and so when I think there's lots of variants and you're learning pretty much on the fly.

162: A new job every week in the beginning later on, maybe every month, every six months.

168: You really have to find other first principles that have accumulated

174: in other parts of your life to really get you over the hump at times.

179: And so, for me, growing up as a kid, I came to this country with my parents when I was very young I was five years old we immigrated here from China.

187: Mom and dad really came here to make a better life for me and for dad that meant going to school, getting his PhD, and working a job actually.

196: At a restaurant on campus at the University of Illinois, for my mom working three jobs a day for the first 12 years of her journey here.

204: And, so I didn't see them that often and, so most of my time was pretty unstructured and I think as a result that unstructured time was what was able to.

215: Obviously I wasn't thinking about becoming an entrepreneur when I was five years old, nor did I ever think [LAUGH] what would happen from my unstructured time but

222: I think some of the benefits really came. I'm from trialing a bunch of new things and

227: I think that it's that time that really, taught me a lot of lessons about how do you learn things for the first time.

235: For me, that meant language that meant earning income divided Nintendo.

241: That meant many small things, but when I put them together, those are some of the lessons that I remember most.

247: >> You mentioned that your family moved from China to Illinois when you're young and little did you know you would soon move again this time from Illinois to San Jose.

257: And when you got to San Jose, one of your teachers told you that you were two years behind your class.

263: And in response, you said that you will become the valedictorian of your class.

269: What you did what was going through your mind when you said that to your teacher?

275: >> [LAUGH] I don't know what probably went through my head obviously, I wasn't thinking that clearly when that common came out because in some ways

282: he absolutely was right. I was behind I grew up in a community in Illinois that

288: prioritized other things beyond the classroom, a lot of it was in sports, and that taught me a lot.

296: Playing basketball, for instance was really one of the main ways in which I learned English, but when I came to San José in high school,

305: or to finish high school I really was in a world of culture shock.

311: The bay area I later found out [LAUGH] is a very competitive area especially in the classroom and the school I went to I think students

321: were studying all the time in during the school day after hours.

327: I had not really done my homework prior, when I was growing up so

332: I probably didn't have the model preparation for coming into school.

339: But I think one thing that I've always really believed about,

345: both my family as well as, many of the different moves I've made as a kid.

352: Because I moved a lot actually, even when I was in Illinois, different schools, things like this is really that.

362: I gained confidence even back then that I could survive in new environments. Whether that meant making friends, whether it meant becoming a chameleon and

371: learning maybe what it was that was, I guess, important to that community.

376: And so I had confidence that even though I didn't really study versus some of my peers at that time.

383: That if I kind of really put my mind to it and maybe took all the attention, I used to spend, maybe playing sports or

391: putting in towards the classroom that I could actually be successful.

396: So that's probably what I was thinking, I doubt any of that was what was in my mind when the comments slipped out but it ultimately it worked out.

405: >> It's truly impressive to see that you took a lot of what other people might see as obstacles in stride.

412: And then built this drive that we will also see in the story of how you created DoorDash.

418: And Tony, you were mentioning that your mom and your dad had a lot of influence on you as a child.

425: And your mom really is an entrepreneur and she opened her small businesses subsequently that you had worked in.

433: So it seems like they really instilled this entrepreneurial spirit in you with their own examples.

439: I'm curious, how did that shape the type of business that you then wanted to create?

444: >> Yeah, I really don't think this is a story you can tell, back then looking forward, it's only a story maybe you can connect the dots looking backwards.

452: You know, for me, my parents, you know, journey really is the classic immigrant story and

459: many folks can appreciate coming into this country without much.

465: And my parents came here with maybe $200 in the bank, and it was we're going to make it or we're not situation.

472: There wasn't really much of an in between road, but I think when you have nothing to lose You also have a lot of upsight as a result and

479: I think that was probably one of the earliest lessons I took where I think on one hand for me, it was just hanging out with my mom.

