Q3 2025 Earnings Call Transcript

Ladies and gentlemen, thank you for standing by. My name is Desiray, and I will be your conference operator today. At this time, I would like to welcome everyone to the DoorDash Third Quarter 2025 Earnings Call. [Operator Instructions] I would now like to turn the conference over to Mr. Weston Twigg. You may begin.

All right. Thanks, Desiray. Good afternoon, everyone, and thanks for joining us for our Q3 2025 earnings call. I'm pleased to be joined today by Co-Founder, Chair and CEO, Tony Xu; and CFO, Ravi Inukonda.

We'll be making forward-looking statements during today's call, including without limitation, our expectations for our business, financial position, operating performance and profitability, our guidance, strategies, capital allocation and investment approach, our plans and expectations regarding the integration and benefits from our acquisitions, our expectations regarding new product and service initiatives, including our autonomous delivery platform as well as expectations regarding platform safety, our global technology platform and the broader economic environment. Forward-looking statements are subject to risks and uncertainties that could cause actual results to differ materially from those described.

Many of these uncertainties are described in our SEC filings, including our most recent Form 10-K and 10-Qs. You should not rely on our forward-looking statements as predictions of future events or performance. We disclaim any obligation to update any forward-looking statements, except as required by law. During this call, we will discuss certain non-GAAP financial measures.

Information regarding our non-GAAP financial measures, including a reconciliation of such non-GAAP measures to the most directly comparable GAAP financial measures may be found in our earnings release, which is available on our Investor Relations website at ir.doordash.com. These non-GAAP measures should be considered in addition to our GAAP results and are not intended to be a substitute for our GAAP results. Finally, this call is being audio webcasted on our Investor Relations website. An audio replay of the call will be available on our website shortly after the call ends.

Operator, I'll pass it back to you, and we can take our first question.

[Operator Instructions] And our first question comes from the line of Deepak Mathivanan with Cantor Fitzgerald.

Two-part question on the several hundred millions of incremental investments for 2026. First, how much of this is towards tech platform initiatives versus perhaps more direct product and expansion efforts that tend to have a very defined near-term payback? And then the second part is can you expand on the tech platform efforts. Are you sort of essentially rewriting the tech for AI development? Or is it more about integrating AI tools with additional token expenses into the service?

And in addition to sort of opportunity for accelerating the product development, do you also see any potential for cost savings from these efforts over time?

Deepak, it's Tony. Yes, I can maybe start on both questions, and Ravi, feel free to add. Yes, look, on the first question of where are we spending, we're primarily investing in 3 areas and kind of the second part of your question kind of feeds into the first part. So the first major area is building a new global tech platform.

And so this is an effort that's been underway for a couple of years now, but it's coming to a head where we're actually making the majority of those investments in 2026. And what is its purpose? Well, there are several reasons of what we're trying to accomplish. The first thing, what we're trying to accomplish, is we're trying to build a single global tech stack, where when we launched one experiment, the experiments that actually make it to our customers can be shipped at the same time across all of our markets and all of our audiences. Today, that doesn't work that way as we have 3 companies at DoorDash on the restaurant delivery, our marketplace front with Deliveroo, Wolt and DoorDash.

In the new world, we'll actually be able to have 1 feature go live to all audiences; whereas today, that 1 feature would get to have -- would pretty much have to get shipped 3 times, which is very inefficient in how we do that. Another goal of the tech platform is to, like you said, in your setup, to make it AI native. And so there will be lots of tooling there where -- I may have mentioned this in the previous call before, where if we were starting a company over again today, I think we would write software pretty differently from how we used to do it. And so there's a lot of work in order to architect how we set up the architecture, so we can manage both agent workflows as well as how you would deploy software, test software, write software, what the role of the engineer is in that new paradigm.

All of that is getting constructed as part of this work. Yes, and I do actually think that, at the tail end of this work, you'll see, on a go-forward basis, not only will we ship faster and ship improvements across the board globally. We'll actually be more efficient, and we'll have freed up engineering capacity to do a lot more work. And so that will allow us to not only have a better cost structure but really just be able to do a better job in solving the next problem for customers.

So the first area of spend is in building our tech platform. The second area is investing in new products. And so we announced probably, well, several launches in our dashboard product events, which happened at the end of September. And we're really excited about them.

A lot of times, when you're building a company, what you're really doing is you're starting with lots of experiments. Some of those experiments make it into products. Some of those products then can be graduated into commercialization where you're trying to test whether or not you have a good business. And then some of those candidates then ultimately yield big businesses that generate the cash flows to allow you to invest in the next set of products and experiments.

And I think we're really fortunate at DoorDash in a couple of ways. One, in the area of local commerce, there's just a lot of different problems. And one is -- and the other way in which we're lucky is that a lot of the experiments, many of these have been running for years, are now coming ready for more investments. And so we're very excited to be investing behind them.

We announced, for example, a lot of work on in-store and building several products there with going out, reservations and our CRM platform behind SevenRooms. We talked about DoorDash Dot, which is the customized, purpose-built for delivery autonomous vehicle, the first in the world to drive on the road, sidewalk and bike lanes in order to make that happen. We talked about DashMart Fulfillment Services, where we're creating an ability for any retailer to offer same hour or same-day delivery with near-perfect accuracy. So these are some examples of the new products that we're talking about.

