DoorDash Inc (DASH) 2025 Q4 法說會逐字稿

Hello, everyone. Thank you for joining us, and welcome to the DoorDash Q4 2025 earnings call. (Operator Instructions) I will now hand the call over to Weston Twigg. Please go ahead.

大家好。感謝您的參與，歡迎參加 DoorDash 2025 年第四季財報電話會議。（操作員指示）我現在將把電話交給韋斯頓·特威格。請繼續。

All right. Thanks, Elizabeth. Good afternoon, everyone, and thanks for joining us for our Q4 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, profitability, our guidance, strategies, capital allocation approach and the broader economic environment.

好的。謝謝你，伊麗莎白。各位下午好，感謝各位參加我們2025年第四季財報電話會議。今天我很高興能與聯合創始人、董事長兼首席執行官徐東尼先生和首席財務官拉維·伊努孔達先生一起出席。在今天的電話會議上，我們將發表一些前瞻性聲明，包括但不限於我們對公司業務、財務狀況、經營業績、盈利能力、業績指引、策略、資本配置方法以及更廣泛的經濟環境的預期。

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.

前瞻性陳述存在風險和不確定性，可能導致實際結果與所述結果有重大差異。這些不確定因素中的許多都在我們的美國證券交易委員會文件中有所描述，包括我們最新的 10-K 表格和 10-Q 表格。您不應將前瞻性陳述視為對未來事件或績效的預測。除法律要求外，我們不承擔更新任何前瞻性聲明的義務。

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.

在本次電話會議中，我們將討論一些非GAAP財務指標。有關我們非公認會計準則財務指標的信息，包括此類非公認會計準則指標與最直接可比較的公認會計準則財務指標的調節表，可在我們的收益報告中查閱，該報告可在我們的投資者關係網站 ir.doordash.com 上找到。

A 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.

非GAAP指標應與我們的GAAP績效一併考慮，並非旨在取代我們的GAAP績效。最後，本次電話會議將在我們的投資者關係網站上進行音訊網路直播。通話結束後不久，我們將在網站上提供通話錄音回放。

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

接線員，我把電話轉回給你，我們可以開始回答第一個問題了。

(Operator Instructions) Shweta Khajuria, Wolfe Research.

（操作說明）Shweta Khajuria，Wolfe Research。

Thanks a lot for taking my questions. Let me try two, please. One is on competitive intensity globally and specifically in Europe, if you're seeing anything different, and if so, it hasn't been rising.

非常感謝您回答我的問題。請讓我試試兩個。一是全球，特別是歐洲的競爭強度，如果你看到任何不同之處，那就是競爭強度並沒有上升。

And then second is on investments. This may be for Ravi, how should we be thinking, there's a lot of maybe debate that's going on in terms of the level of investments not only in 2026, but to the degree that it may continue into 2027, especially as it relates to tech replatform. So could you please, Ravi, help us think through how we should think about it? Is it a onetime in '26 or should we expect some elevated level of spend in '27 too because it could bleed into next year?

其次是關於投資。這可能是拉維的想法，我們該如何思考？目前有很多關於投資水平的爭論，不僅是 2026 年的投資水平，而且這種爭論可能會持續到 2027 年，尤其是在技術平台重構方面。拉維，你能幫我們理清一下該如何看待這件事嗎？這種情況是 2026 年的一次性事件，還是應該預期 2027 年的支出水準也會有所上升，因為這種情況可能會延續到明年？

Thanks a lot.

多謝。

Yeah, hey, Shweta, I'll start by talking about Europe. Overall, we feel really great about our position in Europe. We're the leading player in many of the countries in that continent. And we're off to a really great start with our Deliveroo acquisition.

好的，嗨，Shweta，我先從談談歐洲開始吧。總的來說，我們對自身在歐洲的地位感到非常滿意。我們在非洲大陸的許多國家都是行業領導者。我們收購 Deliveroo 的開局非常順利。

With respect to Deliveroo, we are growing much faster at the same profit contribution that we expected before the acquisition. We're gaining share in its largest markets. We're doing the same on the Wolt side. And so when I overall look at our business outside of the US, what I see is faster growth than what we see in the US, which, by the way, the US had two of the fastest-growing quarters in 2025 in the last four years. So to trump that is actually quite impressive. And we're continuing to improve our unit economics across the board. So I feel really strong about our position overseas.

就 Deliveroo 而言，我們的成長速度遠超預期，利潤貢獻也達到了收購前的預期水準。我們在其最大的市場中不斷擴大市場份額。我們在Wolt這邊也採取了同樣的措施。因此，當我整體審視我們在美國以外的業務時，我發現其成長速度比在美國更快。順便一提，在過去的四年裡，美國在 2025 年經歷了兩個成長最快的季度。所以能勝過那項成就，確實相當令人印象深刻。我們正在持續提升各方面的單位經濟效益。所以我對我們在海外的地位感到非常有信心。

Shweta, on your second point, right, I'll make a couple of points. One, my expectation for the full year EBITDA for '26 has not changed since the last call. The way I think about it is 2026 EBITDA margin is going to be up slightly compared to 2025, excluding ROO and ROO to produce about $200 million of EBITDA, like we said. So that remains very consistent.

Shweta，關於你的第二點，好的，我再補充幾點。第一，我對 2026 年全年 EBITDA 的預期與上次電話會議以來沒有改變。我認為，2026 年的 EBITDA 利潤率將比 2025 年略有上升，不包括 ROO 和 ROO，從而產生約 2 億美元的 EBITDA，正如我們所說。所以這一點始終非常一致。

Look, as we're thinking about investing, there's three major part of investing that we called out. The quantum of investment dollars is also very similar to what I had expected at the time of the last call. One of the major areas of spend, as you called out, is our global tech stack.

你看，當我們思考投資時，我們提到了投資的三個主要面向。投資金額也與我上次通話時的預期非常接近。正如您所指出的，我們的全球技術堆疊是主要的支出領域之一。

Look, we're happy with the progress. We're making good progress there. There are components of that spend that are redundant, especially as we take on the cost of running both tech stacks in parallel. Majority of that should be in '26, some of that will be in '27, which will come off, but that's a smaller component of the overall spend. The other two areas are -- both autonomy as well as merchant services. We are expanding both of those areas, we are investing. We like what we are seeing there.

你看，我們對目前的進展很滿意。我們在那方面進展順利。這項支出中有些部分是多餘的，尤其是當我們同時承擔運行兩種技術堆疊的成本時。大部分支出應該在 2026 年完成，部分支出將在 2027 年完成，這部分支出將會減少，但這只佔總支出的一小部分。另外兩個領域分別是－自主權和商家服務。我們正在拓展這兩個領域，我們正在進行投資。我們很喜歡那裡看到的景象。

And in fact, I mean, if we continue to make more progress from a customer benefit perspective, our goal is to continue to invest more. Look, I mean, our goal has always been to maximize long-term free cash flow. We believe these investments are the right investments, especially as we think about becoming the operating system for local commerce. As we make more progress, we'll continue to invest behind them because ultimately, they lead to more areas where we can invest behind as well as improve overall free cash flow generation.

事實上，我的意思是，如果我們從客戶利益的角度來看繼續取得更多進展，我們的目標是繼續增加投資。我的意思是，我們的目標始終是最大限度地提高長期自由現金流。我們相信這些投資是正確的投資，尤其是當我們考慮成為本地商業作業系統的時候。隨著我們取得更多進展，我們將繼續增加對這些領域的投資，因為最終，這些進展會帶來更多我們可以投資的領域，並改善整體自由現金流的產生。

Okay, that's very helpful. Thanks, Tony. Thanks, Ravi

好的，這很有幫助。謝謝你，托尼。謝謝，拉維

Michael Morton, MoffettNathanson.

Michael Morton，MoffettNathanson。

Hey, good evening. Thanks for the questions. Maybe one for Tony to start. In the press release, you talked about investing to support growth in longer distance and higher effort deliveries. I was wondering if you could provide some more details on what type of deliveries those are related to? And just following up on that, like Dash has made a lot of investments in DashMart and DashLink.

嘿，晚上好。謝謝大家的提問。或許可以先讓托尼來開局。在新聞稿中，你們談到了投資以支持長途和高投入交付業務的成長。我想請問您能否提供更多關於這些配送類型的資訊？另外，Dash 對 DashMart 和 DashLink 進行了大量投資。

And Tony, I'd love to hear maybe if it's related to that longer distant investment, but how you see the e-commerce landscape evolving? Do you expect to see local inventory worked more into the kind of consumer shopping experiences? And anything around that would be great.

東尼，我很想聽聽你對電子商務格局未來發展的看法，這或許與那項長期投資有關？您是否預期本地庫存會更多地融入消費者的購物體驗中？任何相關的內容都非常棒。

And then maybe a quick one for Ravi. I love the commentary on the inflection in unit economics for your new verticals business. Could you speak a little bit about some of the drivers of that improvement? Is it scale and also may be less investment requirement? Anything in that would be great, too. Thanks.