487: That was really what I was doing when I was working inside of a restaurant or washing dishes alongside her and things like that.

493: But on the other hand, I learned that this was the way that you can earn your way into better things.

500: For me, it's why I worked at a really young age. It was to save up enough money to buy Nintendo,

507: to buy the games that came within Nintendo. So I'm not talking about big things per se, but

512: that's what I thought a lot about as a kid. It's why I decided to mow lawns, when I was nine years old.

518: And effectively knock on doors so that I can, you know, create different shapes because if you cut grass at different heights,

525: that's how you can create different shapes. And in places like Illinois where there's a lot of land,

531: there's a lot of grass you can cut. Which means that you can earn your way to more video games and so on.

536: I think that these are small things when I look backwards. And I think what it taught me was that if you put your mind to which

545: ever new thing it is, and you may be, either because of reality or

550: because of putting yourself in that frame of mind think that you have nothing to lose, there's a lot of upsight.

561: >> Yeah. And we were just talking about how this young enterprising side of you and

567: I thought that's a great segue into how you later than also learn about other aspects of entrepreneurship at the GSB.

574: And I know many of us in this audience are taking startup garage right now, which is famously the class where you started DoorDash,

583: back then was still delivery.com. And true to this Lean Startup MVP methodology, you and

589: your co founders were really scrappy at the beginning. One example that I really loved was how you guys track drivers with the find my

597: friends function on the iPhone instead of building out a very sophisticated back end technology for it.

605: What were some lessons that you learned from that class that stuck with you? >> Yeah, well, I think the first lesson that we learned is what's ultimately the,

616: almost overly simplified problem that you're trying to solve.

622: And if you can ask yourself that question very crisply,

627: I think that's a very clarifying way to really test ideas very, very quickly.

634: I think one of the earliest things that made DoorDash,

640: I think a fast moving place was that we always try to simplify the question we were trying to solve.

649: We knew, for example, that we wanted to ultimately build last mile delivery. That's way too big of an endeavor for,

656: just four of us inside of a classroom or inside of my apartment.

661: And so instead, we asked ourselves three questions. We asked ourselves, is this a service that students or

668: people, customers would actually want and would they actually pay for it? And I believe at the time we said, would they pay $6 for it?

677: Is this a service that restaurants would actually pay for and I think at the time, I'm trying to remember this is, in 2012,

686: maybe the startup garage class that we took. So, almost eight years ago. I think it was like 15% or something like that with restaurants.

695: And would drivers want this job enough to earn, to see it as an opportunity to work.

700: And that was it, those were the three questions that we were trying to answer.

707: And I think the first lesson again is to really simplify what you're trying to do.

713: Because I think oftentimes when you try to solve a really hard problem as the first,

719: I guess, quote-unquote, go to market to test your ideas, you might not even know what it is that you're actually trying to solve?

729: And then the second thing is just that you learn by doing. I think none of us really had a background in logistics or delivery.

739: We learned by doing deliveries. We learned by working alongside restaurateurs.

744: We learned by doing customer service. And when I think of,

750: that's kind of what we were doing when we were in startup garage. I don't necessarily know if we linked those two ideas when we're doing it but

759: learning by doing and the simplifying the question that you're trying to solve,

764: I think, were pretty important principles. Yeah, I think it's really interesting because they're almost a contrast where

773: you're trying to really simplify but also, at the same time you want to embrace the complexity of all the operations that is entailed in the business.

781: And I'd like to expand on the second point a bit too. Because like you mentioned, the four of you drove as drivers for FedEx and

788: Domino's to learn about how delivery works and and you drove all the delivery for DoorDash in the in the first year to a year and a half.

798: You have now synthesise this into a core principle for DoorDash that you call getting to the lowest level of detail.

805: Could you help us first understand what this principle means? >> Yeah.