But if you think about it in each case, we're running the business exactly as we always have, where the goal is we want to make sure that we always can solve the most number of problems for our customers in the highest quality ways, and we manage our projects carefully to milestones. And as they deliver upon each milestone, we grow them into businesses, and we continue to invest behind them. I think our track record in investing in the areas that we currently have operating, whether it's U.S. restaurants, U.S. new verticals, the international business, our commerce platform, our ads business, have suggested that we've had some success in repeating this playbook, and we're doing this now for future growth.

Yes, Deepak, I mean just to add to what Tony talked about, right, look, our core business is continuing to do really well. I mean you're seeing that come through in terms of the numbers where, if you're thinking about it from a growth perspective, growth accelerated for the fourth straight quarter. Overall unit economics are improving across the business as well as the profit dollars continue to increase. This is giving us the ability to reinvest back in the business.

If you think about our operating philosophy, when the GOV is coming in ahead of expectations, when the unit economics are improving, our philosophy has always been to reinvest back in the business. If you're looking at the results today, it's a combination of these decisions that we've taken over the course of the last decade. That philosophy and how we are operating is not changing. Look, we're very disciplined in terms of how we operate.

We're thinking in terms of IRR. All of the investments that we talked about, we think they're going to extend the duration of growth as well as drive strong IRRs for us. And if you're thinking about '26, what I would say is that the EBITDA margin for the existing business, including the investments, think of that as the overall business excluding Deliveroo, I would expect those margins to be up slightly compared to 2025. Hopefully, that should give you a sense of how we're thinking about investing. I mean net-net, look, the business is growing.

The business is growing exceedingly -- exceeding our expectations. And our goal is to continue to reinvest, and we are very excited about the investments that we're making.

Our next question comes from the line of Shweta Khajuria with Wolfe Research.

Could I at least try 2? First one is also on investment. Could you please talk about where you're planning to invest as it relates to Deliveroo, what your goals are for the first year in terms of order of business and then -- and strategic focus areas? And then the second one is on automation.

Could you please talk about where you are in terms of expanding your robots and your third-party partnerships? And how do you think about scaling it and deploying the opportunities over the next year or maybe 1 to 3 years?

Yes, Shweta, I'll start and Ravi, feel free to add. First question is about investing in Deliveroo. This is kind of similar to what I said about our general investing philosophy and kind of what Ravi said about why we feel like now is the right time to continue to invest for the long run, which is whenever we see -- it starts with making sure that we can build the best-in-class product experience measured in terms of the retention, the frequency of use and the engagement from the audiences. We think that we found a business actually in better shape than we had expected with Deliveroo, and we think that there's actually a lot of opportunity to add to the foundation that they already have and make continued product improvements so that it will achieve the best-in-class metrics along those dimensions.

I think if we can do that and at the same time, improve the unit economics, I think that that's what will allow us to generate the greatest service for our audiences and also the greatest long-term business and returns for all of our shareholders. And so the first order of business really is making sure that we continue to invest in the product, make those improvements. And then, of course, over time, you'll see cost efficiencies come out of the business because we'll be combining 2 European teams on the same continent. But I think the first order of business continues to be adding to the strong foundation that the product has and making sure that we can make it best in class given the learnings that we've had with Wolt and DoorDash and then grow from there.

On the second question around autonomy, the way that we're thinking about this is really pragmatic. And I think the most important thing is understanding that the vision for autonomy is really going to be a multimodal world where you're going to see different fulfillment methods. Some of those fulfillment methods will happen with Dashers. Some of them will happen with vehicles on land.

Some of them will happen with vehicles in the air. Some of them will happen with vehicles built by DoorDash. Some of them will be filled by vehicles built in partnerships. And really, what's important is DoorDash has the luxury to create our own autonomous delivery platform where, depending on whatever the use case is or whatever the customer need is, we can solve it with the highest quality, the lowest cost and the best service, and that's really the goal.

And so we'll be fairly pragmatic on how we do this. So a lot of it, we'll be testing within one market, getting that right, figuring out the go-to-market motion because there's many parts you kind of have to get right, everything from the manufacturing when it comes to making these products in-house to partnerships, to integrating the hardware and the software, to making sure you get the repair and the teleoperations correct, to making sure that you work well with city governments. There's a lot of things you kind of have to get right, and so for us, '26 will be the year where we're ready to commercialize some of these efforts.

But I think this is going to take a while. This is not something that's going to happen overnight. It does require making investments upfront because sometimes these decisions are required upfront, and you don't really get the product until later. But for us, we're pretty excited about what we see as the potential.

We also are very, very excited about this autonomous delivery platform, where we're going to be able to inject whatever the right fulfillment method is in order to give the best service.

And Shweta, your point around Deliveroo, look, I mean, we're very excited with the partnership with Deliveroo. Look, I'll wind us back to the deal thesis that we have, right? This gives us the ability for us to operate in really large attractive markets where we could deploy our operational playbook, our product playbook to drive operational rigor. If you think about it, the focus for us is -- what we're learning is there's an opportunity for us to continue to improve product.

There's an opportunity for us to improve the consumer experience. To your point around the focus, that's what we are focused on. Why is that important is because it drives scale by improving retention and order frequency, which will drive more gross profit dollars. That's honestly how we operate DoorDash.

