Record Orders, Shrinking Take Rate: DoorDash Buys Growth With Margin and the Tape Notices

Results vs. Consensus

Q1 2026 Scorecard

DoorDash publishes GAAP diluted EPS only; there is no company-defined adjusted EPS. A second consensus provider carried $0.41 for the quarter against the $0.36 the tape traded on, which is a vendor-normalization gap rather than a data dispute. Both providers clustered revenue at $4.12B to $4.15B and Total Orders at 954M to 956M, so the direction of the two misses is not in question.

Year-over-Year Comparison

Sequential Comparison

Revenue: the wedge between order value and revenue

Marketplace GOV grew 37% and revenue grew 33%, and the four-point gap is the whole story of the quarter. Net Revenue Margin, which is simply revenue divided by GOV, fell to 12.8% from 13.1% a year ago and from 13.3% last quarter. The company attributes the decline to lower fees charged to consumers as a share of GOV and to the Deliveroo mix, partially offset by advertising and by lower credits and refunds. Two of those three drivers are permanent: Deliveroo carries a structurally different take rate and is now nine percent of the revenue base, and a decision to charge consumers less per order is a pricing posture, not a timing item.

Underneath the acquisition, the legacy business is decelerating on the metric that matters most. Order growth excluding Deliveroo ran 16% in Q1 against 20% in Q4 and 21% in Q3. Management sized the winter-storm drag at roughly one point of GOV growth, which leaves a three-point deceleration unexplained by weather. The company's own release is candid about the restaurant category: U.S. restaurant GOV growth was faster than a year ago and slightly above the sixteen-quarter average, but slower than Q4. Three separate framings, all pointing the same direction.

Margins: a take-rate problem, not a cost problem

The instinct on seeing Adjusted EBITDA margin compress is to look for spending indiscipline. It is not there. Bridging the 17 basis points of year-over-year compression in Adjusted EBITDA as a share of GOV shows the entire decline arising above the operating-expense lines, with sales and marketing actually giving back leverage:

Read that table the way an operator would. Marketing efficiency improved materially, R&D and G&A cost a combined twelve basis points as the platform build and legal accruals scaled, and adjusted gross profit per dollar of order value fell nineteen basis points. In the company's own GAAP framing the same picture holds: cost of revenue fell to 6.3% of GOV from 6.5%, sales and marketing to 2.4% from 2.5%, and R&D and G&A were each flat at 1.3% and 1.4%. Nothing in the cost base got worse per unit. What got worse is how much revenue each order carries.

Assessment: this is the single most important disclosure in the quarter and it is the one least discussed on the call. A take-rate decline driven by consumer-fee reductions is a competitive-intensity signal dressed as an affordability strategy. If DoorDash is trading take rate for volume, the correct question is not whether volume is growing, it is whether the volume it buys retains at a rate that makes the trade accretive. Management's cohort commentary asserts that it does. The company does not publish the cohort data that would let anyone check.

EPS and the acquisition-accounting drag

GAAP net income of $184M is 5% below the year-ago quarter on 33% more revenue, and the bridge is entirely in non-cash and one-time lines. Depreciation and amortization added $117M year over year, $83M of it amortization of acquired intangible assets from Deliveroo, Wolt and SevenRooms. Restructuring added $47M. Those two items alone are $164M pre-tax, against a $9M decline in net income. On the company's own full-year framing, roughly $450M of the $1.1B to $1.2B of 2026 depreciation and amortization is acquired-intangible amortization, so this drag persists through at least 2027 regardless of operating performance. Investors who anchor on GAAP earnings for this name are measuring purchase accounting, not the business.

Segment Performance

DoorDash reports as a single operating segment and does not publish segment profitability. The two disclosures that carry analytical weight are the geographic revenue split in the quarterly filing and the Deliveroo carve-out in the release.

Revenue by Geography

Deliveroo Carve-Out

United States

U.S. revenue of $3,104M grew 17%, a rate that sits meaningfully below the 24% consolidated GOV growth excluding Deliveroo and reflects both the domestic take-rate reduction and the faster growth of the international base. Within the U.S., management describes restaurant growth as accelerating year on year and running slightly ahead of the sixteen-quarter average, while grocery and retail added more new consumers in the quarter than in any prior quarter and improved order rates among mature cohorts. Neither category carries a disclosed dollar figure.

Assessment: the U.S. is doing what a mature share leader does, which is trade some price for occasions. Seventeen percent revenue growth on a $12B annualized domestic base with a fee reduction embedded is a respectable outcome, but it is the slowest-growing part of a business the market values on 30%-plus GOV growth, and it is the part that will still be here after the Deliveroo comparison laps in Q4.

International

International revenue nearly tripled to $932M as Deliveroo entered the base on October 2, 2025. Excluding Deliveroo, management says international unit economics improved year on year, that constant-currency international GOV growth decelerated versus Q4 but remained above the U.S. marketplace, and that Deliveroo itself is running at its highest growth rate in four years with accelerating GOV in the U.K., France and Italy. The filing notes that no single country outside the United States reached 10% of consolidated revenue, so there is no way to size any individual market.

