Prepared remarks
Hello, everyone. Thank you for joining us, and welcome to the DoorDash Q2 2026 Earnings Call. I will now hand the call over to Weston Twigg. Please go ahead.
Thanks, Connor. Good afternoon, everyone, and thanks for joining us for our Q2 2026 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, profitability, our guidance, strategies, capital allocation approach, and 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-Q. You should not rely on 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 webcast 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.
Questions and answers
The first question comes from Michael Morton of MoffettNathanson.
I wanted to ask about the grocery business as you've talked about improving the unit economics. From our understanding, there are some grocers on the platform who are paying effectively zero or very low take rates; they came on looking to see if you could drive demand and how well they could work with DoorDash. I was wondering if that's the case and then what the opportunity is to reprice these relationships going forward, maybe pushing some of the affordability burden that's leaning on DoorDash right now back to the grocers.
Yes. Michael, it's Tony. I can start and Ravi can chime in. What I would say is we see extremely strong performance in our grocery business. It's the fastest-growing part of our marketplace business, and we have very healthy relationships with all partners on the platform. In terms of the economic relationships, I'm not going to comment about anyone in particular. What I will say is that when you are the fastest grower in the market for them, and you are their source of growth — put another way, we might represent 100% of the growth that they see in their actual business — you certainly have opportunities to grow your business with them as well as improve those relationships. If you look at our business as a whole, one of the things you see from this quarter, and frankly, many of the periods leading up to now, is that there are many sources of improving economics. We've improved unit economics across all of our categories. We have improving unit economics in our restaurant business, too. We have improving unit economics in our different geographies in which we operate. We have increasing adoption of our DashPass program as well as accelerating growth in our ads business. When you add all of that together, we have a business with many levers we can control to shape our financial profile and make great investments. Ultimately, we're always seeking the next best investment. It doesn't mean that we always make those investments, but when we see the opportunity, we're leaning in. That includes all of the work we're doing in grocery, which we think has a long runway, as well as the other opportunities in front of us.
And Mike, just to add, if you take a step back and think about our overall grocery business as well as new verticals, we became order volume share leaders in Q4 and have continued to extend that lead. Second, when you look at underlying growth in MAUs — the number of users that use categories outside of restaurants — that number and the frequency are growing. We noted in the letter that basket sizes are growing. If you look at our historic cohorts, consumers are using us for more use cases, which is driving overall basket sizes higher. Last call, I mentioned we expect our overall new vertical business to be gross profit positive; we're on track for that in the second half of the year. If we think about the business as a whole, we focus on retention and order frequency as well as underlying improvement in unit economics, and they're all headed in the right direction for us.
The next question is from Mark Mahaney from Evercore ISI.
I'll ask a question about Deliveroo. You've had now three quarters in a row of accelerating growth in orders and in GOV and revenue. So just peel that back a little bit — what different opportunities have you had, what have you been able to pull and change to deliver that better performance? And is it also showing up on the bottom line? Have you been finding ways to improve the profitability of Deliveroo as well?
It's Tony. What I would say on Deliveroo is it's really a story that probably started back in 2021 when we made our first large acquisition overseas, Wolt. We've learned a ton about building our U.S. business and about how Wolt has operated across Europe, and we've integrated lessons learned while recognizing that some lessons don't apply everywhere. What you're seeing in Deliveroo is accelerating performance across the board. That's exciting because it's validation that our integration work is working and that lessons from building marketplace businesses around the world translate into meaningful geographies where we're seeing growth in all our major international markets. This doesn't even include the majority of benefits we expect to see once we finish building a single tech stack. As that work comes more fully online, towards the beginning of next year, we expect to see even more benefits over time.
And Mark, to add, when you look at the actual performance of Deliveroo itself — volume growth, MAU growth, subscription growth — year-over-year it's been the highest we've seen in the last couple of years. On your question about profitability, we've increased unit economics as well. The way we're operating the business is similar: we're finding opportunities to drive investment back into selection and quality. Subscription is a big area of focus for us, and we'll continue to invest back in the business. My expectation is we'll continue to drive higher top line while meeting the profitability targets we set out in the last letter.
The next question is from Nikhil Devnani of Bernstein.
I'll stick with the theme of international. There's a common perception that international might be lower quality growth because maybe you're not number one everywhere or the competitive set is better funded and more consolidated today. I'd love your perspective on that sentiment and how you structurally see the longer-term earnings power or quality of growth out of these international markets relative to the domestic business. How important is it to have a pure market share number versus minimum viable scale in jurisdictions that allow you to operate well?