That's honestly how we operate Wolt. Deliveroo is going to be no different for us. What we see in the business is, from an EBITDA perspective, look, we are very comfortable with the profit generation of the business. It is in line with what we underwrote when we did the deal, which assumes some level of investment.

Now what you're seeing in the business is the growth is exceeding our expectations. The business is growing double digits, which is giving us confidence to invest back in the business. The focus for us continues to be investing behind the team, investing behind product, which will ultimately drive your long-term free cash flow generation in that business.

Next question comes from the line of Ross Sandler with Barclays.

Glad to be back on the call. I guess just following up on just the broader consolidation that we've seen from you guys and from Prosus acquiring JET, how do you see the overall landscape in Europe evolving next year on the back of all of this? And Ravi, it sounds like you're investing in Deliveroo to grow a little bit faster. I assume that the $200 million of EBITDA is reflecting that investment and the platform consolidation investment that Tony just talked about a few questions ago would be kind of a separate thing, not included in reducing Deliveroo's run rate from what it was before to this $200 million based on some kind of like allocation of that platform investment.

Just any clarity on that, that minutia would be helpful.

Look, on the first question on the European landscape, I think we have a great opportunity to be the leading local commerce platform there. And I think a lot of the confidence of what inspired us to pursue the Deliveroo acquisition really came a couple of years ago after gaining confidence in working with Wolt. The first test for us was whether or not we can actually take some of the lessons that we've learned at DoorDash and also combine them with the lessons that Wolt has learned in building their markets, to see if we can create the best-in-class product. And I think when you look at our growth rates, which continue to exceed those of our peers as well as just the retention and frequency levels and how they're progressing nicely as well as the unit economics improving at the same instance to all-time highs, we gained further confidence that we could kind of take on the next project.

And that really is kind of how Deliveroo came into the puzzle. So when you add up our presence in Europe, you're talking about presence in over 20 countries and with the strongest position in the cities with the biggest profit pools, and so I think there's a lot of strong foundation to build from. And so in general, what I see is can we solve the most local commerce products in Europe the way that we're trying to do that in the States and in other countries. And so far, as I mentioned, we've been fortunate where some of the experiments that we've been running for a while now are now coming into fruition into becoming real products, and we're starting to invest behind those products, which is some of the investment commentary that we outlined in the letter.

And that will translate over to Europe as well.

Ross, to the second point, yes, I mean, the $200 million, think of that as the contribution from Deliveroo to the overall EBITDA. And think of that as the investments that I talked about earlier, right? It's investments that we're making behind product. It is investments in selection, quality as well as people. You should think of that as the contribution of Deliveroo EBITDA.

Next question comes from the line of Josh Beck with Raymond James.

I wanted to kind of go back to the tech platform that you've been building in the background. I'm curious kind of what you've learned thus far and kind of why this moment in time was the right moment to inflect upwards. There's a lot of external changes in the AI stack. The international scale of the business is obviously very different.

Robotics is having breakthrough. So I'm just kind of curious if there was maybe a smaller list of items that drove the step up. And then with respect to the integration, I think in some cases, they can be messy. You have obviously a lot of consumer-merchant-Dasher ecosystem.

How do you minimize the disruption and kind of keep the strength? It sounds like the Roo business maybe is kind of in the double-digit range, so it seems to be in a pretty good spot. How do you maintain that throughout the integration?

Yes. I can take that. I guess with respect to the tech platform, I think this was something that always was in our heads when we're thinking about the acquisition of some of the companies that we're talking about, right? I mean, like, by and large, these companies perform the same service in each of the geographies.

Now there are a lot of, obviously, local differences, but by and large, the service themselves are the same. And I think that's one of the things that screams that we should be building a single-type platform to actually make sure that all of the products are under the same data models, the same architecture, the same UX. But when you're doing that, especially when you have like a change as big as AI happening externally, you obviously have to take that into consideration. So I think some of these things kind of came together in the '24 time period, where it was exceedingly obvious like what the right thing to do is, and then now you just have to go and do it.

And that's kind of the work that we've started here in '25. It will really take shape in '26, which is, again, a large part of the investments. Obviously, those investments will come off, but like -- but that's -- the timing of that really makes sense for us. On the integration of Roo, I mean, you're absolutely right. I mean we obviously -- one of the reasons why we're incurring extra cost is because we are making extra tech investments to make sure that Roo can continue to perform well on its own even as we're building the single tech stack.

And so as a result, when you have these added investments, that's what explains it, and -- but that's also how you protect the experience and the service that the Roo customers get to see even as you're building this new foundation.

Next question comes from the line of Jason Helfstein with Oppenheimer.

Just one and a follow-up. How should we think about advertising broadly? I mean it obviously comes in at a high incremental margin. There's flow-through versus reinvestment.

So just how are you thinking about how we should think about how that flows through? And then second, there's been some questions, a report floating around about kind of documented workers and the government kind of cracking down undocumented workers and certain reports like talked about what percentage. Just any color you can have about how you manage making sure that you don't have undocumented workers on the platform and how you manage through that? And any exposure there?

Sure. I can take both of those questions, Jason, and feel free to add in here, Ravi. Look, on ads, I mean, we're pretty excited about the ads business, right? I mean it's -- it was the fastest business, ads business in history to get to $1 billion of annualized revenues.