"I mean we've never been stronger is the short answer in Europe. I mean, if anything, Deliveroo is seeing the highest growth rate it had in the past 4 years, and it's actually been reaccelerating in growth each of the months in which we've been operating it." — Tony Xu, Co-Founder, Chair and CEO

Assessment: the Deliveroo integration is going better than the base case a market that spent 2025 worrying about the price paid would have assumed. That said, "highest growth rate in four years" is a comparison against a company that was, on any read of its standalone history, growing slowly. The credible tell is not the growth rate, it is that management chose to exit four countries within five months of closing rather than defend them. That is the behaviour of an owner willing to shrink the map, and it is the more useful datapoint about integration discipline.

Unit economics per dollar of order value

Assessment: contribution profit per GOV dollar held flat year over year but gave back 30 basis points sequentially, and every other line moved down. For a platform whose entire equity story rests on operating leverage arriving as volume scales, a quarter in which volume scaled 37% and none of the per-unit profitability measures improved is the quarter the leverage thesis has to answer for.

Key Topics & Management Commentary

Overall Management Tone: confident on demand, deliberately unspecific on cost. Management fielded every question on growth, product and competition with detail and conviction, and answered every question that touched the size or duration of the 2026 investment load with framework language and a reiteration of the full-year margin frame rather than a number. There is no defensiveness in the posture; there is a consistent refusal to be pinned, and it was most visible on the replatforming spend and on how long gas relief runs.

Note also the format: DoorDash publishes a shareholder letter in lieu of prepared remarks and opens the call directly into questions, so the entire hour of management commentary below is responsive rather than scripted.

1. The take-rate reduction nobody asked about

Net Revenue Margin fell 30 basis points year over year and 50 basis points sequentially, to the lowest level in the five quarters the company presents. The release names the drivers plainly: lower consumer fees as a share of GOV, and the Deliveroo mix, offset in part by advertising and by fewer credits and refunds. Across eleven analyst questions, not one asked management to size or defend the consumer-fee reduction.

Assessment: the market spent the call litigating agentic commerce and the platform rebuild while the company quietly disclosed that it is keeping less of each order. Over four quarters, 30 basis points on a $130B annualized GOV base is roughly $390M of revenue. That is a larger number than the gas relief, the storms and the restructuring charge combined, and it is the only one of the four that does not have an end date attached.

2. Deliveroo is carrying the headline and reaccelerating underneath it

Deliveroo contributed $362M of revenue, 12 of the 33 points of revenue growth, and roughly 11 of the 27 points of order growth. Management reports accelerating year-over-year growth in Deliveroo monthly active users and Total Orders, and accelerating GOV growth in the U.K., France and Italy. Integration work on the shared platform has begun, and the acquired business is expected to contribute approximately $200M to 2026 Adjusted EBITDA, a figure carried forward unchanged from the prior guide.

"At Deliveroo, early output from our investments and integration work has been promising. In Q1 2026, we drove accelerated Y/Y growth in Deliveroo MAUs and Total Orders, and accelerated Y/Y growth in Marketplace GOV in the U.K., France, and Italy." — DoorDash Q1 2026 shareholder release

Assessment: the acquisition is working on the metrics the company can see and we cannot. The reiterated $200M contribution is the one auditable commitment, and it is worth roughly six percent of our full-year Adjusted EBITDA estimate. The comparison laps in Q4 2026, at which point the consolidated growth rate steps down toward the ex-Deliveroo rate mechanically. Anyone modelling 2027 off a 37% GOV growth rate is modelling an acquisition, not a business.

3. Organic order growth decelerated five points in two quarters

Excluding Deliveroo, Total Orders grew 16% against 20% in Q4 2025 and 21% in Q3 2025. Asked directly about the deceleration, the CFO sized the winter-storm effect and redirected to input metrics, describing record membership sign-ups, an all-time high in monthly active users, rising order frequency and share gains in most markets.

"The impact purely from a winter storm perspective is roughly about 1% on a year-over-year growth perspective from a GOV basis. Look, when I look at the underlying demand, Nikhil, it continues to be very good, right? We've talked about MAUs reaching an all-time high." — Ravi Inukonda, Chief Financial Officer

Assessment: weather explains roughly one point of a five-point deceleration and management did not attempt to explain the rest. The input metrics cited are all real and all directionally supportive, and none of them is published as a number, so the deceleration is the only quantified fact in the exchange. We treat 16% organic order growth as the honest run rate entering Q2 and would need to see it stabilise before crediting the demand narrative in full.

4. Dasher gas relief: an unguided cost absorbed inside the guide

Gasoline prices rose sharply during the quarter following conflict in Iran, and DoorDash introduced a Dasher gas relief program that was not contemplated in the February guidance. The Q1 impact was roughly $50M and the company anticipates a gross cost of over $50M in Q2. Management found offsets by pushing unspecified first-half investments into the second half, and made no commitment on how long the program runs.

"Q1, we obviously had the impact from both winter storms as well as the introduction of gas rewards. Q2, we did extend the gas rewards program. Rough impact of that in Q1 was about $50 million. The projection for the impact in Q2 is also going to be about $50 million. Like I said earlier, we did find offsets in the business." — Ravi Inukonda, Chief Financial Officer

Assessment: absorbing $50M of unbudgeted supply-side cost and still landing Adjusted EBITDA in the upper half of the guided range is the most impressive operational fact in the release, and it is under-appreciated because the reported dollars do not show it. The offset mechanism is where the caution belongs: pushing investment from H1 into H2 improves the first half and loads the second half, which is the same second half that has to deliver the margin inflection the full-year frame requires.