810: So I, you know I am, I mentioned that we didn't really, we weren't really students

816: or we weren't really experts in logistics per se, so we had to be students of it.

822: And the only way we thought we could be students was by actually doing the work. And so when we started doing deliveries, we started noticing that, wow,

830: some restaurants are really, really fast and efficient and other restaurants take a little bit longer.

836: Some Dashers, the drivers on our platform, are really fast and accurate and others are not as much.

846: Some customers really seem to care about a certain type of salad for example, versus,

856: what others may just consider it to be a Monday meal.

862: And you start realizing that flee to bring something to you, something especially that perishes in minutes, whether it's ice cream melting or

871: pizzas getting cold, and doing that over and over again consistently, because you know this is something that you don't really get any points for

881: the first set of deliveries that you got right. If you got the current one wrong, that tends to be what customers remember.

889: There's a lot of steps, we ultimately identified about 20 steps in the process.

895: And it really came from doing the work. And when problems arose it classic asking five why's analysis to

902: get to the lowest branch of what the actual problem is. Because most of the time what we realized was in something that seems as,

912: quote unquote, easy or mundane as delivery, if you're at the surface level,

918: you're never going to actually realize what the problems are. They're always hidden somewhere.

926: And it maybe a restaurant is delayed one day actually because someone didn't show up to work.

933: But you never would have guessed that, if you're just looking at the data of how long it takes them to prepare something, for example.

939: That would never probably register on your dashboard. So operating at the lowest level level of detail is pretty much trying to find

946: your way to the right problem. And it almost never is at the surface.

952: That's really interesting insight, and I'm curious once you have learnt this, how do you then impart this on the employees

960: to make sure that this principle actually makes the company more successful?

966: >> Yeah, I don't think it's necessarily one mechanism or two mechanisms.

971: But it's something certainly over the years that I would say lots of great people at the company have developed and

977: kind of put together different things that we do. One of the things that we found effective is really writing things down.

984: DoorDash is a pretty strong writing culture, we tend to start meetings by reading documents that people have written.

991: And really these documents are you could think of them almost as hypothesis trees, almost, where we're guessing what the problem is.

998: And we're trying to get to the lowest level branch and to really understand and we're really looking for the disconfirming evidence to try and see whether or

1007: not we're just confirming our own bias or actually, we're on the right path to seeing what the right problems are.

1014: I would say, other mechanisms that DoorDash have found to be maybe

1020: helpful in this endeavor, is to actually do the work.

1025: So, this is why for instance, we have everyone at the company do deliveries once a month or do customer service once a month.

1033: I do customer service actually every day because sometimes it's really hard just looking at the data to get to the digging.

1042: And when we find that the data I guess, conflict with what customers are saying that it just probably means we're missing something,

1052: we're missing the measurement of something. That's most of the time what is actually happening.

1059: And so this teaches us how to improve our audit mechanisms and our measurement methods.

1064: And so I don't think there's a silver bullet here. I think it's a maniacal focus on making

1071: sure that we can work the inputs to each one of these branches. And that's really I think the spirit of how DoorDash operates.

1080: >> Yeah, that's a really great insight because it's something that's seemingly unscalable on a surface, but it actually has been one thing that's crucial to

1088: the way DoorDash has been successful. >> Yeah, and I think over time you find that there are certain patterns that you

1095: can actually build products for build tools to automate. But I think if you started that way, it would be really, really difficult because

1103: most likely you would be automating something that does not even exist. >> Yeah, I think we are really

1109: obsessed with data today and >> [LAUGH] >> It is really good to keep that in check and know that it is really worth it to deliver and order once in a while.

1118: I know Tony, many of us have been following DoorDash for a while and seeing the wildly successful IPO that you recently had.

1125: Congratulations on that by the way. >> Thanks. I mean big team effort. >> [LAUGH] >> I think sometimes we forget that

1132: DoorDash wasn't always a sure bet, and in fact, the company almost ran out of money by the end of 2017.