然後也許可以給拉維快速地來一杯。我非常同意您對新垂直業務單位經濟效益拐點的評論。您能否談談促成這項改進的一些因素？規模是否夠大，所需的投資是否較少？其中任何一項都很好。謝謝。

Yes, I'll start. On the question with respect to Dashers and the evolution around what DoorDash is doing in the world of commerce, I mean, I think Ravi kind of touched nicely about this on the previous question about how we're building the operating system for local commerce. And so I'll maybe take a couple of minutes to expand upon that a little bit, and then I'll answer directly to your question about how Dashers and frankly, some of our other audiences play into this broader ecosystem.

好的，我先來。關於送餐員以及 DoorDash 在商業領域所做事情的演變問題，我認為 Ravi 在先前關於我們如何建立本地商業作業系統的問題中已經很好地談到了這一點。因此，我可能會花幾分鐘時間稍微展開一下，然後我會直接回答你的問題，即 Dashers 以及坦率地說，我們的一些其他受眾是如何融入這個更廣泛的生態系統的。

When I think about like what is going to be required in order to allow all the small, medium and large physical businesses to become omnichannel businesses and compete against one or two large behemoths. It's going to require software. It's going to require warehousing and physical infrastructure. It's going to require the lowest cost of delivery at the highest quality, and it's going to require amazing software.

當我思考要讓所有小型、中型和大型實體企業轉型為全通路企業，並與一兩家大型巨頭企業競爭，需要做些什麼時，我就感到很困惑。這需要用到軟體。這需要倉儲和實體基礎設施。這需要以最低的成本提供最高品質的服務，而且需要非常優秀的軟體。

And if you think about a lot of what we announced actually in 2025 at our Dashboard product event in September as well as the investments we're making into '26, I think they line up nicely to give you a view of what we're building. So if you think about it, we're starting by obviously building software for every small, medium and physical -- and large physical business, whether that means helping them and adding them into the DoorDash app.

如果你仔細想想我們在 9 月的 Dashboard 產品發布會上宣布的 2025 年計劃，以及我們對 2026 年的投資，我認為它們可以很好地讓你了解我們正在建立什麼。所以仔細想想，我們顯然是從為所有小型、中型和實體（以及大型）實體企業建立軟體開始的，無論是幫助他們還是將它們添加到 DoorDash 應用程式中。

One of the things that we've been seeing is just continued growth across all categories at DoorDash. Now around 30% of customers are ordering outside of the restaurant category, especially as we broaden our reach into grocery, retail, where we've become the leading third-party transaction platform in the US, and we're growing very, very fast outside of the US as well in that dimension. So we're bringing software in that dimension. We're bringing software also in the dimension of all of the B2B products we're shipping, where we're helping businesses offer delivery through their own channels. We're helping businesses also stand up their own e-commerce solutions.

我們看到的一個現像是，DoorDash 的所有類別都持續成長。現在約有 30% 的顧客在餐廳以外的類別下單，尤其是在我們將業務拓展到雜貨店和零售領域之後，我們已成為美國領先的第三方交易平台，而且我們在美國以外的這一領域也發展得非常非常快。所以我們正在引入這方面的軟體。我們也把軟體應用到我們推出的所有 B2B 產品中，幫助企業透過自己的管道提供交付服務。我們也幫助企業建立自己的電子商務解決方案。

We're building warehouses to bring inventory closer to where consumers live. So we announced Dashmart Fulfillment Services last September. We're very excited to be partnering with companies like Kroger or CVS to really offer the highest possible quality, the perfect inventory accuracy at the lowest cost and the fastest speeds. So that they can compete and offer same hour, same-day delivery against their peers.

我們正在建造倉庫，以便將庫存更靠近消費者居住地。因此，我們在去年九月推出了 Dashmart 物流服務。我們非常高興能與 Kroger 或 CVS 等公司合作，以最低的成本和最快的速度，真正提供最高品質的服務、完美的庫存準確性。這樣他們才能與同業競爭，提供當小時送達、當日送達服務。

That's something that we're doing. And we're also investing in how to do this in the future, right? And that's really our investment into autonomous vehicles where we announced DoorDash Dot as well as some other projects that we're working on in order to make sure that no physical business will be at a disadvantage when it comes to offering the best possible delivery experience at the lowest possible cost.

這是我們正在做的事情。我們也在投資研究未來如何做到這一點，對吧？這就是我們對自動駕駛汽車的真正投資，我們推出了 DoorDash Dot 以及我們正在進行的其他一些項目，以確保任何實體企業在以盡可能低的成本提供盡可能好的配送體驗方面都不會處於劣勢。

So when you think about how Dashers kind of fit into the system, one of the things that has to be true and that is true, I mean, it's been happening throughout last year, the years before and obviously, this year and into the future is that Dashers will have more choice in terms of the types of orders they can take on. And that's why one of the investments we're making this year is actually directly into Dashers so that as they do more types of grocery retail orders, which can travel longer distances, which are usually more complicated because there's a shopping nature involved, although not always. But because of the greater complexity, we want to make sure that the pay models reflect that. We want to make sure that the app experience actually reflects that.

所以，當你思考送餐員如何融入這個系統時，有一件事是肯定的，而且也確實是肯定的，我的意思是，這種情況在去年、前幾年以及今年和未來都會發生，那就是送餐員在接單類型方面將有更多選擇。因此，我們今年的投資之一實際上是直接投資於 Dashers，以便他們能夠處理更多類型的雜貨零售訂單，這些訂單可能需要更長的配送距離，通常也更複雜，因為其中涉及購物性質，儘管並非總是如此。但由於情況更加複雜，我們希望確保薪酬模式能夠反映出這一點。我們希望確保應用程式體驗能夠真正體現這一點。

And there's a lot that we're also doing in cataloging all of the physical information that exists nowhere on the Internet that we're also partnering with Dashers to do as well. So there's a lot of things that we're building, but we're building all of the fundamental building blocks at the end of the day to enable local commerce. So that for consumers, they can get anything inside the city delivered to them at the best possible experience at the lowest possible price.

我們還在做很多工作，例如整理網路上不存在的所有實體訊息，我們也在與送餐員合作進行這項工作。所以我們正在建造很多東西，但歸根結底，我們正在建立所有基本組成部分，以促進本地商業發展。這樣一來，消費者就可以在市內以盡可能低的價格獲得盡可能好的購物體驗，並享受送貨上門服務。

Hey, Mick, on your broader question around new verticals, right, our retail and grocery business, maybe I'll zoom out and just talk about the performance and what we're seeing in that business. Look, I mean, new verticals had a really strong quarter as well as the year. We are the fastest growing in the US as it relates to third-party peers.

嘿，米克，關於你提出的關於新垂直領域的更廣泛的問題，對吧，我們的零售和雜貨業務，也許我可以跳出這個框架，談談該業務的業績以及我們在該業務中看到的情況。你看，我的意思是，新興垂直領域在本季以及全年都表現得非常強勁。就第三方同業而言，我們是美國成長速度最快的公司。

Look, today, Tony mentioned the fact that 30% of our MAUs in the US order from categories outside of restaurants. And our focus has always been how do you get that 30% to be 100% over time. And what we're seeing is that we improve selection as we are investing behind making the product better, whether it's quality, affordability, more and more of our MAUs continue to order from categories outside of restaurants.

今天，Tony提到，我們美國30%的月活躍用戶從餐廳以外的類別訂購商品。我們一直關注的是如何讓這 30% 隨著時間的推移變成 100%。我們看到，隨著我們增加對產品（無論是品質還是價格）的投入，我們的產品選擇也在不斷改進，越來越多的每月活躍用戶繼續從餐廳以外的類別訂購商品。

In fact, the number of new consumers that join and start their journey with new verticals is also improving on a year-over-year basis. At the same time, we've been focused on improving the efficiency of the business. And look, when I look at the profitability, I mean, the team has made really good progress in unit economics year-over-year. I expect our entire retail and grocery business to be unit economic positive in the second half of the year.

事實上，加入並開始體驗新垂直領域的新消費者數量也逐年增加。同時，我們一直致力於提高業務效率。你看，從獲利能力來看，團隊在單位經濟效益方面取得了非常好的逐年進步。我預計我們整個零售和食品雜貨業務在下半年將實現單位經濟效益為正。

And look, when I look at that, I mean, it's just continued execution, right? It's a combination of steady progress that we're going to make over a bunch of different fundamentals, whether it's scale, density, continue to improve the efficiency on the logistics side, basket sizes are getting larger. There's no one thing, which is a step function change. It's continued execution, finding basis points, largely how we operate our entire business and our primary focus continues to be, hey, how do you actually get the business to scale. You know exactly what we need to do on the unit economic side. How do we get 100% of our MAUs to use grocery and retail. That's the real focus that we've had as a team.

你看，我看那不就是持續執行嗎？我們將在許多基本方面取得穩步進展，無論是規模、密度，還是在物流方面不斷提高效率，以及購物籃容量越來越大。沒有哪一件事是階躍式變化。持續執行、尋找基點，這在很大程度上決定了我們整個業務的運作方式，而我們的主要關注點仍然是，嘿，如何真正實現業務規模化。您非常清楚我們在單位經濟效益方面需要做什麼。如何讓 100% 的月活躍用戶使用食品雜貨和零售服務？這正是我們團隊一直以來真正關注的重點。

Thank you.

謝謝。

Nikhil Devnani, Bernstein.