Nikhil, a couple of points. Our aspiration is to be the global leader in local commerce, and we think we're well-positioned given our deep and broad portfolio of products to serve those audiences. This is a game played locally; there are no global network effects in these businesses. Sometimes there's not an obvious tie between market position and economic profile because it is a minimum viable scale business. That said, the vast majority of our international business is concentrated in our top 10 markets outside the U.S., and in those markets we are the leader or a very strong number two, gaining share in markets like the U.K., Italy, Germany, the Nordics, Israel, and Canada among others. We're making fundamental improvements to core propositions for all audiences: wider selection, better prices, better quality of delivery in terms of reliability, accuracy and speed, and improved customer service. We also see runway to bring our portfolio of B2B products, which have done well in the U.S., to markets outside the U.S. where digital technology adoption in restaurants and retail is more nascent. The potential is very large.
Nikhil, the results are clear. On the Deliveroo side, growth is accelerating and underlying cohorts show some of the highest growth we've seen in the last couple of years. Excluding Deliveroo, MAUs are growing and order frequency is growing. Wolt+, our subscription program in Wolt, had a record quarter in paid subscriber growth. It's not just about growth; we're improving unit economics across Deliveroo and Wolt. Gross profit and contribution have continued to improve year-over-year.
The next question is from Deepak Mathivanan from Cantor Fitzgerald.
Great. Tony, last month Andy Fang talked with Boris at Claude about how DoorDash is aiming to translate AI spend into outcomes closer to business metrics while also letting employees experiment aggressively with AI tools. Can you talk about where you're seeing this attribution clearly now? How should we broadly think about AI spend at DoorDash over the next 12 to 18 months? And Ravi, a question for you: U.S. restaurant GOV acceleration — can you expand on the drivers of the growth? I know weather was disruptive last quarter, and there were factors like the World Cup. Curious about the drivers of acceleration in Q2.
On AI, a couple of things. First, any technology we adopt, including AI, must be rooted in delivering a better customer experience. If it's not solving a real problem, there's little purpose. One recent product is DoorDash Ask, an ordering agent that helps customers discover restaurants similar to ones they've ordered before but new to them, and helps build a grocery cart in under two minutes — directly solving real pain points in an increasingly larger and more diversified marketplace. On the merchant side, we've automated building catalogs for retailers and menus for restaurants, including photos and metadata, enabling faster merchant onboarding and same-store sales growth. For Dashers, we're seeing routing improvements and ways to help Dashers find the best areas to Dash. These are examples of AI applied to drive customer outcomes that ultimately deliver business results. We also balance efficiency with invention: we've built internal tools like DashBench to model the appropriate tools and give teams control over token spend and intelligence required. Our approach to AI is disciplined while allowing customer-focused invention.
Deepak, on restaurant growth, Q2 was quite strong and growth accelerated from Q1 to Q2. Much of the growth is coming from increased DashPass subscribers. We added more DashPass subscribers in the last year than in the two prior years, and paid subscriber growth was one of the highest we've seen in the last couple of years. That reflects an improving underlying product and increased investment in selection and quality. Mature cohorts are engaging at higher rates, new consumers are strong, and all of this is driving the restaurant growth you see. Also, Q2 of last year was unusually strong, so comping against that and still delivering strong numbers is testament to the demand and product improvements.
The next question is from Dominic Ball of Rothschild & Co Redburn.
Interesting commentary about investing more in merchant services and software. DoorDash has been testing its POS product in a few U.S. markets with both SMBs and enterprise restaurants. How are these test trials going, what products and features are resonating, what drives some restaurants to choose DoorDash here, and how should we think about a broader commercial launch?
Our vision is to be the best partner to every local business. We want to give every business the same tools we built for ourselves so they can grow their digital businesses. We operate in an ecosystem with at least three offerings: our marketplace; tools to build digital businesses for restaurants and retailers — that business serves over 150,000 businesses and grew 40% year-over-year in the quarter; and products that drive customers inside stores, including go-outs and reservations. On the B2B side, we have SevenRooms, which we acquired about a year ago. This ecosystem allows us to help customers build relationships with local businesses and forge regulars for them. We can do this because we have the biggest scale and deepest customer datasets to drive engagement. A customer may start by ordering delivery, then sign up for a merchant's first-party loyalty program we've built, and both DoorDash and the merchant can incentivize the customer to visit the store. This interaction effect enables us to be the best growth partner for businesses while giving consumers choice in how to interact. The proof points highlighted in our quarterly update are strong, and this ecosystem has a long runway.