And it's also a business where we have extra budgets wanting to spend more on the Dash platform than we kind of give ad space to. And that's because, when I think about the ads business, I think the reminder I always give to our teams and I would say again here, is that you kind of have to solve for all of your audiences. On the one hand, you obviously have to maximize the return on ad spend for advertisers. On the other hand, you have to make sure that you are not degrading the consumer experience.

And that's hard to do. There are real conflicts and real tensions and real trade-offs and negative consequences, I think, if you get that incorrect, and that's why I've always believed the right order to sequence things is to first build a very healthy marketplace, and then the monetization opportunities including ads will follow. And they have. I mean, look at our business.

I mean we've continued to grow faster at bigger scale for many quarters in a row now, and you're seeing the bottom line contribution margins and EBITDA margins grow in the same instance. And so I think all of that is suggesting that, I think, we have the right trade-off, and we just have to make sure that we have the discipline to maintain it. On the second question, I mean, Dasher supply has never been healthier. It's always been something that's obviously very top of mind for us, which is making sure that we get correct all of the Dasher authenticity pieces.

There's a lot of work that goes into it. We -- something that we've been investing into for many years now and will continue to. We've seen no challenges to our Dasher funnel or supply irrespective of whatever reports there might be out there, and it continues to be something that we've upheld the highest standards and something where it's -- we've always done it the right way, I guess.

And Jason, just on your point on the ad business, right, just to clarify, I mean, I think the ads business is growing. It's growing quite nicely. We don't differentiate that from, okay, are we thinking about it just purely from a flow-through perspective because, for us, every dollar that we generate, our goal is to reinvest back in the business at healthy rates. That applies for improvements in unit economics or applies for dollars that we generate from ads.

We think of efficiencies that we could generate so that we can put that back in the business. But net-net, ads business is growing, and it's growing quite nicely.

Next question comes from the line of Michael Morton with MoffettNathanson.

I guess maybe one for Ravi to start. If I'm just doing like quick math here, looking at your guide and kind of what it means for the core business, it seems like this incremental step-up cost could maybe be like $100 million a quarter. You feel free to correct me on that. And then if -- Tony highlighted the buckets where the spending is going, and sorry, it's a stupid question, but from the outside looking in, just understanding maybe like where the platform development spend, like how that hockey sticks so much if it's something that's been going on for several years.

And then what I found interesting was the DashMart Fulfillment Services comment he made. And I was just looking for any more details if you're really stepping on the gas there and if this is like maybe part of a partnership with some of like the, I'd just say, large AI platforms, where maybe you work local commerce more into these e-commerce searches. And is there more cost involved as you build out DashLink? Anything there would be really helpful.

Yes, Mike, I'll take the first one. I think to your point around -- I mean, look, we are still early in terms of how we're thinking about planning. We'll give more precision as we give quarterly guidance. What I would say is if you're thinking about 2026, to the earlier point, what I made to Deepak, is I would think of margin for the existing business, including the investment areas, think of that as the existing business plus investment, excluding Deliveroo.

I would expect the margins to be up slightly compared to 2025. Hopefully, that should give you a sense of how we're thinking about margins as well as investment into '26.

Yes, Mike, I guess what I'd add to some of your other questions is, first, the reason why the tech investments go up in '26 is because that's when they're actually happening, right? Like the way -- if you think about how would you build new software, the first thing you would do is you first have to architect it, right? And so you're not actually really doing much actual coding. But like, for instance -- but once you're ready to code, what do you have to do?

Well, okay, well, one thing you have to do is you have to spin up some cloud instances. And if you're going to maintain multiple stacks as you're building a new one, you're going to incur extra kind of temporary cloud instances, right? And so that's one of the reasons why the costs go up. And that's -- so it's just basically like when you're actually ready to deploy the software and actually get everything onto the same tech stack, that's what's adding to the kind of temporal costs.

With respect to DashMart Fulfillment Services, it's less about working with AI companies, although -- and more on how we work with our existing retail partnerships, right? So what is Dasher Fulfillment Services? Well, the goal is we want to -- one of the challenges in delivering from just third-party stores is that not every store and most stores actually, I would argue, do not know their inventory, and this is for a whole host of reasons. And even though we believe we have leading accuracy and the quality of delivery when it comes to any category, including those outside of the restaurant category, it's still not good enough because it's not perfect, right?

And therefore, why would a customer pay a premium to get not what they ordered or to get a substitute of what they ordered? But if we could manage the inventory systems ourselves and actually run the fulfillment setup end to end from the warehousing to the inventory to the fulfillment, we get almost near-perfect accuracy. Okay. So how do you translate that and do it for every retailer, especially retailers who may not have the density of stores or the coverage in the country so that they can offer something like same-hour or same-day delivery?

Well, I think what I would argue is you can literally turn every physical retailer into a omni-channel player. Now that's going to take time, and this is one where it takes time to set up supply chains with retailers, to test the right markets, so on and so forth. But what we're very excited about is that the quality is best in class. Now what we're doing is we're adding selection that's never been made available to customers before and bringing it close to where they are so that these retailers can offer same-day delivery, and that combination is now happening.

Next question comes from the line of Andrew Boone with Citizens.

I wanted to ask about new verticals. You guys talked about the fact that unit economics are still negative. Can you talk to us about the path and maybe the visibility that you have to breakeven? What are the key kind of operational things do you guys need to do to get there?