1140: So if you could take us back with you to that moment, the investors have soured on the delivery space, and you had spent six months

1148: trying to get a lead investor for the next round and all you heard was no.

1154: What kept you going during that time and how do you convince the others to follow you?

1162: >> Yeah, I think again there isn't a silver bullet and you're right.

1168: DoorDash had many ups and downs and we still do.

1173: I don't think the journey to solving some of the problems we want to solve, transforming every brick and mortar business, for example, is going to happen

1182: overnight and you're undoubtedly going to get your doubters along the way.

1188: And so when I think back on some of our toughest years, I think the first

1193: really centered around this value of the company in which we call truth seeking.

1200: And at the company we tend like to say bring your data and

1205: your insights, but not necessarily your opinions.

1210: [LAUGH] And when we looked at what was happening with the company, the data and the insights would suggest that we had a very healthy company.

1220: And that the products that we had been building were very differentiated and

1225: industry leading. Now on the flip side, you're right, a business like DoorDash requires quite a lot of capital in order to scale.

1235: And, we were effectively operating a portfolio of markets, some of which are much, the majority of which were longer and therefore not yet profitable.

1245: And so that's why we needed capital. And I think because of the transparency that we had for

1250: many years even prior to the difficult moments that DoorDash had. I think people in the company knew that there was a disconnect between what was

1259: actually happening with our customers and the growth of the business and maybe what some outsiders had written about us or thought about us.

1268: I think that was important. I think just being transparent with information,

1273: which was something that happened far before I think we had difficult moments than when the difficult moments inevitably came.

1283: And then I think the second thing was, there isn't anything you can fully do to

1289: control this in terms of what external validation or lack thereof hits you,

1295: but for us it was this value of choosing optimism and building a plan.

1300: We were prepared for pretty much any scenario, regardless of of what amount of capitalism come into the company.

1308: I think by demonstrating that not just to a small group of people at the company but

1314: really to pretty much everyone at the time. I think people had confidence that regardless of which way things

1322: were going to go in terms of what was going to happen externally, that we would find a way out of it.

1329: And so I think that's the best you can do sometimes. Because you don't get to control everything when you're not yet

1336: profitable as a company. And and so I think the second best thing when you don't have

1343: full control is that you plan for the variance, and whether that's up variance or down variance you should be prepared.

1354: Whether that means it's difficult to raise financing or whether a global pandemic hits.

1359: And I think those are the things that are really, really important when the tough moments come.

1364: >> I love your choice of wording, which is you can choose optimism and I think that sometimes we tend to forget that because of the outside validation

1373: that we inevitably crave that we forget that it is actually something we can consciously have, and then just focus on doing what is right to move forward.

1382: [CROSSTALK] >> I think it goes both ways, right? And when things are going really, really well, you should also make sure you're

1389: truth seeking, because you're [LAUGH] never as good or as bad as people say you are and I think that maybe the customer information is probably the most telling.

1398: But even that's sometimes can be swayed by macro events like a pandemic, for example.

1406: So I think it's really important to be intellectually honest. And I think if you can do that, and if you can set that at the beginning of your

1413: journey, if you're building a team or building a company over time, I think that lays the foundation to carry you through the really difficult times.

1424: And speaking of things going really well, after that difficult period for

1429: you, DoorDash was on this trajectory of rapid growth where, if I remember the numbers correctly, I think DoorDash's market share grew from

1439: 17% to 50% in just two years, which that kind of growth is just crazy.

1445: And so for you personally, how did you make sure that you could constantly learn and adapt as the CEO?

1453: >> Yeah, I think it's always focusing on the inputs, not the outcomes. In some ways, much of the performance or the out performance of the company

1461: came in many years before I think when you saw some of the results.

1467: The way I like to explain the things usually everyone when you start, we're all born with zero market share [LAUGH].