5. The global technology platform: three stacks running at once

The single largest component of what management has described as "several hundred million dollars" of platform investment is a new global technology stack intended to replace three separate codebases across DoorDash, Wolt and Deliveroo. Design and domain mapping are complete, production traffic is ramping across all three brands, and foundational infrastructure is live across payments, fraud, support, subscriptions, merchant tooling and logistics. Asked whether the P&L is currently carrying redundant costs for all three stacks, the CFO confirmed it is and gave a phasing rather than a figure.

"There is a portion of the spend, which is redundant in the sense that we are going to run all 3 tech stacks in parallel while we're working on the global new tech stack. That is going to phase in. It's going to phase out. My expectation is majority of that will run through '26. Maybe some portion will bleed into early 2027, but that will bleed out." — Ravi Inukonda, Chief Financial Officer

The strategic case is velocity and quality: one build instead of three, and the ability to port the best of each regional product into a shared base. The CEO's example was logistics, where the constraints of dense older European cities produce techniques that do not arise in gridded American ones, and the reverse for long-distance and large-catalog problems.

Assessment: the logic is sound and the execution appears to be on schedule, which is more than most platform rewrites of this scale can claim two quarters in. The problem is that a project consuming the majority of several hundred million dollars of investment, carrying duplicate run costs through 2026 and into 2027, has never been given a bounded number by the company. Investors are being asked to underwrite the payback without the cost. That is a governance choice, not an accounting constraint.

6. Four country exits and a $48M restructuring charge

Between the Q4 print and this one, DoorDash announced the wind-down of Deliveroo and Wolt operations in four countries and the closure of an engineering hub, taking a $48M restructuring charge in the quarter consisting of $33M of employee termination costs and $15M of other closure costs. Exits were substantially complete by March 31, and most remaining cash payments are expected to conclude by year end. The action was framed as a focus decision rather than a retreat, and management stated at the time that it did not expect a material change to the outlook given in February.

Assessment: a company that has spent five years acquiring its way into international markets choosing, within five months of its largest deal closing, to shrink the map rather than defend marginal geographies is a positive signal on capital discipline. The $48M is real cash and it depressed GAAP earnings, but the correct read is that it bought a permanently smaller loss-making footprint. We would rather see this than a fifth consecutive quarter of "investing in emerging markets."

7. Grocery and retail: volume leadership on an unfinished product

Grocery is the category management is most willing to talk about and least willing to quantify. DoorDash claims volume share leadership in U.S. grocery delivery, says it added more new grocery and retail consumers in Q1 than in any prior quarter, and says roughly one in every two consumers entering the industry for the first time enters through DoorDash. The CEO's framing of the opportunity was notably unsatisfied.

"I actually do stand by the statement that we are miles to go on building an experience that I think can outcompete you going into a grocery store and buying the items yourself because that is still the winning product if you just look at all of the data out there." — Tony Xu, Co-Founder, Chair and CEO

The identified constraint is inventory truth. Grocery supply chains, inventory systems and payment systems do not reliably reconcile, so the platform cannot always know what is actually on a shelf, which produces substitutions, out-of-stocks and cancellations that the in-store experience does not have. Separately, the CFO reiterated that the new-verticals portfolio is on track to be gross-profit positive in the second half.

Assessment: the honesty is the useful part. A CEO claiming share leadership while saying the product cannot yet beat walking into a store is describing a category where the winner is not yet determined by product quality, only by distribution. That cuts both ways: the runway is genuine, and so is the possibility that whoever solves inventory truth first takes it. Gross-profit positive in H2 is the one falsifiable commitment; we will grade it in the Q3 print.

8. DashMart Fulfillment Services: the inventory bet

The company's answer to inventory truth is to take on fulfillment itself, building inventory management and fulfillment operations with grocery and retail partners rather than picking from partners' own stores. It is live with a handful of partners and deliberately unscaled.

"We like what we see with a handful of partners we have, but kind of in the spirit of all things at DoorDash really, where we really want to make sure we nail the experience before we scale it because I think this is both quite disruptive in a positive way to the customer experience. It's also quite disruptive to how retailers are used to working." — Tony Xu, Co-Founder, Chair and CEO

Assessment: this is the most consequential unquantified initiative on the platform and the one that most changes the capital profile of the business. Owning fulfillment converts an asset-light marketplace into something with inventory risk and physical footprint. Management is right to scale it slowly. Investors should note that the grocery profitability case and the DashMart case are the same case, and that neither has a disclosed dollar attached.

9. Advertising: a record with no number

Advertising was described as at a record high and growing faster than in any prior year, with the unlock coming from moving beyond small-business advertisers into large restaurant and retail advertisers and, most recently, consumer-packaged-goods advertisers. Management emphasises a lower ad load than peer platforms and treats advertising economics as interchangeable with any other unit-economics improvement, to be recycled into the business rather than dropped to the bottom line.

"I mean on the ads question, it's never gone better for us in the world of ads. Ads is at a record high, continuing to grow extremely fast compared to any previous year." — Tony Xu, Co-Founder, Chair and CEO

Assessment: advertising is the highest-margin revenue in the model and the only disclosed offset to the take-rate decline, and the company will not size it. In a quarter where Net Revenue Margin fell 30 basis points despite an advertising tailwind, the unstated ads number is the difference between a manageable pricing decision and a structural one. This is the disclosure we would most like to see added.