The next question is from Jason Helfstein of Oppenheimer.
Two questions. One, you talked a little about Dot deliveries in the release. Anything you want to share on how you're thinking about unit economics today versus long term and how that impacts demand and elasticity? Two, is it possible to share the AOV for the most recent period in the chart on Page 3 comparing restaurant versus grocery and retail?
I'll start on DoorDash Dot. Our vision for autonomous delivery is to offer AVs throughout the network to deliver the best service: fastest delivery, most affordable delivery, and delivery from the widest selection, including faraway places. The key is marrying operations with technology; that's the complexity and the secret sauce. An AV alone will not scale autonomous deliveries meaningfully — you must solve operational problems: loading at the merchant, estimating prep times or inventory, handling drop-off issues like doormen or complex gates. We've encountered many such issues in our test markets, and reaching our current milestone with meaningful scale in our test market is a huge accomplishment. It gives us confidence that scaled autonomous delivery is possible if you master both operations and technology. We run the network and are building the technology, which gives us a detailed understanding of how to make this happen. We also partner across land and air modalities. Our autonomous delivery platform solves these operational problems for partners: merchants take their existing DoorDash integration and get access to any AVs, and customers will gain the future benefits of these technologies. Regarding cost profile, we're excited by what we see — exponential progress — but the first milestone was commercializing in a real way, similar to what some robotaxi providers have done in rideshare; we've begun to do it with delivery.
Jason, on the AOV and baskets: as we expand new verticals and grocery, as the product gets better, basket sizes increase. This was expected: as we add selection and improve product quality, customers use us for more use cases, driving higher basket sizes. We're seeing this in older and newer cohorts, visible in the overall business. Our thesis and the chart in the letter used an older cohort to represent the trend: consumer spend on restaurants and new verticals is increasing, and DashPass penetration is increasing. This is a self-reinforcing loop where the product gets better, consumers habituate to DashPass, they use the product more, driving more growth and more profit dollars. That's been our focus, and we're pleased with the cohort performance underlying the business.
The next question is from Shweta Khajuria of Wolfe Research.
Two questions. First, on DashMart Fulfillment Services since your launch and working with a handful of partners, what have you learned so far that you can quantify or share as tangible learnings — for example, magnitude of customer experience improvement? What metrics do you look at to decide to scale DashMart Fulfillment Services? Second, on overall EBITDA growth as we move into the back half of the year and in light of investments this year, how are you thinking about balancing growth and EBITDA growth given current demand trends?
On DashMart Fulfillment Services: we want every local business to succeed. Grocery delivery has a structural challenge: grocers often don't know their inventory because consumers move items around, making it difficult to offer a premium delivery experience while ensuring customers get exactly what they ordered. DashMart Fulfillment Services manages warehouses where we control inventory and sell exactly what's in stock so we can offer near-perfect accuracy and fast delivery. Since announcing last fall and launching with several partners, we're seeing lots of incremental demand; these warehouses run near 24/7 versus traditional store hours, dramatically increasing TAM and solving customers' timing needs. We're also seeing 10x better error rates because we're managing inventory, so customers get exactly what they ordered. All the signs are positive for scaling DashMart Fulfillment Services. It's complicated and involves physical infrastructure plus inventory management technology, but it can power a city's needs and unlock much greater selection for customers — today DoorDash delivers perhaps 1/10 of available selection in most cities, and Fulfillment Services can expand that dramatically while maintaining quality and speed.
On balancing growth and profitability: demand trends remain strong. We focus on cohorts — MAUs are at all-time highs, subscriptions domestically and internationally had one of their best quarters, restaurant growth accelerated, and Deliveroo continues to do well. Product improvements drive retention and order frequency. The quarter was strong from an underlying profitability perspective as unit economics improved. Our philosophy hasn't changed: we consistently work on efficiency across the P&L and reinvest in the business. We're investing in autonomy, the unification of the global tech stack, and merchant services; these are on budget and on plan. These investments increase the surface area, but the goal is to drive both growth and profit dollars. As we make the product better, we've been able to do both.
The next question is from Josh Beck of Raymond James.