And then in terms of just U.S. kind of growth overall, now gross adds were higher year-to-date versus 2024. Can you guys just talk about the opportunity of where you guys are finding new users? I know we've talked about this on past calls. But can you just revisit what pockets you guys unlock in there?

Yes, Andrew, let me start. Look, I mean, talk about the overall performance of the new vertical business, right? Like look, new verticals had a really strong quarter. The business is growing really fast.

We're the fastest growing. We believe we are leaders in terms of order volume share ahead of our expectations. Remember, if you recall, Q4 of last year, we talked about the fact that 1/4 of our users order from new verticals. That number has continued to grow quite nicely.

Miles have increased. Order frequency is increasing. The overall basket size continues to increase, which just tells us that consumers like the product. The usage of the product then continue to go up.

On the unit economics front, just to be clear, I mean, unit economics continue to improve. They've improved sequentially Q-on-Q as well as year-on-year. We're very comfortable where the unit economics are. We are comfortable on what it needs to get to, to be breakeven.

Look, a lot of that is going to come from scale as well as continued improvements, whether it's quality of the product that we continue to improve. What we're focused on right now is scaling the business. We think there is an opportunity for us to continue to improve the product. As long as we continue to improve the product, this is going to be a large business for us, which will drive more free cash flow generation in the future.

Yes. And then your second question with respect to the U.S. strength, I mean, you're absolutely right. I mean we're thrilled with the performance of the U.S. marketplace. I mean 4 quarters in a row of increasing strength on a bigger base, especially at the scale that we're talking about certainly is something that we're very proud of.

There's no one thing. I mean a lot of the performance that you're seeing today really were the result of actions probably initiated 3 years ago, where it's this continued maniacal focus on improving the inputs of the experience. How do we improve the selection, the quality, the affordability, the service levels? And when you do that -- and that's because the most difficult thing that we always compete against are increasing consumer expectation.

You talked in your question about why are we seeing increasing monthly active user penetrations or perhaps why are we ahead of schedule on some of these forecasts. It's because we're delighting each cohort of customers, right? And each new cohort of customers obviously have a higher expectation bar than the previous cohorts. And if you can keep delighting the new customer cohorts where we are still the leader in acquiring new customers, whether it's in the restaurant category, in the grocery category, in any retail category, well, then you're also delighting all of your existing customers.

And so you kind of have this bow that lifts all of your cohorts, and that's because of product improvements. And -- but a lot of the stuff started probably 3 years ago, and now we're making investments into things that we -- hopefully will have an impact 3 years from today.

And Andrew, when you look at the underlying cohorts, right, the demand on the underlying cohorts continues to be very strong. Both MAUs are growing. Order frequency is growing. Even subscription, both in the U.S. as well as international, had a record quarter in terms of DashPass subscribers as well as Wolt+ subscribers.

What we see in the business is even existing cohorts, cohorts that are quite old, they're also continuing to engage with us even more. All of this goes back to, look, I mean, improvements in product, improvement in selection as well as quality that we've been continuing to work on not just in the last quarter, right, over the last several years.

Next question comes from the line of Nikhil Devnani with Bernstein.

I had a clarification on the investment commentary. Is the bulk of the spend fixed cost investment that you get leverage on as you compound the top line? Or is it a step-up in variable costs as well? And then on new verticals beyond grocery and convenience, which categories of the retail or local commerce opportunity, do you feel, are showing the most promise from a demand perspective that becomes the next big category for DoorDash going forward?

Yes, Nikhil, I'll take the first one. On the spend and the investment areas, look -- I mean, our goal is we are spending up and down the P&L. Some of that is going to come through in terms of cost of sales. Some of that is going to sit within sales and marketing as well as some of that is going to sit within R&D and OpEx.

Look, the areas that we're investing behind are we're trying to increase and scale our autonomous program, obviously, doing it in a very disciplined manner. We're growing our software business. Both our digital ordering as well as our SevenRooms business falls under that. They're starting to generate revenue.

That's going to continue to increase. And finally, some of the tech stack work that we are doing, which you talked about, that's going to hit across the P&L. So you should think of that as, a, some of that is going to drive leverage as we go through as we scale the business.

Nikhil, on your second question with respect to which categories are we seeing growth, we're actually seeing growth quite a lot across a lot of categories outside of -- multiple categories outside of grocery, convenience and alcohol. And it tends to be somewhat probably what you may expect that comes with time of the year. For example, pets is a category that's kind of a all-season category. As we head towards the holidays here in the U.S., you see categories like electronics really spike in terms of the gifting use case.

We've seen growth in health and beauty. Home improvement has been a very big surprise to us, seeing growth where we're delivering, believe it or not, thousands of pounds of mulch per day. I guess, that more happen maybe in the summer time period. Again, so some of this has to do with when customers need different things.

But I think what's really interesting, even without just looking at the performance of these categories, is just actually looking at where things are going, like the trend and the trajectory and the input metrics such as the searches for different types of products. And we're effectively becoming the everything inside your city store, and as a result, we're seeing growth across the board. And so sometimes, some of these new customers are coming in -- we talked a little bit earlier about MAU growth or monthly active user growth. They're coming in for the first time outside of the restaurant category.