10. Agentic commerce and the disintermediation question

Two separate questioners pressed on whether AI agents could insert themselves between the marketplace and the consumer, reducing DoorDash to logistics infrastructure. The CEO's answer was historical rather than defensive: large top-of-funnel platforms have repeatedly driven traffic to commerce destinations without retaining the customer, and the durable advantage is a proprietary structured catalog of the physical world that cannot be scraped.

"For at least a decade, you can argue companies like Google or Apple and many other large platforms were top-of-funnel drivers to a lot of different commerce platforms, ours included. … Yet the retention of that traffic was a fraction of what platforms like DoorDash saw." — Tony Xu, Co-Founder, Chair and CEO

Assessment: the analogy is well chosen and the catalog argument is the strongest structural claim management made all call. It is also unfalsifiable on a quarterly cadence. What can be watched is whether the take rate holds: if agents commoditise discovery, the pressure shows up first as fee compression, and fee compression is exactly what this quarter reported. We do not think the two are connected yet. We would revisit that if Net Revenue Margin declines again in Q2.

11. Autonomy: building rather than partnering

DoorDash Dot is live in a small number of markets, and management describes the ambition as an autonomous delivery platform spanning multiple formats on land and in the air, built in-house as well as through partners. The CEO drew a sharp distinction between demonstrating a vehicle and operating a fleet at scale under all conditions, and framed 2026 as the year of climbing that curve across autonomy, hardware, remote operations and city-by-city regulation.

"There's a pretty big difference between just shipping a vehicle or having a vehicle ready for a demonstration and a vehicle that can really operate at scale under any condition and is really battle tested." — Tony Xu, Co-Founder, Chair and CEO

Assessment: this is a multi-year option with no disclosed spend, no market count, no unit economics and no sizing of the addressable delivery volume, and it sits inside an R&D line already carrying the platform rebuild. We assign it no value in the current framework. It is a reason to watch the R&D-to-GOV ratio, which rose eight basis points year over year.

12. AI inside the company, and what it does not yet change

Asked whether AI changes the headcount model, the CEO said yes in principle and then declined to draw the conclusion, describing a company that is measurably more productive without yet knowing how that should reshape team design.

"We're seeing, for instance, a lot of productivity gains right now from AI, about well north of half of our code, as an example, probably closer to 2/3 of our code is written by AI today. But that doesn't alone kind of articulate how workflows and team setup ought to change." — Tony Xu, Co-Founder, Chair and CEO

The CFO's framing was explicitly "do more with more" rather than do the same with less, and he pointed to a near-term operating-expense growth expectation in the 2% range, consistent with prior guidance.

Assessment: two-thirds of code written by AI with headcount philosophy unchanged is the clearest statement available that the productivity gains are being reinvested rather than harvested. That is a defensible choice for a company that believes it is early in a large category. It is also the reason nobody should model AI-driven margin expansion at DoorDash in 2026.

13. Capital allocation: buybacks that do not shrink the count

DoorDash repurchased 1.1 million Class A shares in the quarter at a weighted-average price of $146.93, spending $162M, and had bought back 1.4 million shares for $205M year to date through May 5, leaving approximately $4,795M of the February 2025 $5.0B authorization. Weighted-average diluted shares nonetheless rose 1.6% year over year, to 442.3 million from 435.6 million.

Assessment: at the current pace the buyback is an anti-dilution program that is not fully anti-diluting. Full-year stock-based compensation is guided to $1.3B to $1.4B against a repurchase run rate that annualizes near $650M. The repurchases were executed well, at an average price 14% below the post-print close, and management deserves credit for buying into weakness. The share count still went up.

Guidance & Outlook

The Q2 GOV guide is the headline the stock traded on. Its low end implies 34% growth against the year-ago quarter and its high end 38%, with a 36% midpoint, so the company is guiding to essentially no deceleration in consolidated GOV even as it laps a quarter of Deliveroo ownership for the first time. Against that, the Adjusted EBITDA guide is doing more work than it appears: the $820M midpoint is 2.49% of the $32.9B GOV midpoint, an improvement on Q1's 2.39% but still below the 2.7% posted in Q2 2025, and it carries over $50M of gas relief that the year-ago quarter did not.

Implied second-half ramp: the two guided quarters produce $1,574M of first-half Adjusted EBITDA. The full-year framing requires margin as a percentage of GOV to finish slightly above 2025's level excluding Deliveroo, and the CFO stated he expects both second-half dollars and second-half margins to exceed the first half. Layering the second half at a 25% to 30% step-up over the first, in line with the 23% step DoorDash produced in 2025, points to $1.95B to $2.05B in H2 and $3.5B to $3.6B for the year. That is the number the second half has to deliver, and it has to deliver it while carrying the investments management just pushed out of the first half.

Street at: consensus was $830M for Q2 against an $820M guided midpoint, a gap small enough to be noise but pointed in the wrong direction on the only line where the Street was above the company.