1475: And you don't tend to just do that overnight. There's probably lots of other things that were happening that maybe had not yet

1483: demonstrated the fruit yet of the progress. Because, I think this is the hardship about starting a company really, or

1491: really frankly doing anything that will take a period of time, which is what you do today may or may not pay off right away.

1499: And I think your ability to I think articulate that to those around you is really important.

1509: And I think the best way to make sure that you're

1514: on the right path is to look at the inputs. Because the outcomes may not always be there every month or every week.

1523: And so at DoorDash, for us it's always been about improving the selection that

1529: we offer our customers, the quality of the delivery, in terms of our timeliness and

1534: our speed and our accuracy as well as the affordability of the service. That's what we were focused on.

1540: We weren't looking necessarily at outcome metrics per se. We were looking a lot at the input metrics.

1546: And the maniacal focus on that was the bet that we made that ultimately would translate into things that maybe others might appreciate or care about.

1556: Because at some point, you do have to show your cards, so to speak. And make sure that maybe some of the ideas you had would translate into real results.

1565: And for us, I think we always understood that relationship between what we were working on and what the outcomes would be.

1573: Maybe they wouldn't happen in the same time period that we were working on improving the inputs. But I think having that understanding and then having the patience, if you will,

1583: to keep working the inputs and then letting the score take care of itself, I think those are things that we both had to solve.

1593: >> And as that business grow and, like you mentioned, at the very top it's a different company every six months, right?

1599: So how did you consciously think about adapting your leadership style, and making

1605: sure that you are still the best person for this company for this specific stage?

1612: >> Yeah, well, I think in the beginning, it's really about learning how to build a product, and how to build a product that people really want.

1620: It took some time to figure out how do you actually improve these inputs? And you're almost making tweaks along the way.

1627: And over time you're trying to build a repeatable system in which you can take that product to more and more segments of the population.

1635: For us that happened city by city but for other types of products that can happen maybe customer group by customer group, something like this.

1641: And then I think the next evolution was really learning about skills and different teams.

1647: I'd never worked a day in finance before in my life. So it's really hard to hire a CFO sometimes.

1653: Yes, I know I went in to the GSB but I probably am not a finance expert.

1660: So what does it mean to be a world class CFO? What does it mean to be a world class head of engineering or head of sales?

1669: This is what I mean by skills. These are not necessarily skills I had. But these are skills that I had to go and try to learn from those who were

1677: the best to form my own view of what that meant, and how that might translate into a successful outcome in an environment like DoorDash is.

1687: And then over time, it's about learning how do you build systems. Because once there's a sufficient number of people,

1696: it's pretty hard to just rely on your memory or on I guess your stories to,

1702: I think, allow people to make great decisions very quickly with high quality.

1708: And so the obsession then morphs towards what are the mechanisms we're going to build to allow speed,

1716: as well as the audit mechanisms that we're going to create to maximize quality?

1722: And so I don't think there's necessarily one set of things you have to learn.

1729: I think ultimately, the way that I try to measure myself

1734: is how are we actually doing along these dimensions?

1739: Maybe it's tough to see in a one day or one week grading period. But you can se it, I think, over six to 12 months.

1747: And you can see it year on year, are you improving on the product? Are you improving on the skills?

1753: Are you improving in some of the systems? And then what morphs even over time is how do

1759: you actually scale this culture that you built? Some of it intentionally and

1765: I'm sure some of it co-created with all the folks that are joining the company. And so I think there are ways in which you can look at rates of progress.

1773: But I think it's first and foremost recognizing what is it that you're actually trying to get better at? Because it's hard to get better at all of those things at once.

1781: But that's at least what I found in our journey thus far. >> That's a really good mental model to have.

1787: And, Tony, I know you've said a lot that you really started DoorDash out of a passion to help small business owners like your mom.

1794: And today with more than 300,000 restaurants and 200,000 dashers

1800: on the platform, this passion has also really evolved into a responsibility.