And then what you see happening is they'll then shop in the restaurants and then back and forth. And this is what we've always believed could be the case, where if you start with the highest frequency category of restaurant food, which is where we are the leader, you just get the most shots on goal. And then now we're also the leader in terms of transactions outside of restaurants, both in grocery, convenience as well as the categories outside of that. Then you really get this multishopping behavior across the board.

And if you can do that the most number of times, which I believe is what we're on track to do, you'll create the most valuable membership program, which is DashPass.

Next question comes from the line of Lloyd Walmsley with Mizuho.

I wanted to go back to the DashMart Fulfillment Services and just better understand the plan there in terms of integrating all this 3P inventory, especially on the grocery side. Is this going to require like a lot of build-out of new facilities? Or do you just sort of take control, pick and pack inside some of the partner facilities? And then as we think about the time line of that, and it seems like this could be a really big and attractive area of investment, is that something -- is this something where '26 is sort of continuing to experiment on this and maybe you scale it more in '27?

Just like anything more you can help us understand like how that will work and the time line around that would be great.

Yes. Lloyd, yes, on DashMart Fulfillment Services, we're obviously super excited, and we share your optimism that this could be a very attractive area of investment. I think the short version of this, it really depends on the retailer we work with, right? Some retailers have different goals.

For example, some retailers may want to launch new geographies. Some other retailers may want to densify existing geographies. Other retailers may want to find attractive uses for less productive existing square footage inside their existing stores. I mean there's a whole host of different goals, and this is why the question is a little bit hard to answer.

There is no one-size-fits-all solution, I guess, is what I'm trying to say. And so as a result, I think what's most important is you have to build different capabilities and be flexible. What we're stubborn on is we're stubborn on building the best possible experience where you bring in selection. That's never been made available to customers before, right?

If you think about it, there's -- roughly speaking, in cities, there's like tens of millions of items out there, and today, only a single-digit percentage of them are being delivered through DoorDash. And if you compare that to -- but DoorDash almost already carries the most amount of inventory for same-hour delivery. I just think there's a massive headway and runway in front of us in terms of how much more of that inventory we can bring and fulfill on behalf of customers and bring everything inside the city. Now this though is not our inventory, right?

This is the key challenge. It's a retailer's inventory and the challenge is, well, one, like does the inventory exist. And that's one of the reasons why we're building DashMart Fulfillment Centers -- Services because sometimes the inventory can exist and it happens to be inside stores, but other times, it doesn't exist or it's missing or it's in delay somewhere in the first mile or the middle mile. And so it's complicated.

But the simple thing that we want to give customers is we want to give them selection that's never been made available before to them, done same day or same hour. We're bringing a new e-commerce capability to a lot of physical retailers as a result on the retail side. But the answer to your question is a little tricky because each retailer may have their own goals, and so there may not be one answer that, that is the setup. And that's why we have a handful of these retailers that we're working with, and we're customizing the solution that makes the most sense for them.

And of course, we're making sure that's the best-in-class experience for consumers.

Next question comes from the line of Youssef Squali with Truist Securities.

Tony, one subcategory we did talk about is perishables. I wanted to just pick your brain on how you think the entry of Amazon in that space is likely to kind of impact you guys. Not even sure how big perishables is to you. I'm assuming it's small again, but maybe you can help clarify that.

And just how are you guys kind of positioned to kind of defend your turf? And then maybe just comment on the change in the guards at least in some cities like New York after the win of the Democrats last night, Mamdani, and how that potentially could impact Dashers pay, eventually Dashers organizing and just really not just in New York City but in other big cities as well?

Yes, I'll take those. Look, on the first question, look, I guess 2 sentences have always been true for the history of DoorDash, which is, one, it's always been a competitive space no matter which space we're talking about; and number two, DoorDash just continue to grow and I guess now even grow faster at bigger scale. And so I guess, like you may wonder how is that possible. How can you square away some of these things?

Well, I would say a few things. I would say the first thing is the market is still very not penetrated, right? If you look at something like grocery penetration, it is so low relative to something like restaurant delivery. And the reason, again, has to do with this inventory fulfillment challenge that I've talked about probably for several years now.

Actually, when we launched DashMart in 2021, this was the original vision for DashMart Fulfillment Services, which I'm glad that we're now making come alive. But it's can we get you exactly what you ordered, and if you can't get customers exactly what they ordered, guess what, they're not going to order. And that's really the challenge, I think, when it comes to the grocery delivery space or the perishable delivery space. With respect to retailers and folks coming in, well, I think at the end of the day, it's really about consumer choice, right?

I mean consumers -- DoorDash was created so that we want to give and create a world where consumers can choose from any retailer of their desire. And we believe that maximal choice versus just choosing from 1 or 2 retailers is something that is not only both good for all audiences, but it's great for cities. I mean that's the whole point of why DoorDash exists, to connect every local business to every local consumer. And if we can do that, we're going to grow the GDP of those cities and create more jobs for everybody and make better neighborhoods and all the good stuff that you'd want to see inside the city that you live in.

I guess that leads me to the second question, which is around what's happening with some of the different cities and kind of recent elections and stuff. My position on this has always been that governments and businesses should always work together. And that's what produces the best possible outcomes for all constituents. And I think this, again, is where several sentences can be true at the same time, even though they sometimes come into tension, I believe that all audiences deserve to be treated fairly.