Nikhil Devnani，Bernstein。

Hi there, thanks for taking the question, Tony, I wanted to ask about agentic commerce and where you see things going longer term. I think today, it's clear that Dash has very strong direct relationships with consumers. And even with where things currently stand with AI chatbot experiences, that feels well intact. But there is a debate around how this might change in an agentic future where search, discovery, transaction flows start to compress and get more automated. You might still be part of the transaction, but perhaps the economics look different is the debate.

你好，托尼，感謝你回答這個問題。我想問一下關於代理型商務以及你認為它長期發展的方向。我認為，如今很明顯，Dash 與消費者之間建立了非常牢固的直接關係。即使在人工智慧聊天機器人體驗目前所處的階段，這一點仍然保持完好。但對於在智能體主導的未來，搜尋、發現和交易流程開始壓縮並變得更加自動化，這種情況可能會如何變化，目前存在爭議。你或許仍然是交易的一部分，但爭論的焦點在於，其中的經濟效益可能有所不同。

So how do you think about positioning DoorDash for that future as consumer behavior evolves. Do you want to integrate with third parties? Does it make more sense to double down on your own vertical solutions maybe you disagree with that altogether and think AI -- Agentic AI's upside as a broad premise. So just curious for any thoughts you have on that? Thank you.

那麼，隨著消費者行為的演變，您認為應該如何定位 DoorDash 以適應未來的發展？您是否希望與第三方整合？加倍投入自己的垂直解決方案是否更有意義？或許你完全不同意這種觀點，認為人工智慧——智能體人工智慧的優勢在於其廣泛的前提。所以，我很好奇你對此有什麼看法？謝謝。

Sure. No, I think it's a great question. And the first thing I always come to when it comes to any new technology, whether it's autonomous vehicles or agentic commerce, and kind of how that interacts with LLMs, it's really how well does it solve the end-to-end job for a customer, kind of becomes the lens in which I approach all of these things.

當然。不，我覺得這是一個很好的問題。每當我接觸到任何新技術，無論是自動駕駛汽車還是智慧商務，以及它們如何與生命週期管理 (LLM) 互動時，我首先想到的總是：它能否很好地為客戶解決端到端的問題？這逐漸成為我看待所有這些問題的視角。

And I actually think that maybe the best way to think about this is to look a bit through history and maybe offer you a couple of examples and then maybe work our way towards the present and think about the future. So when I think about these terrific products, whichever chat assistance that you love using or just protocols with agentic commerce that you like using, I kind of view them very much as almost like the new forms of the Googles or other large kind of top-of-funnel channels, if you will.

其實我覺得思考這個問題的最好方法是回顧歷史，舉幾個例子，然後慢慢過渡到現在，再展望未來。所以，當我想到這些優秀的產品時，無論你喜歡使用哪種聊天輔助工具，或者你喜歡使用哪種代理商務協議，我都會把它們看作是谷歌或其他大型漏斗頂端渠道的新形式。

And if you think about -- I'll give you a couple of examples from history that kind of maybe touch on your question. One is, look at what happened to product search over the 2010s. You saw over time, companies like Amazon take increasing share of product search from traditional search engines because they did the end-to-end job for customers because searching to buy an item is only one task, but perhaps reading reviews or tracking the package or returning the package or any other form of customer support are also part of the end-to-end job that you have to solve for customers.

如果你仔細想想──我舉幾個歷史上的例子，或許能解答你的問題。其一，看看 2010 年代產品搜尋發生了什麼變化。隨著時間的推移，像亞馬遜這樣的公司從傳統搜尋引擎手中奪取了越來越多的產品搜尋份額，因為它們為客戶完成了端到端的工作。搜尋購買商品只是其中一項任務，但閱讀評論、追蹤包裹、退貨或任何其他形式的客戶支援也是必須為客戶解決的端到端工作的一部分。

And over that decade, you saw, I think, companies like Amazon and others take increasing share when it comes to something like product search. Another example that comes to mind and maybe closer to home, is something that actually Google launched was called Google Food Ordering, which they launched in, I believe it was like 2016 or something like this, where they offer the ability for restaurants to offer delivery directly through Google Maps and other Google surfaces.

在過去的十年裡，我認為，像亞馬遜這樣的公司和其他公司在產品搜尋等領域佔據了越來越大的市場份額。另一個例子，可能更貼近我們的生活，是谷歌推出的一項名為「Google Food Ordering」的服務，我記得好像是在 2016 年左右推出的，這項服務讓餐廳能夠直接透過谷歌地圖和其他谷歌平台提供送餐服務。

And look, they drove the ton of traffic, multiple fold traffic of what DoorDash could generate to these restaurants. But when you looked at the retention and the frequency of use of that channel versus companies like DoorDash, it really was a fraction of what DoorDash could provide.

你看，他們為這些餐廳帶來了大量的流量，是 DoorDash 能帶來的流量的數倍。但當你比較該管道的用戶留存率和使用頻率與 DoorDash 等公司時，你會發現它提供的服務與 DoorDash 相比簡直是天壤之別。

Why is that? Well, because after a checkout, things can happen in the physical world. For example, a driver might be late, or an item might be missing or some substitution on a spoiled carton of milk needs to be made or not the exact brand of whatever produce that a customer was looking for needs to be changed.

這是為什麼？因為結帳之後，現實世界可能會發生一些事情。例如，司機可能會遲到，或者物品可能會丟失，或者需要用其他物品替換變質的牛奶，或者需要更換顧客想要的某種農產品的品牌。

So the end-to-end job at the end of the day, I think, is how customers will ultimately judge where they do their shopping. And at the end of the day, wherever the customers keep going back to, that's where the audiences will flow and where the audiences flow so will the advertiser budgets as well as the interest along that dimension.

所以，我認為，歸根結底，最終決定顧客選擇在哪裡購物的是整個流程。歸根究底，顧客不斷光顧的地方，就是受眾流向的地方；受眾流向哪裡，廣告商的預算以及對這方面的興趣也會隨之而來。

And so when I put all this into perspective, the historical context into where I look at DoorDash today, I think, DoorDash is really well positioned because we're actually solving the end-to-end job for a customer, which is to get them some item, brought to them in the condition they expect, on time, every time. That's actually really hard to do. You got to map the physical world, all of which that information does not exist anywhere on the Internet. That's data that DoorDash has to collect in a proprietary way.

所以，當我把所有這些放在一起，結合歷史背景來看今天的 DoorDash 時，我認為 DoorDash 的處境非常好，因為我們實際上為客戶解決了端到端的問題，那就是每次都能按時將他們期望的商品送到他們手中。這其實很難做到。你需要繪製出物理世界的地圖，而這些資訊在網路上根本不存在。那是 DoorDash 必須以專有方式收集的數據。

You have to actually be excellent at the execution of the operations. You have to be excellent at collecting all the metadata as well for all of these different items as well as the personalization you can perform if you actually have all of the customers and all of the customer information.

你必須真正精通操作的執行。你不僅要擅長收集所有這些不同項目的元數據，還要擅長進行個人化設定（如果你擁有所有客戶及其所有資訊）。

And I think when you put it all together, we're going to be the best place to solve the end-to-end job for customers. And so long as we are that best place, we will also attract all of the audience and all of the advertiser dollars that comes from that audience.

我認為綜合所有因素來看，我們將成為為客戶解決端到端問題的最佳選擇。只要我們保持最佳狀態，我們就能吸引所有受眾以及來自這些受眾的所有廣告商的資金。

But look, I think with respect to how that informs our partnerships with some of these AI systems, I view them as channel partners, and we'll see how much traffic they can drive in a very similar way to how companies like Facebook and Google did the same for DoorDash in the past.

但是，就這如何影響我們與某些人工智慧系統的合作而言，我認為它們就像通路合作夥伴，我們將看看它們能帶來多少流量，這與 Facebook 和 Google 等公司過去為 DoorDash 所做的非常相似。

Thanks, Tony.

謝謝你，托尼。

Deepak Mathivanan, Cantor Fitzgerald.

迪帕克·馬蒂瓦南，坎托·菲茨杰拉德。

Great, thanks for taking the questions. One for Tony and one for Ravi. Tony, can you talk about the strategy with the autonomous delivery platform. What do you think this platform looks like in two to three years, perhaps between first-party efforts with Dot and maybe through -- and also with some of the other third-party partnerships.

太好了，謝謝你回答這些問題。一個給托尼，一個給拉維。Tony，可以談談你們與自主配送平台的合作策略嗎？你認為這個平台在兩到三年後會是什麼樣子？也許是透過與 Dot 的第一方努力，也許是透過——也可能是透過一些其他的第三方合作夥伴關係。

And then for Ravi, can you elaborate on the reasons why the unit economics improvement in the core US restaurant business will be lower this year than prior years. Is that from maybe slower advertising growth due to scale or perhaps moderation and some efficiency gains or maybe reinvestments. Can you expand on that, if you don't mind? Thanks so much.

那麼，Ravi，你能否詳細說明一下，為什麼今年美國核心餐飲業務的單位經濟效益改善幅度會低於往年？這可能是由於規模擴大導致廣告成長放緩，也可能是因為廣告投放趨於溫和、效率提升或再投資所致。如果您不介意的話，能否詳細解釋一下？非常感謝。

Yes. Deepak, on autonomy you're absolutely right. That the autonomous delivery platform is probably the most valuable part of what we're building. Because the way I kind of view it is that in the future, there will be a collection of different vehicles, a fleet of different vehicles, both by land and by air, some of which we'll build inside our four walls and others of which we will partner. And I think the most important part is actually playing the orchestration of all of the activity and movement and the handoffs.

