how-doordash-turned-food-delivery-into-local-commerce-platform-vassili-samolis-launchpod-logrocket
===

[00:00:00] All right, Vassilis, welcome to the show, man. Glad to have you on. Thank you for having me, Jeff. Good to see you. DoorDash is one that is near and dear to my heart, and not just because I like to order food late at night. What do you order? So I travel a lot when I'm not hosting Launch Pod. We have a lot of dinner networking events for senior product folks like yourself, actually.

I actually end up using DoorDash wherever I am to kind of order dinner. It might be late at night or really early in the morning, and, and it's been a lifesaver a couple times. That's amazing. That's one of my first things, like when I land in a new city, one of the first things I do is I order some food to hit me at the hotel by the second I check in, and it's great.

DoorDash, to me, is, you know, I go on, I order food when I need to. That's great, but there's so much more to it when you think about kind of how it serves restaurants or, or the kind of thought behind where it sits in the restaurant and, and kind of food world and [00:01:00] commerce structure. Maybe we can just dive into that real quick before we kinda do anything else, 'cause I think people don't quite understand the full picture here maybe.

We've come a really long way. DoorDash, I would say our origin story was food delivered, right? Like, we would bring food to you. Mm-hmm. And I think a lot of folks experience DoorDash through the lens of the consumer, right? And they would probably tell you that today, hey, DoorDash brings you pretty much everything your neighbor has to offer.

But the reality is that we serve the communities, not just consumers, and that can mean from your neighborhood bodega to your favorite sushi place. I have a Greek place right next to me, I'm originally from Greece, that I really, really love, and those folks struggle, right? Like, being a local entrepreneur is a really hard thing.

Thousands of restaurants every year in New York go out of business, and I think when you think about the evolution of DoorDash, again, our mission is to empower local communities, and I think we've come a really long way from effectively a delivery platform to a global technology platform, and that's what we're working to do for these folks.

One thing I don't think a lot of people know on my end is I kind of had a side [00:02:00] quest in my career where I ran a really large beer and food festival company in Boston for many, many years, and we got to be really close and friendly with a lot of the kind of food scene and restaurants and restaurateurs across Boston and New England.

And it was always wild to me how hard it was to run a restaurant in any kind of profitable way that you would keep open. That, it's so interesting to see DoorDash go more in that direction or talk about sitting within the kinda local commerce space overall. It's hard, right? It's hard because I talk to local entrepreneurs every week, and I hear these stories, and I think the challenge is that these folks are being asked to be marketers, they're being asked to be operators, they're being asked to cook food in an environment that is highly dynamic.

I don't think the technology solutions that are out there for these folks are that great, which on the inverse is why I think we have a huge opportunity to drive impact for these communities and these entrepreneurs. Let's take five seconds and kind of maybe take an aside in this conversation and give people the [00:03:00] TLDR on how did you get to where you are?

You're VP of product over at DoorDash. That doesn't happen overnight. You got a lot of great companies kind of behind you. There's a lot between growing up in Greece and VP of product at DoorDash. As you mentioned, I grew up in Greece, and that's kind of, like, an important part to this conversation. Mm-hmm.

Greece is kinda a community-based society, and I do think I was longing for that when I moved here. So I've been a builder and a tinkerer, I would say, for the most part that I can remember. When I got my first Mac, I remember dreaming about making it to the Silicon Valley and coming out here at some point in my life.

And lo and behold, you know, be careful of what you dream. You just might get it. Exactly. I found myself here in the United States around 15 years ago, and since then, I've had the incredible luck in working around people in environments that I consider to be amazing. So, um, I worked at Pinterest in the early days as we were building both our advertising platforms and our commerce platforms.

Had a short stint at Instagram in the last five years, which I would say have been the most formative for me in so many ways, have [00:04:00] been here at DoorDash. That just kinda takes us right into the thing I'm most curious about is it's great to say, "Okay You know, local commerce platform, we've moved beyond delivery.

It's another thing to really do it and act on it and make that happen. How has that come to fruition over the five years you've been there? What has that change looked like, and how do you drive from a tool that people use late at night to order food to something that is kind of supporting local commerce and sitting within a ecosystem that supports across all those functions?