For example, one of the hot topics in a lot of these coastal cities today, including the city where I live, San Francisco, is affordability. DoorDash has -- is probably the most aggressive on making sure that we are the most affordable platform. This is something that we've been doing since our history, and most recently, we're -- we took the lead, for example, on making sure the SNAP benefits could continue for those who needed them the most. When it comes to Dasher pay and Dasher protection, that's something -- we were the first platform in 2019 in this country, in the U.S. to offer occupational accident insurance to Dashers without their asking us to.

And it's something that we've always believed in. And I think that what I found to be most productive is finding commonsensical solutions and helping politicians create commonsensical policies so that they actually get what they want. I think what sometimes unfortunately gets in the way is when ideologies or biases come without even evaluating the facts or what audiences want. And I think that so long as we put what audiences want up there, I think that businesses and governments can coexist.

I believe that businesses should be allowed to be for-profit companies, and I believe that audiences deserved to be served.

Next question comes from the line of Justin Post with Bank of America.

I don't think you've really had a chance to outline synergies with Deliveroo on the -- maybe on the order and the top line side revenues, but I would love for you to talk about that if you can. And then maybe talk a little bit about the take rate differences in the accounting. That would be helpful.

Sure, Justin, I'll take both of those. Look, I mean, the focus for us -- like I said on the earlier question about Deliveroo, right, the focus for us is always, on day 1, is to continue to improve the product. It's to continue to improve the consumer experience. Why is that important is because that drives scale and the combination of scale and improvement in unit economics drives more gross profit dollars, right?

For us, the philosophy is to improve the product, to improve the overall gross profit dollars. That's what we are focused on from a day 1 perspective. Look, I mean, obviously, part of the deal thesis was to have cost synergies. We operate a global platform.

We think there's going to be synergies largely from scale as well as cost redundancy. Some of that is going to take some time. We are excited about, a, the partnership as well as what we're seeing in the business. To your second point around some of the accounting differences, the way I would think about it is, from an EBITDA perspective, you should assume roughly an USD 8 million to USD 10 million impact or an expense to EBITDA as you're thinking about modeling going from Roo definition of EBITDA to the DoorDash definition of EBITDA.

Next question comes from the line of Lee Horowitz with Deutsche Bank.

I guess going back to investments. I guess, how should we be thinking about the payback period? I mean a lot of talk about taking time, past dependency and spinning up an environment. So it sounds like payback periods are perhaps getting extended relative to your typical investment plans.

Is there anything in this new bucket of investments that is perhaps more of the traditional quicker payback periods that is part of your typical playbook? And then maybe relatedly, retail obviously sounds like a big and compelling greenfield opportunity for you guys. Any way to contextualize how much of this new investment plan may be specifically targeted at this vertical and driving faster growth outcomes there?

Yes. I can take both, Lee, and feel free, Ravi to add in here. On the payback period, nothing has changed. Our bar for payback period is still the same, and that's true whether we're taking on something in a completely new domain like DashMart Fulfillment Services or autonomy or our tech stack, for example.

It's just that we're taking on more projects now, right? And like I said, it's very hard to kind of forecast the progression of which experiments you ran many years ago to now which experiments have actually now earned their privilege of getting more investment. And so it's just more projects have found product market fit in our portfolio that I believe will achieve our investment philosophy, which is to maximize long-term free cash flow per share. So the payback period and kind of the methodology in which how we think about capital allocation has not changed.

On retail, I think it's probably similar to maybe a previous question about it. We are seeing a lot of growth in retail right now, and I think it's kind of similar to where -- but from a product perspective, I think we, DoorDash, are kind of in retail today where we, DoorDash, were in grocery maybe in 2021 or something like that or the tail end of 2020. So we're just very early on the actual product experience itself, and that's why it's super encouraging. It actually means that the real reason why we think there's a massive opportunity in retail is because consumers are actually pulling us in.

You see this not just in the buying behavior, which is great. But you also see this in the search behavior. You see this in what we talk about with retailers as well. A lot of what DoorDash does is kind of 2 sided, right?

We have a consumer business, which maybe most people know us for, but we also have a commerce platform, which -- and already, we've created 2 of the most successful B2B products out there with DoorDash Drive and DoorDash Storefront. Obviously, we're adding SevenRooms to the mix, which is really in the restaurant category. But in the retail category, we get a lot of requests to help there because I think these retailers recognize that we have a large audience, a large consumer base that we can certainly bring, but we also have capabilities that we can bring, right? And DashMart Fulfillment Services is now part of the capability offering.

Online ordering and fulfillment with Drive always have been there. But -- so as you kind of think about this, DoorDash is a 2-sided player in the retail realm where, obviously, there's the consumer front, but there's also the B2B front with retailers.

Yes, Lee, on the payback period, right, like, look, nothing is really changing about how we operate the business, right? Look, when we think about investments in payback, I think of it in terms of 2 dimensions. One is, is it ultimately improving retention order frequency. The other one is, is it driving free cash flow and IRR.

So that framework and how we operate, that discipline is not changing. Let me give you an example, right? We start with small levels of investment. As we find product market fit, we continue to increase the level of investment.

Look at our new verticals business where we've increased our unit economics year-over-year. Look at our overall international business. That's close to being contribution profit breakeven. That just goes to show you the discipline of us being operators in how we operate the business.