是的。Deepak，關於自主性，你的觀點完全正確。自主配送平台可能是我們正在建構的產品中最有價值的部分。因為我的想法是，未來將會有各種各樣的交通工具，一支由各種交通工具組成的車隊，包括陸地交通工具和空中交通工具，其中一些我們將在公司內部製造，而另一些我們將與合作夥伴共同製造。我認為最重要的部分實際上是協調所有的活動、動作和交接。

Because it's not going to be immediately or at least not immediately obvious to me that autonomous vehicles are going to perform every single type of delivery. There are certain times where you're going to see handoffs between Dashers and AVs.

因為自動駕駛汽車是否會執行每種類型的送貨任務，這一點不會立即顯現，至少對我來說不會立即顯現。在某些時候，你會看到送餐員和自動駕駛汽車之間的交接。

At other times, you're going to see AVs perform deliveries that Dashers don't want to do. At other times, you're going to see Dashers perform deliveries that AVs are not well equipped to do. And so it's really the kind of goal for '26 then is really to figure out what are all of these different use cases. Where can we apply the most pragmatic business impact and customer impact immediately. We're doing that right now in a couple of different markets. So we actually have real live deliveries happening right now with AVs and we're very excited about the future.

有時，你會看到自動駕駛汽車執行送餐員不想做的配送任務。有時，你會看到送餐員執行自動駕駛汽車不太適合執行的配送任務。所以，2026 年的真正目標是弄清楚所有這些不同的使用案例是什麼。我們可以在哪裡立即應用最務實的商業影響和客戶影響？我們目前正在幾個不同的市場進行這項工作。所以我們現在確實已經在使用自動駕駛汽車進行實際的送貨服務了，我們對未來感到非常興奮。

Hey, Deepak, on the US restaurant point, maybe I'll give you like a slightly broader answer. Look when I look at the performance of the US restaurant business, it continues to be quite strong. In fact, in Q4, the contribution margin for the US restaurant business was up on a year-over-year basis. And I do expect us to continue to improve margins in 2026. When I look at the last few years, I mean, growth has been quite strong.

嘿，迪帕克，關於美國餐廳的問題，或許我可以給你一個更全面的回答。從我觀察美國餐飲業的表現來看，它依然相當強勁。事實上，第四季美國餐飲業務的貢獻毛利年增。我預計我們在 2026 年將繼續提高利潤率。回顧過去幾年，成長動能相當強勁。

In fact, in '25 the restaurant business grew faster at a larger scale compared to 2024. And even when I look at the efficiency that we've driven over the last three or four years, it's been quite good. We've driven improvements in margin. Dashers costs have become more efficient, C&R has become more efficient. We've driven leverage in sales and marketing, like you called out, ads is becoming a larger portion of the overall business. And we're still continuing to invest behind that business, whether it's selection, quality, affordability.

事實上，2025 年餐飲業的成長速度和規模都比 2024 年還要大。即使回顧過去三、四年我們所取得的效率，也相當不錯。我們提高了利潤率。配送員的成本控制更有效率，收貨和回收效率也更高了。正如您所指出的，我們已經在銷售和行銷方面取得了顯著進展，廣告在整個業務中所佔的比例越來越大。我們將繼續加大對這項業務的投入，無論是在產品選擇、品質或價格方面。

As I look ahead, I do expect us to continue to improve margins, albeit it will be at a lower pace compared to prior years. Some of that is going to come from DashPass. As you know, DashPass had a record year as well as a record quarter. DashPass is going to have an impact on margin. But look, overall, the ROI is strong because profit dollar production is going to be high.

展望未來，我預計我們將繼續提高利潤率，儘管成長速度會比往年慢一些。其中一些將來自 DashPass。如您所知，DashPass 的全年業績和季度業績均創下新高。DashPass將會對利潤率產生影響。但總的來說，投資報酬率很高，因為利潤額將會很高。

Largely because it's very simple, subscribers retain more, they order more, which means that they produce more gross profit dollars. Look, our focus has always been on overall profit dollar production, which continues to be quite strong. And when I look at the top line and the bottom line of the restaurant business, both continue to be very healthy.

主要原因是它非常簡單，訂閱用戶留存率更高，訂單量更大，這意味著他們能創造更多的毛利。你看，我們一直以來都專注於整體利潤美元產出，而這項指標仍然相當強勁。當我查看餐飲業的營收和利潤時，發現兩者都保持著非常健康的水平。

Right, thank you so much.

好的，非常感謝。

Youssef Squali, Truist Securities.

Youssef Squali，Truist Securities。

Great. Thanks for taking the questions. Maybe, Tony, just a question on competition again. But maybe from a grocery and perishables perspective. So as Amazon is doubling down on those categories, can you maybe talk a little bit about what you're seeing in terms of your growth within that category in Q4? And have you seen any changes in the competitive landscape so far?

偉大的。謝謝您回答問題。東尼，或許我只是想再問一個關於競爭的問題。但或許可以從食品雜貨和易腐品的角度來看。鑑於亞馬遜正在加倍投入這些品類，您能否談談您在第四季度這些品類中的成長情況？到目前為止，您是否觀察到競爭格局發生了任何變化？

And then, Ravi, on the headwinds to Q1 margins, can you just help us kind of quantify the impact of the higher Dashers costs. And I know there's a seasonal effect there, but is it more seasonal this year than in prior years or is just normal seasonality? Thank you.

那麼，Ravi，關於第一季利潤率面臨的不利因素，你能幫我們量化一下送餐員成本上升的影響嗎？我知道這其中有季節性影響，但今年的季節性影響比往年更明顯，還是只是正常的季節性影響？謝謝。

Yeah, hey. On the first question regarding grocery. In short, no, we haven't seen an impact on our growth. In fact, we continued very high growth rates kind of as fast as we've seen in the grocery sector, not just in Q4, but also for this year as well. And I think if you think about perhaps why we continue to see fast growth. I think there are a few points.

是啊，嘿。關於第一個關於食品雜貨的問題。簡而言之，沒有，我們沒有看到這對我們的成長產生影響。事實上，我們繼續保持著非常高的成長速度，其速度之快，不僅在第四季度，而且在今年全年都是如此，堪稱食品雜貨業的最高增速。我想，或許你可以思考我們為何持續看到快速成長。我認為有幾點需要注意。

The first thing I would say is, if you think about what we're trying to create, we're trying to create a world in which there's the maximal amount of choice for what customers can get delivered. And that includes all of the grocers. There's a reason why there exists tens of thousands of supermarkets in the US, not just because it's a large geography, it's because the average customer does buy from a couple of different places when it comes to their groceries, whether it's buying from place A for their meat and fish, place B for their produce, place C for their pantry items, place D for some specialty items, et cetera.

首先我想說的是，如果你想想我們正在努力創造的是什麼，我們努力創造的是一個讓顧客在選擇配送商品方面擁有最大選擇的世界。這其中也包括所有雜貨店。美國有成千上萬家超市是有原因的，不僅是因為美國地域遼闊，還因為普通顧客在購買食品雜貨時通常會去好幾個不同的地方，比如在 A 地買肉類和魚類，在 B 地買農產品，在 C 地買食品儲藏室用品，在 D 地買一些特色商品等等。

And I think DoorDash is a place in which you can get all of this at the best possible price and the highest quality of delivery. And so that's, I think, our angle at it where we think that in the future, so long as you believe that customers are going to want choice, and I think that can be well reasoned when you just look at the landscape of grocery and how there exists so many grocers out there, which indicates that I think, consumers prefer choice, then that will continue to be very strong interest in the DoorDash product.

我認為 DoorDash 是一個能夠以最優惠的價格和最高的配送品質提供所有這些服務的平台。所以，我認為，這就是我們看待這個問題的角度：我們認為，未來只要你相信顧客會想要選擇，而這一點從雜貨行業的現狀就能看出，那裡有如此多的雜貨商，這表明消費者更喜歡選擇，那麼消費者對 DoorDash 產品的興趣就會持續非常濃厚。

And then especially if you add some of the capabilities that we're adding where we're adding fulfillment services with DashMart Fulfillment Services, where we're increasing the quality and doing that for every single grocer. So that they have the capability to compete against companies like Amazon. I think that just bolsters the product offering.

尤其是當我們添加一些我們正在增加的功能，例如透過 DashMart 物流服務增加物流服務時，我們會提高質量，並為每家雜貨店提供服務。這樣他們才有能力與亞馬遜這樣的公司競爭。我認為這只會增強產品競爭力。

Youssef, on your second point, I think, it was on Q1 EBITDA as well as Dashers. Look, I mean, I think there's a couple of factors in Q1. One, it's just phasing of investment, which is somewhat unique this year in the sense that we are investing in ROO. Some of that investment is front loaded. I expect Q1 EBITDA from ROO to be about $25 million lower than Q4, but the full year number is still $200 million, which stays very consistent.