Guidance style: historically DoorDash has exceeded the high end of its GOV range, and management flagged in February that the Q1 plan did not assume that pattern would continue. Q1 GOV landed at $31.60B against a $31.0B to $31.8B range, inside it, exactly as management said it would. That is a meaningful behavioural change: this is a company that has stopped sandbagging its top-line guide, which makes the Q2 range more informative and less of a floor than it used to be.

Analyst Q&A Highlights

DoorDash does not deliver prepared remarks on its call, so the Q&A is the entire management performance. Eleven analysts asked questions across roughly one hour.

Whether AI agents can disintermediate the marketplace

The most persistent theme of the call was whether personal AI agents will sit between consumers and on-demand marketplaces, relegating the platform to fulfillment infrastructure. Management answered with a decade of precedent rather than a product roadmap, arguing that large top-of-funnel platforms have repeatedly delivered traffic without retaining customers, and that the durable moat is a proprietary catalog of physical-world inventory that cannot be scraped from any digital repository.

Q: "As the AI platforms become more capable, there's a concern from investors that like personal agents could layer themselves in between the on-demand marketplaces and the consumer. I would love to know Dash's long-term strategic view on this and if there's a risk to your business to becoming an API or logistics offering to these, and why or why not you'd want to work with one of the third-party AI platforms?"
— Michael Morton, MoffettNathanson

A: "For at least a decade, you can argue companies like Google or Apple and many other large platforms were top-of-funnel drivers to a lot of different commerce platforms, ours included. … And from a traffic perspective, they absolutely could drive a lot more traffic than virtually anyone else could to any one of these restaurants. Yet the retention of that traffic was a fraction of what platforms like DoorDash saw."
— Tony Xu, Co-Founder, Chair and CEO

Assessment: the strongest answer of the call, and the one most dependent on an assertion investors cannot verify. The catalog argument is genuinely differentiated, but the historical analogy assumes agents behave like search boxes rather than like buyers, and that assumption is exactly what is in dispute. Nothing in this quarter tests it either way.

The size and duration of the platform investment

The opening question on cost asked for an update on the depth and duration of the strategic investments, and on how the replatforming changes what the company can build. The answer confirmed the program is tracking to plan on both execution and cost, without restating what that cost is beyond the "several hundred million dollars" figure given two quarters earlier.

Q: "As we get deeper into 2026, any updated views around either the depth or the duration of some of the strategic investments, especially in the platform that we've talked about over the last couple of earnings calls. And more importantly, any updated views on how the tech replatforming might position you for different forms of innovation than you envisioned 6-plus months ago?"
— Eric Sheridan, Goldman Sachs

A: "I talked about the fact that this was the biggest component of the investment that we're making. My view on the overall quantum of dollars that we are investing behind this has stayed the same. It's largely in line with what I had expected 2 quarters ago. And both the program from an execution perspective as well as a cost perspective is going well."
— Ravi Inukonda, Chief Financial Officer

Assessment: "in line with what I expected two quarters ago" is a reassurance about variance, not a disclosure about level. The question asked for depth and duration and received neither. On a program management itself calls the largest component of the investment cycle, this is the gap between a credible plan and an auditable one.

Whether the P&L is carrying three technology stacks simultaneously

A follow-up pressed the mechanical question that the strategic answer had left open: if live production traffic is now running across all three brands on the new platform while the legacy systems remain up, is the income statement absorbing duplicate infrastructure cost, and when does it come out? This produced the only phasing commitment of the call.

Q: "Now that you say that you have live production traffic ramping up across all 3 of the global marketplace brands. Does that mean that you're sort of running all 3 tech stacks now, so we're burdening the P&L with the Max costs and then we should start to turn some of those off in the back half? Or how does this sort of triple platforming down to one platform time line work?"
— Brian Nowak, Morgan Stanley

A: "There is a portion of the spend, which is redundant in the sense that we are going to run all 3 tech stacks in parallel while we're working on the global new tech stack. That is going to phase in. It's going to phase out. My expectation is majority of that will run through '26. Maybe some portion will bleed into early 2027, but that will bleed out."
— Ravi Inukonda, Chief Financial Officer

Assessment: the useful half of the answer is the calendar. Duplicate run costs persist through most of 2026 and taper into early 2027, which caps how much margin recovery is available before the 2027 comparisons. The missing half is the amount, which means the second-half margin inflection the full-year frame depends on cannot be independently modelled.

Gas relief: cost, duration and where the offsets come from

A recurring line of questioning tried to establish whether the roughly $50M quarterly gas relief cost is a run rate, how long it persists, and what gets cut to pay for it. Management gave the Q1 and Q2 figures, confirmed the offsets come from deferring first-half investments into the second half, declined to commit on extension, and anchored back on the unchanged full-year margin frame.

Q: "Ravi, thank you for quantifying the support to drivers for Q2. I think you said $50 million. Obviously, we don't know how long this thing is going to last. But is $50 million a good run rate to assume going out for the rest of the year, just assuming we have status quo on the macro environment?"
— Multiple analysts incl. Youssef Squali, Truist; Josh Beck, Raymond James

A: "To your second point around, are we going to extend it? Look, we've not made any decision. Obviously, we'll monitor the situation very closely, and we'll do what's right for the business. But that said, my broader view on EBITDA for the full year has not changed. … If we do decide to extend the gas rewards program, we'll find offsets in other parts of the business in order to make sure we still feel good from a top line as well as a bottom line perspective."
— Ravi Inukonda, Chief Financial Officer

Assessment: the commitment that matters is that the full-year bottom line is being defended regardless of what happens to gas prices. That is a promise to find offsets, which is a promise to cut something. Investors who like the full-year margin frame should understand they are also buying an option, held by management, to reduce growth investment in order to protect it.