On Dot reaching high single digits within a market by the end of the year, can you describe market characteristics — for example, longer route suburb types? If you add modalities like sidewalk robots, pathways, drones, and AVs, is there a ceiling you have in mind for the percentage of orders that could be filled autonomously? Second, on the AI assistant: is there evidence of better frequency for customers using it? On ad monetization, it seems time spent could be less but conversion better — any observations there?
You asked several questions. On DoorDash Dot: it's representative of a DoorDash market — we're testing in Phoenix with real scale. These are real deliveries for tens of thousands of customers, not demos or fixed routes, and that's a real accomplishment. There are many issues to solve to scale, and that's true whether by land or by air. The ceiling for penetration depends on mastering both operations and technology; we believe we're best positioned because we do both in-house. Dot traverses road, sidewalk and bike lane autonomously; it's the only vehicle in the world doing that autonomously, and multimodality adds complexity. The most interesting thing is our autonomous delivery platform — the brains that decides which vehicles go to which orders, how to mix human dashers and AV legs, package configuration, and more. We reduce that complexity so merchants don't need to change their integration, and customers get access to speed, cost and selection benefits. On the AI assistant, reducing friction leads to more usage: make it easier to build a grocery cart, you get more and larger carts. The same applies to discovering new restaurants: it increases order frequency. We eat 20 to 25 times a week and touch only a fraction of that; there are many meal occasions and shopping needs — north of 100 per month — so there's a large runway. Reducing ordering friction and mastering fulfillment will move all metrics in the right direction.
The next question is from Youssef Squali of Truist Securities.
Can you double-click on margin? Ravi, can you talk about drivers for the material beat in adjusted EBITDA in Q2? It came quite a bit outside the guidance range. More recently you've guided to land somewhere in the midpoint; this quarter is dramatically higher. Were any investments pushed back into Q3 or Q4? Are we still tracking to show higher year-on-year adjusted EBITDA margin for 2026 excluding Deliveroo?
Youssef, at a high level, the core restaurant business continues to perform on all cylinders: growth accelerated, and unit economics improved. New verticals are growing faster than restaurants and are on track to be gross profit positive by year-end. International is doing well, and Deliveroo beat our internal volume expectations and is contribution profit positive, which contributed to upside. Specifically, unit economic improvements came in ahead of expectations in areas like ads and subtotal in the second half of the quarter. ROO beat our internal volume expectations and was contribution profit positive, which led to some of the upside. Our operating philosophy isn't changing: we tightly manage the business, look for sources of efficiency across the P&L and reinvest in the business. We're not optimizing the last dollar from one quarter to the next; we focus on building a durable business that increases profit dollar production over time. Regarding guidance, our focus is to land inside the range, and for the second half from a margin perspective I would expect us to land inside the Q3 guidance range. There will be times like Q2 where EBITDA comes in late in the quarter and we didn't have enough time to reinvest back in the business at our desired levels of efficiency. Q2 was such a quarter, and we were happy to drop it to the bottom line.
The next question is from Ross Sandler of Barclays.
Quick follow-up on AVs: how quickly do you plan to bring Dot to other cities? You mentioned you have a few hundred robots in Phoenix. Also, on your charts in the letter, are there examples of cities or countries where you're demonstrating the same trend as the U.S. — where you've either come from behind or overtaken competitors on DashPass or Wolt penetration for subscribers?
We're excited about what's happening in Phoenix, but you have to nail something before you scale it, especially in autonomy where you're solving problems across many dimensions and need tight orchestration between operations and technology. There's a lot of work to do. We're in parallel securing permits and working with cities to unlock expansion. We have plans to expand and will share more over time. The road ahead is exciting but will take time and great execution between operations and technology.
On subscriptions and the chart: subscription has been a key focus for years and continues to perform well. As DashPass penetration increases, overall gross profit per MAU increases. We're still early on penetration and see similar behavior in international markets, though some are slightly behind due to later subscription launches. For specific examples, on Deliveroo in the U.K. we're gaining share, accelerating paid subscriber and volume growth compared to prior years. Across Wolt countries, we are gaining share and order rate improvements have continued over three- and six-month windows. Across Deliveroo and Wolt, this has driven efficient growth with improved unit economics year-over-year.
The next question is from the line of Brian Nowak with Morgan Stanley.
On the global tech stack unification: now that you've built the stack as of the spring and are testing modules and capabilities, can you give examples where you're seeing early signals with quantifiable benefits from new modules that give confidence you'll get real return and ROIC on these investments into 2027?