Youssef，關於你的第二點，我認為，這與第一季的 EBITDA 以及 Dashers 有關。你看，我的意思是，我認為第一季有幾個因素。第一，這只是投資的分階段進行，今年的特別之處在於，我們正在投資 ROO。部分投資是前期投入的。我預計 ROO 第一季 EBITDA 將比第四季減少約 2,500 萬美元，但全年數字仍為 2 億美元，保持非常穩定。

The second part is there was an impact because of the winter storms in January, that's roughly about $20 million. And finally, to your point, Dashers has been very seasonal, right? Like we saw seasonal impact of Dashers in Q1, every single year over the last several years, 2023, '24, '25. '26, I would expect it to be very similar. But your second point around your broader Dasher.

第二部分是由於一月份的冬季風暴造成的影響，損失約為 2000 萬美元。最後，正如你所說，Dashers 的季節性很強，對吧？就像我們在過去幾年（2023 年、2024 年、2025 年）的每一年第一季都看到了 Dashers 的季節性影響。26年，我預計情況會非常相似。但你的第二點是關於你更廣泛的 Dasher。

I mean, Dasher trends are fairly consistent with what we've seen in years prior, when I look at Dashers cost as a percentage of GOV, we generated leverage in Q4 on a year-over-year basis and I expect us to continue to generate some leverage in Q1 as well when I look at it as a percentage of GOV on a year-over-year basis.

我的意思是，送餐員的趨勢與我們往年看到的趨勢相當一致。當我查看送餐員成本占公司總營運成本的百分比時，我們在第四季度實現了同比增長，我預計我們在第一季也能繼續實現同比增長。

Great. Thank you, both.

偉大的。謝謝你們兩位。

No problem.

沒問題。

Josh Beck, Raymond James.

喬許貝克，雷蒙德詹姆斯。

Yeah, thank you for taking the question. Yes. I wanted to ask, I know it's only been maybe about five months since you've closed Deliveroo. Have there been any standout learnings when you think about the loyalty program or the fulfillment network or merchant terms. Just anything that has really jumped out to you as an opportunity area to lean in towards.

是的，謝謝你回答這個問題。是的。我想問一下，我知道你們關閉 Deliveroo 大概才五個月左右。在思考會員計畫、物流網絡或商家條款時，您有沒有獲得什麼特別突出的經驗教訓？任何你覺得特別有發展潛力的領域，都可以寫下來。

And then on the platform modernization, I know, it's arguably very early there as well. But anything you can update us on in terms of maybe some of the efficiencies that you're expecting to gain and maybe how that could either accelerate velocity or maybe free up investments elsewhere? Just would really like updates on those topics.

至於平台現代化，我知道，目前也還處於非常早期的階段。但您能否向我們介紹您預期能獲得的效率提升，以及這些提升如何能加快速度或釋放其他方面的投資？非常希望了解這些主題的最新進展。

Thank you.

謝謝。

Sure. Hey, Josh, it's Tony. I'll maybe take a crack at both questions, and feel free to jump in, Ravi. On Deliveroo, I think to kind of echo what we said on the call so far, it's just been a great start. I mean, like I think the numbers and the performance speaks for itself, whenever you're growing faster at the same budget, and you see more room for upside.

當然。嘿，喬什，我是東尼。我或許可以試著回答這兩個問題，拉維，你也隨時可以加入討論。關於 Deliveroo，我想正如我們之前在電話會議上所說的那樣，目前為止一切都進展順利。我的意思是，我認為數據和績效本身就能說明問題，在預算相同的情況下成長速度更快，而且發展空間更大。

I think that's always a great place to start. With respect to opportunities to improve, yes, I agree. There's a ton of things we kind of have identified. I think we'll be able to ship many things, this -- I mean, frankly, this quarter to be able to improve. And it's really just across the board.

我認為這始終是一個很好的起點。關於改進的機會，是的，我同意。我們已經發現了很多問題。我認為我們能夠推出很多產品，坦白說，本季我們將能夠有所改進。而且這種情況普遍存在。

I mean, when I look at these kinds of businesses, it's really a bunch of small things that add up to make the difference. There's never like one glaring huge thing because usually when there is that's a very -- actually quite simple fix. It's usually the combined sum of lots of small things that compound that ultimately is what allows us to offer surplus to customers or a deficit and fail at delivering what the customers want.

我的意思是，當我觀察這類企業時，我發現真正起作用的是許多小事，而這些小事加起來就足以產生影響。從來沒有那種特別顯眼的大問題，因為通常情況下，即使有問題，也很容易解決。通常情況下，正是許多小事的累積最終導致我們無法為客戶提供過剩或短缺的產品，也無法滿足客戶的需求。

And so we see opportunity pretty much across the board and we're shipping things literally every single day to improve them. So we've already seen benefits from things that we've taken from our lessons learned at building DoorDash, at acquiring Wolt and have shipped those to our audiences over Deliveroo to see improvements in all of the audiences.

因此，我們看到了各個領域的改進機會，我們幾乎每天都在推出新產品來提升它們。因此，我們已經從創建 DoorDash 和收購 Wolt 的過程中吸取的經驗教訓中獲益，並將這些經驗教訓推廣到 Deliveroo 的用戶群體中，從而在所有用戶群體中看到了改進。

On the second question with respect to the tech platform, you're completely right in the high -- in the kind of premise of the question where I believe right now, if you think about our setup, it's really not ideal. We have -- we operate on three tech platforms, pretty much a very similar business. And that -- what does that mean? That means that you're going to be slowed down because in order to ship one feature, you have to ship that 3 times in slightly different tool calls and kind of processes that make no sense.

關於技術平台的第二個問題，你的觀點完全正確——就這個問題的前提而言，我認為現在，如果你想想我們的設置，它真的不太理想。我們經營三個技術平台，業務模式非常相似。那——那是什麼意思？這意味著你的速度將會減慢，因為為了發布一個功能，你必須使用略有不同的工具呼叫和毫無意義的流程來發布 3 次。

And so what we're doing is we're making this pretty big investment in order to both improve the velocity in which we ship as kind of -- to clear out some of the inefficiencies that I described and also just be more efficient with our global footprint, right?

因此，我們正在進行一項相當大的投資，目的是為了提高我們的發貨速度，消除我所描述的一些低效之處，同時也為了更有效地利用我們的全球業務，對吧？

And I think we'll be able to do both of those things as a result of building this tech platform. But I think above all else, actually, we've already seen this kind of play out nicely in the four months that we've worked together with Deliveroo where -- when we do ship something that has worked in the US for us, or in another part of Europe with Wolt to Deliveroo, it has added immediate impact to the customer audience. And we see hundreds of those opportunities in the platform work that we're doing, moving everyone into the single tech stack that I think customers across all audiences will benefit.

我認為，透過建構這個技術平台，我們將能夠實現這兩件事。但我認為最重要的是，實際上，在我們與 Deliveroo 合作的這四個月裡，我們已經看到了這種模式的良好效果——當我們把在美國或歐洲其他地區透過 Wolt 取得成功的產品推廣到 Deliveroo 時，它立即對客戶群體產生了影響。我們在平台建立工作中看到了數百個這樣的機會，我們將所有人遷移到單一的技術堆疊中，我認為這將使所有受眾的客戶受益。

Super helpful. Thank you, Tony.

非常有用。謝謝你，托尼。

Eric Sheridan, Goldman Sachs.

艾瑞克‧謝裡丹，高盛集團。

Thanks so much for taking the questions. With respect to DashPass, how does that fit within your broader strategic priorities? And when you think about growth investments in the business in terms of incenting DashPass adoption more broadly across your markets? And then in terms of DashPass adoption, how do you think about that as a potential stimulant for order frequency when you think about cohort evolution looking out over the next 12 to 18 months? Thanks so much.

非常感謝您回答這些問題。關於 DashPass，它如何融入你們更廣泛的策略重點？那麼，當您考慮透過投資來激勵 DashPass 在您的市場中更廣泛地普及時，您會如何看待業務成長？那麼就 DashPass 的採用而言，考慮到未來 12 到 18 個月的用戶群演變，您如何看待它作為提高訂單頻率的潛在刺激因素？非常感謝。

Hey, Eric, it's Tony -- maybe start on DashPass, and feel free to add in, Ravi. I mean DashPass is critical to our business. If you look at this program, it's a program that we started at the end of 2017, really shifted in 2018, and it's continued to be the core driver of our relationship with consumers. And I think it's grown in leaps and bounds, not just in some of the numbers that we've reported, but also in terms of the benefits that we're starting to ship to customers.

嘿，埃里克，我是托尼——也許可以先從 DashPass 開始，拉維，也歡迎你加入。我的意思是，DashPass 對我們的業務至關重要。如果你看一下這個項目，它始於 2017 年底，在 2018 年進行了重大調整，並且一直是我們與消費者關係的核心驅動力。我認為它取得了突飛猛進的發展，不僅體現在我們公佈的一些數字上，也體現在我們開始為客戶帶來的好處上。

And our goal right -- with DashPass is to continue to increase the number of benefits in which we can offer. And we think there's a lot to do. For example, a lot of the benefits that you've seen recently has occurred as we've launched all of our non-restaurant categories, right? We are effectively charging the same fee for DashPass. We're adding even more value to get discounted or preferential pricing delivery for more complicated deliveries in retail and grocery and more valued discounts on key value items, things like this. Another potential area that we see a ton of opportunity is all of the work that we're doing with our in-store business.