Order-growth deceleration and what is in the Q2 GOV guide

The one exchange that put the deceleration directly to management drew a quantified weather effect and then a pivot to unquantified input metrics, closing with a forward statement that the current quarter had begun well.

Q: "And Ravi, if I could just follow up on, I guess, the order growth dynamics in Q1 as well. Could you just elaborate a bit on the deceleration there? Is that just weather? Or are there other things you want to call out? And how are you thinking about that as you think about the Q2 guidance you've given for GOV?"
— Nikhil Devnani, Bernstein

A: "The impact purely from a winter storm perspective is roughly about 1% on a year-over-year growth perspective from a GOV basis. … We've talked about MAUs reaching an all-time high. Order frequency is growing. Subscription had a record quarter, in fact, across the board across DoorDash, Deliveroo as well as Wolt. … Q2 is off to a good start."
— Ravi Inukonda, Chief Financial Officer

Assessment: the answer is honest about weather and evasive about the rest. One point of a five-point organic order deceleration is accounted for. The rest is answered with metrics the company does not publish, which is not an answer an investor can hold anyone to. The forward "off to a good start" is the more useful signal, and it is consistent with a Q2 GOV guide that assumes no consolidated deceleration.

European competitive position during a platform rebuild

A questioner put the two risks together directly: with a competitor turning more aggressive in Northern Europe, does the disruption of a full technology replatforming leave DoorDash exposed in exactly the markets it just paid up to enter? Management rejected the premise on both halves, first on outcome metrics and then, more revealingly, by arguing outcome metrics are not what it watches.

Q: "Can you maybe talk a little bit about what you're seeing in Europe, in particular, maybe Northern Europe with Uber becoming a little more aggressive. There is a line of thinking out there that maybe as you guys are going through your replatforming, it may make you potentially a little more vulnerable to competition."
— Youssef Squali, Truist

A: "I mean we've never been stronger is the short answer in Europe. … Wolt has seen the highest share performance in each one of the countries in which we operate. But I think those are just outcome metrics. And candidly, they're not things that I stare at all the time."
— Tony Xu, Co-Founder, Chair and CEO

Assessment: a confident answer that nonetheless concedes the frame. Share performance is described as strong and then dismissed as the wrong thing to watch, which is a rhetorically convenient position for a company that publishes no European share data. The four country exits announced eight weeks earlier are the counterweight the answer did not mention.

Grocery unit economics and the path past gross-profit breakeven

Questioners returned twice to what has to happen for the non-restaurant portfolio to become a profit contributor rather than a growth story, with one framing that advertising might be the necessary lever. Management rejected the premise that any single revenue line has to carry the category.

Q: "To get to the next milestone, what are going to be some of the really important elements? I mean, generically, it seems like within new verticals, advertising is a bit more of a weighting factor there. So just kind of curious how we think about maybe some of the real important drivers beyond where we're scaling into the second half."
— Josh Beck, Raymond James

A: "We're not trying to rely on any one source of revenue like ads, for example, to make grocery profitable. We don't need to. We fundamentally have created, we believe, a lower cost structure that allows us to make grocery delivery profitable. But it's just not good enough yet."
— Tony Xu, Co-Founder, Chair and CEO

Assessment: management is claiming the category is structurally profitable at scale and is currently choosing growth over harvest, which is the answer an investor should want. The falsifiable piece is the reiterated commitment that new verticals turn gross-profit positive in the second half of 2026. That is now the single most gradeable promise on the platform.

Autonomous delivery: how much of the market it can serve

An analyst asked two specific questions about the autonomy program: what share of U.S. delivery volume is addressable by autonomous vehicles, and how the cost-to-serve and consumer-behaviour economics work when a customer has to come out to meet a robot. Neither was answered.

Q: "As you've been operating Dot for a bit now in some cities. Are you willing to share any learnings about what percentage of the U.S. delivery market you think is addressable for AVs and then maybe thoughts on how to incentivize consumers to come out and meet the Dot or where the opportunity costs are around cost to serve with AV?"
— Michael Morton, MoffettNathanson

A: "Look, we are early on this journey. We are scaling. And what we are trying to do is obviously operate at scale, manufacture at scale. That's going to be important for us. We've seen good results. We've launched it in a couple of markets. … It's going to help us with speed. It's going to help us with quality. It's going to potentially help us with overall range of deliveries."
— Ravi Inukonda, Chief Financial Officer

Assessment: a direct request for a market-sizing number and a cost-to-serve framework returned neither, and the follow-up from the CEO addressed durability rather than economics. The program may well be valuable; it is currently unvalued and unvaluable from outside. Treat the R&D line accordingly.

Advertising inside and outside the platform

The final question of the call asked whether advertising can extend onto third-party agentic surfaces, which would be the natural monetization path if agents do take share of discovery. The answer separated the two cases cleanly and deferred the second entirely.