We're still building a single tech stack, and it's an iterative process — more like replacing an engine while flying a plane than following a fixed instruction manual. We're seeing benefits already: conversion wins from search improvements, automation wins in customer support flows by bringing automation from one area to another, and more. The theme is taking best-of-breed features and offering them across our 41 markets. It's not copy-paste; it's combining the best of each into a new engine, which is hard work but justifies the investment because it offers greater future velocity: ship once and it goes everywhere. That's the thesis for return and why we're investing.
The next question comes from the line of Ronald Josey of Citigroup.
On the gross profit and DashPass chart in the letter: some lines are steepening. Tony, can you discuss plans to continue adding value to DashPass, and what that does to gross profit? Also, recently the company launched newer or higher fees for larger delivery radiuses — talk about the reasoning and the benefits?
Ron, subscription continues to do well. Our thesis has been that as consumers habituate and graduate to DashPass, their value increases — order frequency and retention go up, which increases gross profit. The chart used a January 2021 cohort because it's old enough to show trends and is representative of other cohorts. As consumers order more and retain more, they graduate to DashPass, and as they adopt DashPass they continue to spend more. In the U.S., DashPass paid subscriber growth in Q2 was one of the highest we've seen in the past two years; we added more paid subscribers in the last year than in the two prior years. We increase DashPass value by improving selection, product quality and affordability, which drives adoption and engagement. On the delivery radius fee we announced recently, it's largely a realignment of what consumers pay relative to the time and effort Dashers put into deliveries. From a P&L perspective, I wouldn't expect a massive impact — in markets where we've launched it so far, the fee is similar or slightly less for the vast majority of orders.
The next question is from the line of Justin Post of Bank of America.
Any update on agentic traffic? Are you seeing traffic from agentic partners, and given your large merchant scale, could you capitalize on that traffic as they roll out booking capabilities to lower marketing costs?
Justin, volume has been quite low from agentic partners we've tested with so far, which isn't too surprising. For many large platforms, the core focus has been enterprise rather than agentic flows. Structurally, consumers care about fulfilling their needs — getting their burrito or their pair of shoes — not the name of the flow. They care about end-to-end experience: knowing inventory, what's in stock, managing logistics at the merchant and drop-off, and handling exceptions. Those are details you must get right for agentic commerce in our categories, and we are filling that void with products like DoorDash Ask. We're open to testing all incremental traffic sources. I've said before there will be two big battles: attention (chat assistants) and atoms (fulfillment). They will partner; our focus is mastering the physical world so we can be the most useful partner to digital assistants as agentic commerce develops.
The next question is from the line of Doug Anmuth of JPMorgan.
You don't manage for take rate, but can you give more color on net revenue margin and the pickup in 2Q? How much was Deliveroo contribution versus advertising and fee changes, and how should we think about that going forward?
Doug, we're not operating toward a take rate or net revenue margin percentage; our goal is to optimize overall profit dollars. The quarter's Q-on-Q increase in take rate was largely driven by Dasher costs, which are seasonal and lower in Q2 compared to Q1. That contributed to the take rate tick up from Q1 to Q2. For modeling, expect Q3 to be similar to Q2 (flattish) and Q4 to be lower, because Dasher costs are higher in Q4. But net-net, we don't manage to a specific take rate percentage; our goal is to invest flexibly up and down the P&L.
The next question is from the line of Tom Champion of Piper Sandler.
Ravi, what's your big-picture view on AI spending and the ROI you're seeing? Has it impacted hiring plans?
We view AI as both a productivity and product improvement lever. We're encouraging teams to use AI to build better products for customers, which drives growth and profit dollars. Examples include Ask for customers, merchant onboarding automation, and conversational bots for Dashers. Internally, we're using AI in sales, accounting, marketing and finance. We're seeing productivity gains. We built models to route tasks to the right model based on cost, quality and efficiency; we've put caps in place and incorporated AI budgets into teams' existing budgets. We're seeing gains from AI usage. The key question is how we take efficiency gains and reinvest them in the business. We're investing in areas with long-term ROI: autonomy, AI and the unified tech stack. The philosophy is consistent: reinvest efficiency gains to build scale and durability to increase long-term free cash flow. We're pleased with current results.
There are no further questions at this time. This concludes today's call. Thank you for attending. You may now disconnect.