而我們的目標是——透過 DashPass——繼續增加我們能夠提供的福利數量。我們認為還有很多事情要做。例如，您最近看到的許多好處都是在我們推出所有非餐飲類別之後出現的，對嗎？我們對 DashPass 實際上收取的是相同的費用。我們正在增加更多價值，為零售和雜貨領域更複雜的配送提供折扣或優惠價格，並為關鍵價值商品提供更有價值的折扣，等等。我們看到的另一個充滿機會的潛在領域是我們在店內業務方面所做的所有工作。

We announced our in-store business in '25 during the September dashboard product event, where we're starting to drive traffic inside the restaurants now with our 56 million MAUs or over 100 million annual customers and offering them either access or value to restaurants that they couldn't get otherwise if they were not members. And so I think the DashPass ecosystem has a long runway ahead of it. And if I kind of take a step back and kind of just think about that in quantitative terms, right now, the average DashPass customer might be interacting with us a couple of times a week or something like that. But when I think about the number of eating occasions, there are 20 to 25.

我們在 2025 年 9 月的儀錶板產品活動中宣布了我們的店內業務，現在我們開始利用 5,600 萬月活躍用戶或超過 1 億年活躍用戶將客流量引至餐廳，並為他們提供餐廳的訪問權限或價值，如果他們不是會員，就無法獲得這些服務或價值。所以我認為 DashPass 生態系統還有很長的發展道路。如果我退後一步，從量化的角度來思考這個問題，那麼目前，DashPass 的平均客戶可能每週會與我們互動幾次左右。但當我想到吃飯的次數時，大約有 20 到 25 次。

And then when I add in the shopping occasions on top of food consumption, it exceeds 20 to 25. So I think we're a fraction or a single-digit percentage of what DashPass could actually achieve. So there's a lot of work to drive frequency, but that starts with adding more use cases to add more value.

再加上購物次數和食品消費，總數就超過 20 到 25 次了。所以我認為我們只達到了 DashPass 實際潛力的一小部分，甚至不到個位數百分比。因此，要提高使用頻率還有很多工作要做，但這首先要從增加更多用例來增加價值開始。

Ross Sandler, Barclays.

羅斯·桑德勒，巴克萊銀行。

Yes. I just wanted to have two. Can we get an update on the Storefront software business and how SevenRooms and kind of the CRM part of it might be kind of accelerating the efforts on that side. And then back to the replatforming, you guys have done a nice job of laying out how all these components are coming together. Just a question on timing.

是的。我本來只想買兩個。能否介紹一下 Storefront 軟體業務的最新進展，以及 SevenRooms 及其 CRM 部分如何加速這方面的努力？回到平台重構的話題，你們在闡述所有這些元件如何組合在一起方面做得很好。關於時間安排的問題。

So clearly, this will benefit both like speed of new features being rolled out and potentially expanding products or geos. When we kind of see the benefits starting to show up? Is this going to be like a continuous thing in '26 onwards? Or is it more kind of like after this year? Just any thoughts on the timing? Thanks a lot.

顯然，這將有利於加快新功能的推出速度，並有可能擴大產品或地理範圍。我們什麼時候才能開始看到好處顯現出來？從2026年開始，這種情況會持續下去嗎？或者更像是今年之後？關於時間安排，大家有什麼想法嗎？多謝。

Yes. I can start. So on first question with respect to Storefront and software in general. It's going really well. I mean, like if you look at the integration of SevenRooms, that also has been a relatively newer project for us, about six months in partnership with the SevenRooms team.

是的。我可以開始了。那麼，關於 Storefront 和軟體方面的第一個問題。一切進展順利。我的意思是，例如你看看 SevenRooms 的整合，那對我們來說也是一個相對較新的項目，我們與 SevenRooms 團隊合作了大約六個月。

And already, we've been able to tremendously speed up their work. I mean, they're now adding venues 50% faster post-acquisition than before we partnered with them. And if you think about like if we step back and think about the thesis, it's kind of proving the thesis that if you can add the best-in-class CRM software with the largest demand generator platform, which is DoorDash, it is a really valuable asset for these restauranteurs who really want to build more regular customers and have that direct relationship.

而且，我們已經大大加快了他們的工作速度。我的意思是，收購之後，他們新增場館的速度比我們與他們合作之前快了 50%。如果我們退後一步，重新審視這個論點，就會發現它實際上是在證明這樣一個論點：如果能將一流的 CRM 軟體與最大的需求生成平台 DoorDash 結合起來，對於那些真正想要建立更多回頭客並與之建立直接關係的餐館老闆來說，這將是一筆非常有價值的資產。

That is also true for the Storefront products, which is less about in-room dining or in-store dining and much more about the takeaway product. But especially as we expand kind of the merchant cohort in which SevenRooms currently addresses, SevenRooms is kind of serving the higher end kind of cohort of restaurants.

對於 Storefront 產品也是如此，它與其說是客房送餐或店內用餐，不如說是外帶產品。但尤其隨著 SevenRooms 目前所服務的商家群體不斷擴大，SevenRooms 正在服務更高端的餐廳群體。

But as we simplify the features to be able to address a larger segment, that's where Storefront and SevenRooms really can team up to both serve a restaurant's four-wall business as well as their business outside their premise.

但隨著我們簡化功能以服務更廣泛的客戶群，Storefront 和 SevenRooms 真正能夠攜手合作，既服務於餐廳的店內業務，也服務於其店外業務。

Your second question on the timing of the tech stack. Look, I wish that the tech stack were already here. But the truth is you'll see a majority of the tech stack work completed this year. That's at least my expectation. And I think with respect to the benefits, you don't have to wait all at once.

關於技術棧部署時間安排的第二個問題。我希望技術棧已經到位了。但事實上，今年大部分技術棧的工作都將完成。至少我是這麼認為的。我認為就好處而言，你不必一次全部享受到。

In fact, we're already seeing benefits from the work we're doing with the global tech stack in all of our different geographies in which we're shipping features from one market that are working into a different market. So that's already happening. That's in fact, one way in which you can test whether or not the value of the tech stack actually has any positive value to customers. And so we're seeing that. But yes, we expect the majority of the work to be done this year.

事實上，我們已經從我們在全球不同地區使用全球技術堆疊的工作中看到了好處，我們正在將一個市場的功能發佈到另一個市場。這種情況已經發生了。事實上，這是檢驗技術棧的價值是否真正對客戶有任何正面價值的方法。所以我們看到了這一點。是的，我們預計大部分工作將在今年完成。

Ken Gawrelski, Wells Fargo.

Ken Gawrelski，富國銀行。

Thank you so much. Two, if I may, please. First, could you provide some color on what you see in the cohort data that gives you the confidence in continued robust core US restaurant growth. Anything you might share on how later adopters behave differently than the base?

太感謝了。如果可以的話，請給我兩個。首先，您能否詳細說明一下，您從隊列數據中看到了哪些內容，讓您對美國核心餐飲業持續強勁成長充滿信心？關於後期採用者和基礎使用者在行為上的差異，您有什麼可以分享的嗎？

And the second question, I just want to touch one more maybe on where Ross was going. As we look to '27, I know it's really early to speak about '27, and we appreciate the color on the '26 quarters. But any way -- any early take you could give us on the balance between investments and margins beyond '26. Any early thoughts on how you view incremental margins beyond '26 versus maybe historical? Thank you.

第二個問題，我只想再稍微談談羅斯的去向。展望 2027 年，我知道現在談論 2027 年還為時過早，我們很欣賞 2026 年 25 美分硬幣的顏色。但無論如何——您能否就 2026 年以後的投資和利潤之間的平衡給出一些初步看法？對於2026年以後的增量利潤率與歷史水準相比，您有什麼初步看法？謝謝。

Sure. Again, I'll take the first one too. Look, I mean, when I look at the performance of the US restaurant digital, like I said in the earlier question, I mean, '25 grew faster than '24 at a larger scale. I should tell you the health of the underlying cohorts. MAUs continue to be quite strong. In fact, we hit an all-time high in terms of MAUs, and order frequency continues to be quite strong.

當然。我還是選第一個吧。你看，我的意思是，當我觀察美國餐飲業數位化業務的表現時，就像我在前面的問題中提到的那樣，2025 年的成長速度比 2024 年更快，而且規模更大。我應該告訴你這些人群的健康狀況。月活躍用戶數依然維持強勁勢頭。事實上，我們的月活躍用戶數達到了歷史新高，訂單頻率也持續保持強勁勢頭。

When I look at the engagement of new consumers that still join, I mean, that continues to be quite strong. And the other thing that you're seeing is, I mean, subscription continues to be a big driver of growth for us, both Q4 as well as '25, we added a record number of subscribers. What you're seeing in the business is as the product continues to get better, there's more people that habituate, they graduate towards subscription. Subscription, they retain more, they order more as well as they try new categories. So overall, when I look at the mature cohorts as well as the newer cohorts, the engagement level in the US business continues to be strong, not just across restaurants, but even our new verticals business.