It has been an incredible journey, and I do think that people like to think of these infliction points- Mm-hmm ... or these magical moments where somebody had a strategic insight. But the reality is it's been a very kind of like steady and incremental process. At the end of the day, why would somebody with small business think about your favorite restaurant in your neighborhood join DoorDash?

Originally, it was to get growth. It was to reach more customers, right? Like that they might not with their store, and eventually grow their business. So as I think about a lot of the products that we've launched along the way, kind of like our marketing stack or our advertising [00:05:00] business being one of them, they're inherently just amplifying that same core job to be done, if you may, for the customer, which has helped me grow, right?

And folks wanted over time more agency on that growth. But then I would say over time, as you start integrating and supporting these customers more and more, more things bubble up. Mm-hmm. Then that kind of like forks into what I would call helping me run my business, and that's where our reservation product or our online ordering product or our table management products come into play.

But at the end of the day, I don't think that there's been this magical moment. It's just been trying to get one percent better every day and being obsessed about our customer needs. How did you look at that platform, or what was the process to go and see the possibility for things like, I know there's a big play across franchise kind of coordination and franchise advertising opt-in.

But beyond that, there's also ways that it serves kind of the non-franchise, the smaller shops. How do you go about looking at that and driving that product [00:06:00] forward? I mean, Pinterest monetized very well. Were you able to kinda take lessons from there and bring them forward, or was it a completely new process of discovery 'cause it's such a new audience?

I generally try to push my teams to default back to first principles. I think that's the best way you can truly sit down with your customer, figure out what are the unique things that you can do for them, and build incredible products. One of my favorite quotes is from Frank Lloyd Wright, the famous architect, and I hope I get it right.

Like, it was something along the lines of, "Show me an expert, and I'll show you a man who stopped thinking." So I'm always wary of expertise, you know? Yeah. But at the end of the day, the thought process was look, like, we have a certain set of capabilities on the platform. Our customers have a certain set of needs, and let's go back to first principles and figure out how we can build great products for them.

And for me, there's a couple of things that set DoorDash apart as a marketing platform. One of them is, like, we are, I call it human-aligned, right? Most advertising platforms, you're there maybe to watch your favorite movie or TV show, or you're there to catch up with friends, and [00:07:00] the commercial message can interrupt what you're doing.

But when I think about any commercial message on DoorDash, it should amplify and help the job that the consumer, the end customer, is trying to do. The other one is, you know, in marketing speak, closed loop measurement, but in human speak is we drive sales at the end of the day on the platform, and that's where the business and who's advertising in this example and DoorDash is aligned.

That was, like, a very big thing for me early on, which is why we built the true CPA model where we, we get paid if we drive an order. And I think that tied our destiny, right? Because at the end of the day, for me to get paid, I need to drive value for the customer. Another one that is very top of mind is we are accountable.

It's one thing to drive a sale, it's another thing to actually deliver it. So, so I can't go to your restaurant and say, "Hey, Jeff, order it." I have to make sure you get the order. The food is warm, right? So there were a lot of things that I think made the platform unique and therefore were very [00:08:00] core in terms of how we thought about the products that we built.

Assuming rational actors on all sides, it has to end up being for continued investment, a kinda win, win, win. Yes. Where the restaurant is not going to hopefully keep plowing money into getting more in their potential customers' eyes, you know, if it's causing them to lose money on every order. Did that work out that way?

Did you find people were looking at it maybe with different goggles on of build growth now and it'll pay back later or something? Part of what makes this job so fun, and at the same time difficult, it's easy to strap an ad on anything and make some short money. Like, what is hard is to make sure you're balancing these constraints.

Human beings get what they wanna eat or shop in a timely manner that's not distractive, and at the same time, like, businesses achieve the outcomes that they want Balancing those constraints is the hard part. We're taking the long-term view, the 10, 20-year view, and we wanna do right by our customers and build a generational business here.

And it was partly principles, it was partly [00:09:00] aspiration, but it was also partly the fact that we are deep with these merchants and local businesses in many other ways. We charge commissions for the food that we deliver. We have a series of other products that they're using that are adjacent or incremental to advertising, and it was very important for us to do the right thing It is interesting because you look at so many ad platforms, and the great thing here is you spend a dollar, in theory, to get someone who is hungry right now.