1805: So what is your vision and your plan for helping the local businesses and gig workers thrive not just during the pandemic, but beyond?

1816: >> Yeah, so the goal of the company has always been more about food delivery. It's always been about how do we grow and empower local economies.

1824: And the way we do that is really with two types of products. The grow part of our mission is really exemplified by the marketplace,

1832: that's our app. And I think what most consumers know us for, in which we bring you lunch or dinner, or now increasingly so,

1840: convenience items and grocery supplies and things like this. But that's one half of the goal.

1847: The other one half of the goal is to empower these businesses to do it on their own. This is our platform in which we're taking products from our marketplace and

1855: we're giving them to merchants so that they don't have to rely on us, that they can create their own digital channels.

1861: So for example, there are many many many businesses that actually just, they sell through their own channel, they don't necessarily sell through DoorDash.

1871: But they use the dashers on our network and the software that we've created to facilitate their own on demand or same day deliveries.

1879: This is called DoorDash Drive. In the event that they don't have their own engineering teams,

1884: to create their ordering systems, we have direct Storefront that powers them to create their own online stores that will integrate into all of their back of house

1893: systems so that they can compete in e-commerce for the first time. And obviously you're going to have to do more than just logistics or

1901: ordering in order to actually build a full stack service. That we have on our marketplace. There's a lot of work that's left to go.

1909: So our responsibility I think is not just to help grow and build the largest local commerce marketplace.

1918: But also to teach and actually give these tools so that they don't have to rely on us one day, because that's really important for their longevity.

1928: With respect to dashers, I think dashers have taught me probably

1934: the most of any audience over the eight years that we've been working on DoorDash.

1941: So DoorDash has millions of dashers actually every month, and it's been very fascinating to see how the vast majority,

1950: 90% of them worked fewer than 10 hours a week.

1955: This is many ways is at least in what I've observed, never have I seen this type of behavior in any work environment before.

1967: And over 80% of them have full time jobs of some form, and what they're looking for flexibility to pick and choose when they want to work.

1976: And they're looking for value I think in the hours that they're actually seeking to work on the platform.

1983: And so our responsibility is very, very deep not just to taking care of those opportunities,

1990: also making sure that we can work with everyone such as elected officials,

1996: regulators, and anyone else interested frankly in creating the future of work.

2002: Which I think looks very different, I think in many ways dashers on DoorDash look very similar to consumers in the sense that they

2012: value their time as much or sometimes more than money. And they've in effect are choosing some of these part time gig opportunities so

2023: that they can save for a project, whatever that may be. Whether that's buying a gift for someone or starting an orphanage or

2034: using the money and buying their kids something for school.

2041: And so I think that responsibility is very, very serious, and it's one where we have to help them achieve their goals of flexibility, and value.

2051: But also work with everyone else who've created this labor system

2056: over the last seven decades, eight decades in the US, and change

2062: it in a way that actually gives them also the protections that they deserve.

2068: And so that's what we're trying to do. I think it's not going to happen overnight. It's a very important part of what we do.

2075: And it's a very important part of making sure that our work with merchants and

2081: our work with dashers can continue the following statement. Which is that the vast majority of GDP in most

2087: cities is still produced by businesses on the streets, small, medium, and large.

2093: And that's why it's so important to us. >> Yeah, you mentioned the future of work I think In the process of transitioning to

2101: that more flexible kind of work. It's really on us to really think about the dashers, the gig workers that we work

2108: with and think more broadly than just our business but also how we're impacting other people's lives.

2114: So that was a really good reminder to have. And because of the pandemic, Tony,

2120: I think meant all of us had to face unexpected challenges this year.

2125: So in that spirit, I have a closing question for you before we turn to student Q&A.

2132: What are the principles that you rely on as a leader during the toughest times.

2140: >> Well, I think the first is to do the right thing which often is probably the hard thing.