As you think about the investment areas, look, I mean, the software side, the payback period is going to be shorter, but our goal is to continue to invest because that's going to drive revenue. Whereas if you think about our tech stack, all of that is going to increase our tech feature development velocity, which ultimately is going to help us release features faster, which is going to drive retention and unit economics, right? That's the distinction we have as we think about our investment areas, but the core philosophy and discipline in how we are operating the business, that is not changing.

Next question comes from the line of Justin Patterson with KeyBanc.

This is Miles Jakubiak on for Justin. I'd like to start with one on grocery. Just curious if you're seeing -- you had some nice grocer adds during the quarter. Curious if you're seeing any increase from grocers to move a bit faster on the delivery side and coming to the platform.

And then one just on going out or dine-in, in-person dining. Saw you had the Going Out launch and some SevenRooms stuff launched during the quarter, so curious if you could just expand on how you view that dining out experience fitting within the DoorDash ecosystem and the opportunity there.

Sure. Miles, it's Tony. I'll start, and feel free to add in here, Ravi. Look, on the first question with grocery, you're right, I mean, we're super excited about all the selection that we're adding, including Kroger, who we announced in the recent quarter.

I think the short version of this is we've never been in a better position in grocery. And we're also adding across the board, not just the top grocers in terms of kind of the national scale players but also the local grocers and really across the board. I think every grocer is recognizing that DoorDash is now the leader in order volume in this category and the leader in acquiring new customers. And so this is something that I think people understand.

And as a result, it's become more habitual that way. I also think it's interesting that some of these grocers, kind of similar to the comment I made earlier about retail, are starting to ask us to help with other things. And so I think there's an interesting -- there's always this kind of 2-sided opportunity for us really on the -- certainly on the consumer front but also B2B. With respect to Going Out, you're right.

Going Out has been an experiment that we've been running for a while, and we kind of shipped more recently in some select markets. And it's going really well. I mean I think this kind of goes to maybe some commentary. I forget whether it was 4 earnings calls ago or 5 where we talked about there's more than 1 way to connect every local business and every local consumer.

Today, I know we're known predominantly as bring everything to you, whether you're at home, you're in the office or wherever you are, but there's no reason why we can't bring you to everything. And again, the goal of the company is to connect every local business with every local consumer. On the consumer front, we obviously have the largest audience with the most number of kind of frequency not just in terms of buying but also just in opening the app and the searches and everything else that they take action on in the product. But we also have quite a lot of information with merchants because of our B2B commerce platform.

A lot of these merchants are increasingly asking us to help them from not just online ordering, we have hundreds of thousands of those businesses, but analytics and marketing solutions. And I think with the addition of a product like SevenRooms, we can really democratize this ability that I think a lot of tech companies have and give it to every small, medium and large restaurant to be able to understand all of their guests and really allow the restaurants to maximize whatever might be the best thing to maximize for them at that moment in time. Sometimes, it's going to be driving new customers.

Other times, it's going to be driving adoption of a new product. And that's really what the point of Going Out is. It's really to introduce customers to restaurants that they maybe never tried before or add an occasion to build increasing loyalty to an existing restaurant that they do have a relationship with. Today, DoorDash effectively has a lot order frequency on bringing things to you, and we have almost no order frequency on having you go to a store, but we believe that can change.

And our last question comes from the line of Ron Josey with Citi.

That's great. Going back to Roo really quick. Understood the investments needed here and the acquisition literally closed, what, around 4 weeks ago. But would love to hear what you all think or learned thus far on ways to improve the product and the consumer experience as we think about these investments.

And then I believe it was mentioned in the letter or the press release that unit economics were flat quarter-to-quarter for U.S. restaurants. I'm just wondering if that's a change in trajectory or anything to call out there.

Yes, I can start, Ron, on Roo. Maybe, Ravi, if you want to take the unit economics question. On Roo, I think there's a lot that we can do, and I think that's mostly because DoorDash has ran just a lot of experiments at this point, tens of thousands of experiments. And as you can probably guess, the vast majority of those experiments fail and never make it to you as a customer, but we have taken a lot of these lessons that have done well and brought them over to Wolt, for example.

And we think that there's a lot of experiments that Wolt has ran, too, that can be portable to Roo in addition to those that would come from DoorDash. And so there's no like one thing. I think that like I probably bore you with the list of features that, that would be talking about, but this really is the same answer to the question of how is it possible that the DoorDash marketplace can -- the U.S. marketplace, that it can continue to grow faster at higher volumes and increase our penetration with users. And it's because there is no one thing.

There's thousands of small things that add up in terms of how we deliver on the selection, the quality, the affordability and the service. And so it will be a collection of things that we've already identified, and we're excited to ship them.

Ron, on the second point around U.S. restaurants, I mean, look, the business is doing well. We talked about the fact that the business is growing. Unit economics are still progressing quite nicely. Quarter-to-quarter, there's moving parts.

But overall, I mean, really excited about what you're seeing from a unit economic perspective. I think a few years ago, we talked about the fact that the incremental margins over the last 8 quarter average was above 7%. That still continues to be the case. I mean this business is growing at larger scale, accelerating as well as the unit economics continue to progress quite nicely.

So we're really pleased with how the restaurant business is doing in the U.S.

Ladies and gentlemen, that concludes the question-and-answer session. Thank you all for joining, and you may now disconnect.