I guess that's the difference at DoorDash. Yeah. Your users are hungry right now. No one's, like, browsing DoorDash for funsies. The majority of our customers open the app because they need something now, and it can be food. It can be that you started cooking and you're missing an ingredient- Mm. -or it could increasingly be that you need a reservation at your favorite restaurant or something else.

But there's this funny pattern where after people usually order what they want to order, they come back to the app a lot. Mm-hmm. And some of it, checking the map to see, like, "Where's my order?" Some of it is actually, like, browsing, and I think that's, like, a beautiful discovery moment where maybe you've ordered from that place.

What's the [00:10:00] wildest finding you've had or weirdest unexpected discovery that's been made while you've been rolling all this stuff out or, or while this has been evolving over time? The most obvious and craziest nugget was a big hypothesis that we had early on was that these mom-and-pop shops, these local entrepreneurs, would have an appetite to spend aggressively on advertising, which is not something you see, right?

Like, when you talk to them, they have promotion motions. You know, they have happy hour. They have a lunch menu or a lunch specials menu. I would say the most surprising thing for me has been that, the overwhelming reception and appetite for that and the curiosity that came along with it. It's not just a set it and forget it or, you know, it's now part of the DoorDash system or a black box.

I get phone calls every day about folks who wanna learn more about what we're doing, what's next, how we're thinking about it, and they come from all sorts of places. We have more than a million merchants on- Yeah. -the platform across kinda, like, the three logos, which are, like, DoorDash, Wolt, and Deliveroo in Europe with more than four hundred thousand [00:11:00] active advertisers.

So, so just, like, if you wrap your head around those numbers, the reality is that it's, uh, a little bit of everything. One thing that I've found very interesting is that I don't wanna call it managed, but there's a lot of appetite for automation. Like, what you generally hear from the entrepreneurs that you're talking about is there's a little bit of skepticism up front.

It's like, you know, "You're asking for more money. Tell me why, but how does this make sense for me and right here now and my family in many cases?" But once you get them over that hump- Mm-hmm. -and they trust you, they are okay with a machine or an algorithm running these programs, and these can be promotion and affordability programs for them or digital advertising.

There's a lot of appetite because most entrepreneurs that I talk to have, like, an incredible sixth sense about what they're good at and what they're not and what their business needs. If they could find a way to trust somebody to do the things that they're not good at so that they can focus on building food or opening a new location, they will do that.

There is probably [00:12:00] one topic that we end up talking about at all these events, and it, it's become to the point where we could just call it, "Let's come talk about AI and how it's affecting your team." The question people kinda keep landing on is, is there going to be product management in five years?

What's that look like? Off the hip, what do you think the future holds as models continue to get better and people, including us, talk about self-improving, self-healing software and, and stuff like that? The most common question I would get when I interview people would be like, you know, "Tell me about the DoorDash culture," or something related to the environment.

Now it's like, "What do you think is gonna happen with product and AI?" And it's like, "What do I know?" I think before we answer that question, right, like we have to go back to what is the role of a product manager from a s- principal standpoint. Mm-hmm. And from my perspective, the product manager is the representative of the customer in the room, in the room with engineering, with design, and a product manager is successful when their customer succeeds.

When I think about what AI, as it [00:13:00] stands today, and I don't have a crystal ball about the future, has been able to do, it's helped in two ways. I remember as a young product manager spending weekends and nights trying to join tables that should have never been joined, you know? Or like going back to meeting notes and try to project manage next steps and whatnot.

So a lot of that right now we can get more leverage and outsource to our agents or to, like, CoWork and other products that we all use. So I think that's, like, more time back, and I think that's kind of amazing, if you ask me. Mm-hmm. The second piece is a big role of the product manager is to translate those customer needs into products and features that then, in partnership with engineering teams, design teams, data teams, turn into great product.

And the beauty of mostly copilots right now or AI-written code is that code is becoming a lot cheaper. So now you're able to get a lot more product output from leaner teams, which I think in a crazy turn of events or a twist of fate, if you may, puts more pressure back to the product function [00:14:00] because you need more judgment.

And what I'm seeing on my teams is that we're running out of PRDs, so to say. Like, the number one thing everyone will want with anybody on my team would be, "I need more engineering." Now I'm hearing the inverse, which is like, "Hey, we're done. What should we build next? Then why?" So my hot take is that there's never been a better time to be a product manager because now you can spend a lot more time with customers and thinking about what you build, and if anything, the need is growing.