Q: "I just want to follow-up on advertising. How do you think about integrating that with agentic capabilities on your own platform? And is there any way you could generate ad revenues on agentic platforms on other platforms?"
— Justin Post, Bank of America

A: "With respect to agentic commerce, I mean, that's just one way of shopping. And so I don't think it will change in terms of our ability to advertise. It may increase some of the opportunities in service areas, but I think a lot of that remains to be seen … I think with respect to what happens with ads on third-party agentic sites, I think you'll have to ask them."
— Tony Xu, Co-Founder, Chair and CEO

Assessment: the deflection is the content. If discovery migrates to third-party agents, the ad business travels only if those platforms permit it, and management has no view on whether they will. That is an honest answer and a reminder that the advertising line, the only disclosed offset to a falling take rate, is the part of the model most exposed to a platform shift.

What They're NOT Saying

1. The size of the replatforming spend. The largest identified investment of the cycle is described as the biggest component of "several hundred million dollars," and the duplicate three-stack run cost is given a phase-out calendar with no amount. Neither the release nor the quarterly filing quantifies it. Without that figure the second-half margin inflection cannot be modelled from outside, only taken on trust.
2. Any advertising revenue disclosure. Advertising is described as at a record high and is one of only two named offsets to the take-rate decline, and it appears nowhere as a number in the release or the filing. In a quarter defined by a 30 basis point Net Revenue Margin decline, this omission is doing real work.
3. Monthly active user and membership counts. The release cites "record membership signups" and "a new high for monthly active users" and publishes no level for either. The quarterly filing's key-metrics table carries Total Orders, GOV, Net Revenue Margin and margin ratios, and no user or member count at all. Every claim about engagement is therefore unauditable.
4. Deliveroo's standalone GOV and orders. The company reconciles revenue excluding Deliveroo and gives ex-Deliveroo growth rates for GOV and orders, but publishes no Deliveroo GOV or order dollars. That is enough to compute organic growth and not enough to assess what was bought.
5. How long gas relief runs. Asked directly whether $50M per quarter is the run rate for the rest of the year, management said no decision has been made. The Q2 cost is disclosed as "over $50 million" and the exposure beyond that is open-ended, hedged only by a commitment to find offsets.
6. Which investments were pushed from the first half into the second. Management confirmed deferring first-half spending to fund gas relief and named nothing. That matters because the same second half has to absorb the deferred spend and deliver the margin inflection.
7. The bridge to the full-year stock-based compensation guide. Q1 stock-based compensation expense was $231M, slightly below the $235M of a year ago, which annualizes near $0.9B against a maintained full-year guide of $1.3B to $1.4B. Neither the release nor the call addresses the implied step-up across the remaining three quarters.
8. Any addressable-market or unit-economic framing for autonomy. A direct question on what share of U.S. delivery volume is addressable by autonomous vehicles, and on cost to serve, went unanswered. The program has no disclosed spend, no market count and no economics.
9. The size of the California payroll-tax accrual. The filing states that an accrual has been recorded within accrued expenses for the state Employment Development Department assessment on Dasher classification, and does not disclose the amount. Worker-classification exposure is the standing tail risk in this business model, and its only quantified expression is invisible.
10. Full-year revenue or GOV guidance. DoorDash guides one quarter at a time on GOV and Adjusted EBITDA, plus a directional full-year margin frame. That is its long-standing practice rather than a change, but it means the annual figure everyone is underwriting is entirely the analyst's own construction.

Market Reaction

• Pre-print setup: DASH closed at $167.97 on May 6, down 25.8% year to date against an S&P 500 up 7.6%, down 11.6% over trailing twelve months, but up 8.3% over the trailing thirty days. The 52-week closing range entering the print was $146.60 to $281.74, so the stock came in nearer the low than the high, roughly 40% below its own twelve-month peak.
• After-hours move: shares rose roughly 10% to 12% in the after-hours session on the release and the Q2 GOV guide, before the call.
• Next-day session: the stock opened at $181.82, an 8.2% gap up, traded as high as $183.50 (+9.2%) and as low as $165.51 (-1.5%), and closed at $171.35, up 2.0% or $3.38. Volume of 12.0 million shares was 2.9 times the 30-day average.
• Peer and market context: the S&P 500 fell 0.4% on the same session. Across the delivery complex, one large peer fell 3.1% and another fell 5.4%, while a third rose 0.5%.

The shape of the session matters more than the close. A stock that gaps 8% and finishes 2% has given back three quarters of its reaction, and it did so while the intraday low printed below the prior close. Whatever the market liked at 4:05 PM on May 6, it liked considerably less by 4:00 PM on May 7. The gap up priced the GOV guide. The fade priced everything under it.

The specific reversal candidates are visible in the numbers. The revenue and order misses were the first data available and did not stop the after-hours rally, which suggests the initial bid was a guidance bid rather than a print bid. What arrives between the release and the next day's close is the call, and the call is where management confirmed duplicate technology costs running through 2026 and into 2027, declined to size them, declined to commit on gas relief duration, and confirmed that first-half investment was deferred into the second half. Every one of those is a reason to mark down the second-half margin inflection that the full-year frame requires, and marking down the second half is precisely what turns a 20x forward multiple from cheap into fair.

Positioning amplified it. A stock down 26% year to date with an 8% thirty-day bounce into the print is a crowded short-cover setup, and the 2.9x volume against a violent intraday range is more consistent with a positioning unwind than with fresh accumulation. Note also that the complex was broadly lower on the session while DASH held a gain, so the fade was not a sector move: relative to peers, DoorDash outperformed on a day the group was sold. The 2.0% close is the market's considered answer, not a rejection.