當我觀察新用戶的參與度時，我發現他們的參與度依然非常高。此外，訂閱仍然是我們成長的主要驅動力，無論是第四季還是 2025 年，我們都新增了創紀錄的訂閱用戶數量。在商業領域，我們看到的是，隨著產品不斷改進，越來越多的人養成習慣，並逐漸過渡到訂閱模式。訂閱用戶留存率更高，訂購量也更大，他們也會嘗試新的產品類別。總的來說，當我觀察成熟的客戶群以及較新的客戶群時，美國業務的參與度依然很高，不僅在餐飲領域，甚至在我們新的垂直業務領域也是如此。

And I think your second question, Ken was just sort of incremental margins. Look, I mean we're not guiding the business towards incremental margins. We're not operating the business towards incremental margins. Our focus has always been on overall profit dollar production. But just to frame your thinking in terms of overall tech stack and what the impact there is.

我覺得你的第二個問題，肯，只是關於增量利潤的問題。你看，我的意思是，我們並沒有引導業務朝著提高利潤率的方向發展。我們經營業務的目的並非追求利潤成長。我們始終關注的是整體利潤額。但為了幫助你從整體技術堆疊的角度思考問題，以及它會產生什麼影響。

Look, there's a couple of costs in there. One is some redundancy in cost as we try to run both tech stacks in parallel. Majority of that spend will be in '26, some will be in '27. But that will come off. And you should expect that to be a smaller component. But biggest value driver for us like we touched on earlier, is going to be velocity of feature development, you're going to become more efficient as developers work on the same tech stack across all three platforms. And you should see the impact of that from an underlying cohort perspective as well as overall growth in the business.

你看，這裡面有一些費用。其一是成本上存在一些冗餘，因為我們試圖並行運行兩種技術棧。大部分支出將在 2026 年進行，一部分將在 2027 年進行。但那會消失的。而且你應該預料到它會是一個較小的組成部分。但正如我們之前提到的，對我們來說最大的價值驅動因素將是功能開發的速度，隨著開發人員在所有三個平台上使用相同的技術棧，效率將會更高。你應該從潛在用戶群以及業務整體成長的角度來看這個問題的影響。

Thank you.

謝謝。

Bernie McTernan, Needham.

伯尼·麥克特南，尼德姆。

Great, thanks for the question. Just wanted to follow up on the discussion on AVs. Do you think the use case for delivery AVs will be broader than robotaxis and mobility, meaning that at least for the foreseeable future, robotaxis are expected only to be in dense cities. Is there a use case for delivery AVs in the suburbs.

太好了，謝謝你的提問。想就自動駕駛汽車的討論做個後續跟進。你認為送貨自動駕駛汽車的應用場景會比無人駕駛計程車和出行方式更廣泛嗎？也就是說，至少在可預見的未來，無人駕駛計程車預計只會出現在人口密集的城市。郊區是否有自動駕駛汽車送貨的應用場景？

And then as we're asking follow-up questions on the financial guidance, in the press release talked about EBITDA a lot higher in the second half of the year. I would say that's probably typical to normal seasonal trends that we see within the business. Just any additional color you could provide would be helpful. Thank you.

然後，當我們就財務指引提出後續問題時，新聞稿中提到下半年的 EBITDA 將大幅成長。我認為這可能是我們行業內常見的正常季節性趨勢。如果您能提供其他顏色，那就太好了。謝謝。

Yes. Hey, Bernie, it's Tony. I'll take the first question on AVs. I mean the short answer is absolutely. Yes, of course, we think that the delivery vehicles will be able to address both suburbs and city centers.

是的。嘿，伯尼，我是托尼。我先回答關於AV的第一個問題。我的意思是，簡而言之，答案是肯定的。當然，我們認為送貨車輛能夠覆蓋郊區和市中心。

In fact, if you look at actually Dot, it was constructed in a purposeful way to actually serve many of the suburbs. And that's true in its form factor, that's true in how we think about integrating it into the autonomous delivery platform in terms of which assignments it ought to receive versus which assignments it should not receive. That's also true for all of our projects for our drone projects, too, which cover, I would say, even beyond suburbs, but even more rural regions where the distances traveled are much farther and you can go a lot faster in the air sometimes than you can go on land.

事實上，如果你仔細觀察 Dot，你會發現它的建造方式經過精心設計，旨在服務許多郊區。從外形上看確實如此，從我們考慮如何將其整合到自動配送平台中，即它應該接收哪些任務而不是不應該接收哪些任務來看，也確實如此。對於我們所有的無人機項目來說，情況也是如此。我認為，這些項目甚至涵蓋了郊區以外的地區，甚至是更偏遠的農村地區，在這些地區，飛行距離更遠，而且在空中飛行的速度有時比在陸地上快得多。

Hey, Bernie, on your second question, I mean, look, what I'll say is my expectation, like I said, for the full year has not changed compared to the last call. I expect the full year margin to be up slightly compared to '25, excluding ROO and ROO to be about $200 million. You're also right. Look, the shape of the curve for us for EBITDA has always been -- second half is higher than the first half. That's naturally how our business works.

嘿，伯尼，關於你的第二個問題，我的意思是，你看，我要說的是，就像我之前說的，我對全年的預期與上次預測相比沒有改變。我預計全年利潤率將比 2025 年略有上升，不包括 ROO，ROO 約為 2 億美元。你說得也對。你看，我們這裡的 EBITDA 曲線形狀一直都是——下半年高於上半年。我們的業務運作方式自然就是這樣。

It's purely math, right? The volume grows through the year, unit economics grow. You put those two together, you'll have more gross profit dollars as you go through the year, which means second half will be higher than first half. We have delivered on that in '23, '24, '25, and '26 is going to be no different. There are a couple of things which are different. One is investment in ROO. Some of it is front loaded. Like I said, it impacts Q1. I would expect the ROO EBITDA to increase as you go through the rest of the year.

這純粹是數學問題，對吧？銷量逐年成長，單位經濟效益也隨之提高。把這兩點結合起來，隨著​​時間的推移，你的毛利就會增加，這意味著下半年的毛利會高於上半年。我們在 2023 年、2024 年、2025 年都達成了這個目標，2026 年也不例外。有幾點不同。一是投資ROO。部分部件是前置式的。正如我所說，這會影響第一季。我預計隨著今年剩餘時間的推移，ROO EBITDA 將會成長。

The second one that we are seeing in the business is when I look at the pace of expansion of profitability for both our new verticals and international business ex ROO that will increase or the pace of expansion is higher than in previous years. Especially as I expect new verticals to be gross profit positive in the second half and international ex ROO to be contribution profit positive in the second half. So you put all this together, that's what's giving rise to the shape of EBITDA curve where second half is going to be higher than the first half EBITDA.

我們在業務中看到的第二個變化是，當我觀察我們新業務領域和國際業務（不包括 ROO）的獲利能力擴張速度時，會發現其擴張速度比往年更快。尤其是我預計下半年新業務板塊將實現毛利為正，而國際業務（不包括海外收入）下半年也將實現利潤貢獻為正。所以把所有這些因素綜合起來，就形成了 EBITDA 曲線的形狀，即下半年的 EBITDA 將高於上半年的 EBITDA。

Very helpful. Thank you both.

很有幫助。謝謝你們兩位。

Andrew Boone, Citizens Bank.

安德魯·布恩，公民銀行。

Thanks so much for taking the questions. Ravi, I wanted to stay on costs and the intensity of investments. If I look at R&D and G&A, it's 211 basis points of GOV. We've talked about kind of that 2% target historically. Should we expect GOV to kind of grow when we just grow into this higher fixed cost? Or how do we think about the fixed cost component of the business on a go-forward basis?

非常感謝您回答這些問題。拉維，我想繼續關注成本和投資強度。如果我看一下研發和一般管理費用，政府支出為 211 個基點。我們之前也討論過2%左右的目標。隨著固定成本的增加，我們是否應該預期政府支出也會隨之成長？或者，我們該如何看待企業未來發展的固定成本部分？

And then going to grocery, is there any detail that you can provide us in terms of the graduation of customers into the Sunday Shop. Can you talk about just the evolution of customers and whether you're starting to capture larger baskets?

然後是雜貨店，您能否提供一些關於顧客如何逐步過渡到週日商店的細節？您能否談談客戶群的演變，以及您是否開始獲得更大的客單價？

Thanks so much.

非常感謝。

Yes, Andrew, I'll take the first one, right. Look, when I look at OpEx in Q4, I think that's probably what you're asking about as well. There's inclusion of ROO. So you should take that into consideration. When I look at 2026, I would expect OpEx to be roughly about 2% of GOV that we talked about.

是的，安德魯，我想要第一個，好嗎？你看，當我查看第四季的營運支出時，我想這可能也是你所問的。包含 ROO。所以你應該把這一點考慮進去。展望 2026 年，我預計營運支出將大約占我們之前討論過的政府支出的 2%。

Look, we're being very disciplined. We're investing in areas where we're improving the product to ultimately drive both scale as well as profitability. You're seeing that in terms of the overall growth as well as the profit dollar production. Look, I mean, our goal is to continue to generate leverage, right. Like OpEx, I think of it as cost of doing business, and our goal is to continue to generate leverage on it just like any other part of the P&L.