2148: And during this pandemic the right and the hard thing to do was to shelve some of our business plans.

2157: And actually make sure that the community was safe, that the dashers would be safe, that merchants would have liquidity,

2167: that we can help the community, any organization that may need our help.

2173: Whether it's delivering meals to children who no longer can get them anymore because schools are closed ,or to give healthcare workers free

2183: meals because they're working 24/7. So that's the first thing. The second thing is that you have to have a bias for action.

2195: Because you can't let the uncertainty if you will

2200: necessarily I guess stop all things or stop all activity.

2207: And most of the time that that bias for action means it's just focusing on what you can control.

2212: It's not trying to say, by doing this we're going to control some outcome. But by doing this that we will have some influence and

2221: maybe that work combined with maybe luck a lot of times, that that's what will yield the good outcome.

2229: And for us, for example, at the beginning of the pandemic it wasn't

2234: obvious to me that even the restaurant kitchens would be allowed to stay open.

2241: Actually if you looked at what happened across the world, some countries whether it's China and some others, like the UK

2247: actually shut down restaurants altogether completely, including the kitchens. Which meant the takeout or the delivery business would be stopped as well.

2257: And those are things you have to be prepared for. And so this is what I mean when I say you can't influence that,

2266: you can work alongside governments and elected officials on that.

2271: But you can't control that outcome, but you have to be prepared for what the consequences are.

2278: I think the third thing is you have to trust your team.

2284: Because no one person is in get you through any sort of difficult situation, and certainly not a pandemic.

2290: To give you a sense of what it felt like at DoorDash, I meat it pretty much was 24/7 from March through June,

2300: where we had twice a day meetings across probably 20 or so work streams.

2307: When you have something like that, it's not possible to know everything and control that decision making.

2313: And so I think a lot of this obviously starts from preparations that you made a long time ago, such as building great people up,

2324: allowing them to make many decisions prior to when a big moment like this happens.

2330: But you have to trust the team and you have to trust the process that they built to solve the hard problems.

2337: And that's what I'm most proud of when I look back at 2020. It wasn't necessarily what happened with an IPO.

2345: It was how we responded during a difficult time for everyone.

2351: >> That's really valuable advice. And Tony, thank you for sharing your grit, your scrappiness, and

2357: growth mindset with us. With that, we'll now turn to student Q&A, I believe that Jenna is up first.

2364: >> My question is about cities and how they'll evolve. And so you've mentioned a little bit about the future of work,

2369: about sort of changing regulatory environment. I'm curious how else do you see cities evolving over the next ten years,

2376: and what is DoorDash's role or place within that evolution?. >> Sure, well the first thing I would say is,

2382: I think cities are always going to get better. I think I forget who studied this, maybe I'm going to blank on the author but

2391: cities have been one of the most enduring, I guess, organizations that I've ever made

2397: it through the history of time far more enduring than businesses have and

2403: I think again I mentioned earlier that most of GDP still happens inside local communities, I think in the US that's not of 50%.

2412: That number has been true in pretty much every decade that it's been measured and

2418: I think it shows you the resilience of how cities go through obviously economic cycles as well as even pandemic cycles.

2426: So, I'm very optimistic and I think what you're going to see is, I think you're going to see more consistency across cities,

2434: I think we used to maybe think, well you must live in a certain city in order to access certain things, you must live in New York City to access delivery,

2443: you must live in San Francisco to work in technology. I think you're going to see more of that spread, you're going to see a greater

2451: distribution of all the things that I think we each want, convenience for example, or access to different types of jobs in more places.

2460: And this kind of growth of almost hubs, if you will, of all the things that could

2465: because we're all human, we all want similar things at the end of the day, I think is going to be more consistent, not less consistent.

2474: And so I think our role is to play our small part in making sure that we can make every small medium and large business you know,

2482: brick and mortar business make this transformation from doing pretty much everything from customer acquisition to customer service,

2492: inside stores to a world where they do this also with digital channels and

2497: so I think we have a long ways to go and we're excited about the journey ahead.