Street Perspective

Debate: Is the take-rate decline a strategy or a symptom?

Bull view: the bull case being made on the Street holds that lowering consumer fees is a deliberate affordability lever that buys occasions, frequency and share, and that the company has repeatedly demonstrated it can recover margin through advertising, ad-load discipline and Dasher-cost efficiency once volume is captured. Contribution margin still expanded 60 basis points year over year, which is the sign a pricing investment is being funded from real efficiency rather than from the bottom line.

Bear view: the bear camp contends that fee reductions in a two-player domestic market are what competitive pressure looks like when neither side will say so, and that Net Revenue Margin at a five-quarter low alongside a five-point organic order deceleration is the signature of a category slowing, not of a share war being won. On this read the affordability framing is post-hoc.

Our take: the bulls have the better of it for now, because sales and marketing as a share of GOV improved 14 basis points in the same quarter, and a company fighting a price war does not simultaneously spend less on customer acquisition per unit. But the argument has an expiry date. If Q2 shows Net Revenue Margin down again and organic order growth below 16%, the bear reading becomes the base case.

Debate: Does the second-half margin inflection arrive?

Bull view: some sell-side desks argue the full-year frame is conservative given the company just absorbed $50M of unbudgeted gas cost, roughly $20M of storm impact and a $48M restructuring charge and still landed Adjusted EBITDA in the upper half of its range. Duplicate platform costs peak in 2026 and taper, Deliveroo's contribution steps up through the year toward the $200M commitment, and new verticals turn gross-profit positive in the second half. All three unwind in the same direction.

Bear view: a growing consensus view holds that the first half was flattered by deferring investment into the second half, so the second-half bar is higher than the seasonal pattern implies, and that a company unwilling to size either the platform spend or the gas-relief duration is a company that has kept the option to move the goalposts. The Q2 EBITDA guide midpoint below consensus, in a quarter that management said would show sequential margin improvement, is an early sign.

Our take: we side with the bears on timing and the bulls on direction. The inflection is real and mechanical, but the deferred first-half spend plus an open-ended gas-relief cost make the Q3 print, not the Q2 print, the first honest test. That is two quarters of waiting for a stock trading at roughly 20 times a full-year number that the company has not itself given.

Debate: Is agentic commerce an existential threat or a distribution question?

Bull view: the optimistic camp accepts management's framing that agents are a top-of-funnel channel like search and app stores before them, that fulfillment quality determines retention, and that DoorDash's proprietary physical-world catalog is the asset agents cannot replicate and will have to route through. On this view agents expand the addressable market by lowering ordering friction.

Bear view: the skeptics argue that agents differ from search because they transact rather than refer, that a consumer who never sees a marketplace interface cannot be retained by it, and that a platform reduced to price-and-speed comparison in an agent's ranking loses both its advertising business and its pricing power at once. The fact that the take rate is already falling is offered as evidence the erosion has started.

Our take: the bear case is theoretically stronger and empirically unsupported at this stage, and we do not think this quarter's fee reduction has anything to do with agents. The catalog argument is the right defense and it is genuinely hard to replicate. The watch item is not agent adoption, it is whether advertising revenue can be shown to be growing faster than the take rate is falling. Until the company discloses an advertising number, that debate cannot be settled with data, which is itself a reason for the multiple to stay compressed.

Model Update Needed

Valuation. At the May 7 close of $171.35 and 442.3 million diluted shares, the diluted market capitalization is $75.8B. Cash and equivalents of $4,575M, short-term investments of $958M and long-term investments of $849M against $2,725M of convertible notes leave $3.66B of net cash, for an enterprise value of $72.1B. That is 24.5 times trailing-twelve-month Adjusted EBITDA of $2,943M, 41 times trailing free cash flow of $1,752M, and roughly 20 times our FY2026 Adjusted EBITDA estimate of $3.55B.

Valuation impact: we establish a fair-value range of $177 to $189, built as 21 to 22.5 times FY2026E Adjusted EBITDA of $3.55B plus $3.66B of net cash across 442 million diluted shares. The $183 midpoint implies roughly 7% upside from the close. That is a market-like return for a business carrying an un-quantified 2026 cost stack, which is what a Hold looks like in arithmetic. We would move the multiple toward 24 times, and the range toward $200, on evidence that second-half Adjusted EBITDA margin is inflecting and that Net Revenue Margin has stopped falling.

Thesis Scorecard Post-Earnings

This is our first published assessment of DoorDash, so the pillars below establish the thesis rather than grade a standing one. Each will be carried forward and scored in the same form every quarter.

Overall: thesis established with the demand pillar confirmed and the margin pillar challenged. The unusual feature of this quarter is that the two are not in tension in the way they usually are: the cost base behaved well and the revenue per unit did not. That makes the take rate, not spending discipline, the variable that determines whether this business earns a premium multiple.

Action: hold. Own it for the category position and the de-rated multiple; do not add until the second-half margin inflection appears in reported numbers and Net Revenue Margin stabilises. The Q3 print, when the deferred first-half investment lands and the new-verticals gross-profit commitment comes due, is the honest test.