It's easy to build something now, easier than it's ever been. And I think the thing people forget is, cool, you cooked up a vibe-coded thing where engineering moved a lot faster to make actually production level thing, not even vibe coded, and they can. Our engineering team is moving several X faster than they ever have.

But it still comes down to everything you make still has to be maintained, and there is some cost to doing that. You have customers who have to use something, and they are only [00:15:00] going to absorb so much knowledge from you and so much of what you're putting out. And, and so if you're building 10X faster and shipping 10X faster, you have a high likelihood, it seems like, of a lot of that getting missed.

And it might not be the ones that you want to get missed. And even if you do it all right- You still have a high chance of them missing that stuff. So what you build and how you promote it and what you promote and what you choose to focus on, I think in a world where there's 10X more just crap flying at you- Yeah

is gonna be even more important. Could not agree more. Could not agree more. So what does that look like on your team now? How have you kind of worked with your team and, and coached people on your team to, to actually take advantage of this? My approach is, like, I've treated this internally, and I'm responsible for AI deployment internally at the company, not just on my team.

I approach it almost as I would with a product launch, which is, look, at the end of the day, AI is a tool, and the tool is only as useful as the job it helps you do. We've looked at what we need to do or what we do every day as product managers, and we're like, "Okay, what can we do 10X better because we now have AI?"

And then we go job by [00:16:00] job, and we make sure we have a repeatable system and playbook across the company. So I'll give you an example, right? Historically, the workflow, unless, like, you had a very technical PM, was like you would partner with an analyst. You would kind of send them off to do kind of like this and this and come back with an insight.

Now, with AI, as long as your data infrastructure is good and your tables are well labeled, you're able to very conversationally in any agent, right, like retrieve, visualize, synthesize the information that you need. That's, like, a great job, right? We worked on that, and now I would say everybody at the team is data empowered and uses this tool that we built internally daily.

A lot of that project management and accountability is done by agents where, you know, where teams are in Slack groups or other, you know, tools that they communicate in. They get the updates. They get the nudges, right? Nobody's taking notes. I think that's where you get to great impact, which is, like, take jobs one by one- And quote unquote outsource them or delegate them to your agent or an AI workflow.

And that's how over time you get incredibly more leverage. [00:17:00] Looking at all the learnings you've had in advertising across DoorDash now, from a product standpoint and, and decision-making standpoint, do you generally lean into there's a whole new market, there's a whole new opportunity here, we can go develop something brand new from the ground up, a new revenue opportunity maybe?

Or is it, "Hey, there's really good traction here, let's double down and continue down this path." Do you have a predilection there? It's hard to not answer both, right? And I think that's the art. Like, it's easy to get carried away by the new sexy thing. It's also very easy to just get addicted on those incremental benefits that you get from optimizing, right?

Like, and optimizing again, because like one percent on a really big basis can be a lot of money, can be a lot of growth. For me, and that's part of like the art and maybe a little bit of science of product leadership, is trying to figure out that balance, which is especially in a more established environment like DoorDash, it's how can you make sure you're continuously improving the [00:18:00] experience for any and all customers, but at the same time, you're taking some wild new swings that are gonna be your S-curves ahead, right?

Like, how are you disrupting yourself? And I wish I could tell you I've, I've nailed this. I'm still trying to figure it out. Is there a way to blend the faster engineering capability with the kinda taste in picking right from product to give us the ability to get a little bit further down that decision path before we have to veer one way or the other?

Experimentation has been like such a powerful tool in, in tech the last ten to fifteen years. And I do think that thanks to AI, right, like, and code being cheaper, we're now in this unique position where you can explore, exploit a little bit faster and try different things, but you can't skip the taste part.

Yeah. You can't just say generate like hundreds of hypotheses and test them. I think we are seeing a resurgence of real human, real thought. Vasilie, it's been great having you on, man. I appreciate you coming on, talking product, talking through what you've learned across all this and how you're looking at problems I think a lot of people are, are sharing.

Thank you so much for [00:19:00] having me, and thank you for everything that you do for the community. Really appreciate you, Jeff.