2503: >> And what other industries or verticals could you see door dashes

2508: last mile fulfillment model A plus B applied well and why? >> Sure, well again, crawl, walk, run, as I like to say and

2518: it's tough doing one thing well, and so as you kind of go forward,

2523: you want to sequence that appropriately. But what one of the reasons why we started with some,

2529: there's a few reasons why we started with restaurants. You know, one of the reasons why we started with restaurants was if the goal

2538: ultimately is to bring you everything inside your city in minutes, not hours or days, we thought it would make sense to start with what we had

2548: guessed to be the highest frequency category first, so that we have lots

2553: of activity because when you have lots of activity in the same geography,

2558: you have basically, the greatest density of work, if you will. The greatest density of work is what allows you to be more efficient and

2567: also achieve a lower cost way of bringing things to places. The other reason why we started with food

2574: is we knew that something like restaurant food is very perishable.

2579: Obviously, you have minutes to get this right and if you get it wrong, you kind of have to solve it then in there, you can't wait to give your password the next day or

2587: something, it's not like that, you kind of have to solve it there and then and so on.

2592: We thought that if we can tackle the hardest problem first, that other categories, at least from a pure last mile logistics perspective,

2601: would be easier to do. Obviously, if you have more time to deliver something less perishable, that's easier than the more time pressure, and greater perishability.

2609: And so, I mean for us, we found that to be true as now we're starting to deliver convenience items from places like CVS, Walgreens 711 as we're

2618: delivering grocery items from places like smart and final or Meijer, Hyvee and other places that seems to be true, but we have a lot of work to do.

2627: I mean, I think one of the hardest things of what we have to do is we have to digitize the physical world.

2633: There does not exist for instance, some easy structured catalogs where you know how long it takes to bake a deep dish pizza versus making a salad or

2643: how many apples are left in aisle six or what are the prices for this item as they are changing.

2649: Maybe by the hour by certain stores because of a promotion or something like that.

2656: I mean, I know we're in a pandemic, but post pandemic, where's the last parking space in San Francisco or Los Angeles?

2663: I mean, these are tricky questions, where we first have to almost study the world first before we start, hopefully,

2670: you know bring value to different people inside that world. So that's still a big part of the journey for us right now, but

2677: I'm confident that you know, based on where we started, that that gives us quite a lot of opportunities to deliver everything else.

2688: >> Thank you both for the questions. Tony, I like to wrap up with a customer review from the top lightning round.

2693: >> All right. >> I think we have maybe 30 seconds left, so I'll be quick. It's only three questions.

2699: The first question is, what is your favorite class of the GSB?

2705: >> I don't know if it's taught anymore. It was called, I think it was called aligning startups in their markets taught

2711: by Andy Ratcliffe and Mark Leslie, I don't know if that class is still there but

2716: though that was those a highlight. >> And what is your go to food delivery order?

2722: Actually I don't have one. So my wife and I, we actually try a different place almost every time where we

2728: were, so we've ordered from about 1100 places so far in the Bay Area.

2734: I guess the most recent order was a new sushi place that we discovered in San Francisco called Sasa Japanese restaurant.

2742: >> Nice. Last question. In just one sentence, what would be your word of advice for

2748: aspiring entrepreneurs at the GSB?

2754: >> Well, if you're an aspiring entrepreneur, I would say

2761: find something you're obsessive about and then that's where I would start.

2768: Because, the journey to building anything of meaning, I believe will take years,

2774: maybe decades and so I think unless you are maniacally focused on it,

2780: I think it is difficult to maybe make it all the way through all the ups and

2785: downs and so find some find something that you're obsessive about.

2790: >> With that, Tony, It's been truly a pleasure. Thank you again for joining us.

2795: >> Of course. Thanks for having me.