你看，我們紀律嚴明。我們正在投資於能夠改進產品的領域，以最終推動規模化和獲利能力的提升。從整體成長和利潤額來看，這一點都顯而易見。我的意思是，我們的目標是繼續創造槓桿效應，對吧。就像營運支出一樣，我認為它是開展業務的成本，我們的目標是像損益表中的其他任何部分一樣，繼續利用它來產生槓桿效應。

Andrew, on the second question with respect to grocery, yes, is the short answer. We are seeing the evolution of customer behavior both now incorporating kind of the middle of the week run, which is kind of how DoorDash started five years ago in the grocery category to now the larger baskets that happen on the weekends here in the US and in other parts of the world, it has a slightly different behavior, but we do see both behaviors now where people use us for both kind of the quick runs as well as the stock-up use cases.

安德魯，關於第二個關於食品雜貨的問題，簡短的回答是：是的。我們看到客戶行為正在演變，既包括現在常見的周中購物（這與 DoorDash 五年前在雜貨類別中的起步方式類似），也包括現在美國和世界其他地區週末的大批量購物。雖然這種行為略有不同，但我們現在確實看到這兩種行為，人們既使用我們的服務進行快速購物，也使用我們的服務進行囤貨。

You see this, in fact, happening faster and faster with each successive cohort and you see each existing cohort actually increasing their spend and their overall share of wallet when it comes to grocery with us.

事實上，你會發現這種情況隨著每一代顧客群體的更迭而愈發迅速，而且你會發現，每一代現有顧客群體在我們這裡購買食品雜貨的支出和總支出份額都在不斷增加。

Thank you.

謝謝。

Lloyd Walmsley, Mizuho.

勞埃德‧沃姆斯利，瑞穗銀行。

Great, I was just wondering if beyond the first quarter guidance you can help us with just how to think about a framework for GOV growth for the balance of the year. I appreciate the comments on the EBITDA cadence, but anything you can help us out with on GOV growth?

太好了，我只是想問一下，除了第一季的業績指引之外，您能否幫助我們思考今年剩餘時間政府部門成長的框架？感謝您對 EBITDA 成長節奏的評論，但您能否就政府支出成長方面提供一些幫助？

Yes. Look, I mean, I think a couple of things, right? Like one, growth in the business continues to be quite strong. And we are seeing that from both existing consumers as well as new consumers where MAU growth continues to be strong, order frequency continues to be strong. Like I said in one of the earlier questions, I mean, DashPass had a record year as well as continues to drive overall growth.

是的。你看，我的意思是，我覺得有幾件事，對吧？與第一點類似，該業務的成長動能依然強勁。我們看到，無論是現有消費者還是新消費者，每月活躍用戶數 (MAU) 成長依然強勁，訂單頻率也依然強勁。就像我在之前的問題中提到的那樣，DashPass 不僅創下了歷史新高，而且還在繼續推動整體成長。

For us, the way I think about it is as long as we are continuing to improve the product, the underlying cohorts are responding, as you can see from engagement as well as sort of retention and for us, I feel pretty good about the overall growth, not just in Q1 but for the rest of the year as well.

我認為，只要我們不斷改進產品，用戶群就會做出積極回應，正如你從用戶參與度和留存率中看到的那樣。因此，我對整體成長感到非常滿意，不僅是第一季度，而且對今年剩餘時間也是如此。

All right. Thank you.

好的。謝謝。

Lee Horowitz, Deutsche Bank.

李‧霍洛維茨，德意志銀行。

Great, thanks for taking the question. Maybe building on an earlier one. 2026 is obviously a big year for investments in sort of the software and services stack for your merchants. I guess looking beyond this year, where do you see sort of the natural adjacencies that you can build on top of once you've sort of rolled out this new software service stack for your merchants that will drive more value to both your merchants and the consumers in the coming years.

太好了，謝謝你回答這個問題。或許是在前期工作的基礎上進行。顯然，2026年對商家而言，是投資軟體和服務體系的重要一年。我想展望今年之後，您認為在為商家推出這套新的軟體服務體系之後，有哪些自然而然的鄰近領域可以作為基礎，從而在未來幾年為商家和消費者創造更多價值。

Yeah, hey Lee, I can start. I mean, the short version of this is -- I mean '26 in many ways, is like a setup year of building like a new company that is now a global company operating in 40-plus geographies around the world. But doing the same thing. We just have like different countries and different markets to consider and there are also at different maturities that when we've ship certain products, right? So there's a lot of different things.

好的，李，我可以開始了。我的意思是，簡而言之——我的意思是，2026 年在很多方面就像是一家新公司的籌備年，而現在這家新公司已經是一家在全球 40 多個地區開展業務的全球性公司。但做的是同樣的事情。我們需要考慮不同的國家和不同的市場，而且當我們出貨某些產品時，它們的成熟度也各不相同，對吧？所以有很多不同的東西。

And I think a lot of the adjacencies to your question can be derived from what we're shipping at DoorDash, right? I mean -- so if you think about the different missions we have. One is we want to bring you everything inside the city. So a lot of the work is going beyond restaurants. I think we've certainly proven ourselves capable, at least in the categories of grocery, convenience and some of the early innings of other retail categories. So we've got a lot of work there to do.

我認為，你提出的許多問題都可以從我們在 DoorDash 推出的產品中找到答案，對吧？我的意思是──想想我們肩負的不同任務。第一，我們想把城裡的一切帶給你。所以很多工作都超出了餐廳的範疇。我認為我們已經證明了自己的能力，至少在食品雜貨、便利商店以及其他一些零售類別的早期階段是如此。所以，我們還有很多工作要做。

A big part of that has to do with also building fulfillment services in which we can forward deploy inventory on behalf of retailers. So that they can offer same-hour, same-day delivery and be competitive with other big companies. We also have to invest in autonomous technologies so that these companies can do it at the highest quality and the lowest cost.

這其中很大一部分原因在於，我們還需要建立物流服務，以便代表零售商提前部署庫存。這樣他們就可以提供當小時、當日送達服務，並與其他大公司競爭。我們還必須投資於自主技術，以便這些公司能夠以最高的品質和最低的成本完成這項工作。

So that's one big mission in terms of how those adjacencies work. The second one is how do we actually build software such that these companies can be omnichannel businesses? We talked a little bit about this on this call where we talked about Storefront, SevenRooms. But there's also a DoorDash Drive and offering delivery as a service. There's other services that we ought to be building as you think about how a physical business must effectively adopt in order to become an end-to-end digital business.

所以，從這些鄰近關係如何運作的角度來看，這是一項重大任務。第二個問題是，我們如何真正建立出能讓這些公司成為全通路企業的軟體？我們在這通電話中稍微談到了這一點，當時我們談到了 Storefront 和 SevenRooms。但也有 DoorDash Drive 提供外送服務。還有其他一些服務我們也應該著手構建，因為在思考實體企業如何有效地轉型為端到端的數位化企業時，我們需要思考這一點。

So we have a lot of merchant services work that we have to do. That's also connected to the third mission of actually driving in-store traffic to merchants, right? We're starting this with restaurants in the form of two products going out in which we're offering value to customers to discover new restaurants for casual dining. And we're also doing it in the form of access where we're offering reservations to some of the best restaurants that work with SevenRooms. And so I think the -- there's a ton of work there to do as well.

所以我們有很多商家服務的工作要做。這也與第三個任務——實際引導客流量到店消費——有關，對吧？我們先從餐廳著手，推出兩款產品，為顧客提供價值，幫助他們發現新的休閒餐飲餐廳。此外，我們還透過提供預訂服務，讓用戶能夠預訂與 SevenRooms 合作的一些最好的餐廳。所以我覺得──那方面也有很多工作要做。

So I think those are three big missions that will take us quite a long journey in terms of building the operating system for local commerce and connecting consumers and merchants in more ways than we currently do. And if we can do this across every geography in the way that makes the most sense for that geography, I think, it's a very exciting future for DoorDash.

所以我認為這三大任務將帶領我們走上一條漫長的道路，以建立本地商業作業系統，並以比現在更多的方式連接消費者和商家。如果我們能夠以最適合當地情況的方式，在每個地區都做到這一點，我認為，DoorDash 的未來將非常令人興奮。

Justin Patterson, KeyBanc.

Justin Patterson，KeyBanc。

Great, thank you. Good afternoon. On the ad side, it looks like you made some nice progress with Symbiosis. I'd love to hear more about how you're evolving the ad product this year and some of the key stats you're taking to capture more grocery and retail advertising dollars.

太好了，謝謝。午安.在廣告方面，你們似乎在Symbiosis方面取得了一些不錯的進展。我很想了解你們今年是如何改進廣告產品的，以及你們正在採取哪些關鍵措施來吸引更多食品雜貨和零售廣告預算。

And then separately, I know this is not related to replatforming, but we have seen a lot of companies see benefits from agentic coding. Curious how -- what type of efficiencies, if any, you're seeing from that and how that might fit with the broader replatforming initiatives? Thank you.

另外，我知道這與平台重構無關，但我們已經看到很多公司從智慧編碼中受益。很好奇具體是如何實現的——你們從中看到了哪些效率提升（如果有的話），以及這與更廣泛的平台重構計劃有何關聯？謝謝。

Yes. Hey, Justin, this is Tony. Yes, maybe I'll take both. On the question around Ads, our Ads business is growing really, really fast, and it's probably something I should have added in terms of just when I think about the record year that DoorDash had in 2025, Ads was a big part of it in terms of -- it's more of a derivative or an output in terms of the growth from our marketplace. But it really had a strong year.

是的。嘿，賈斯汀，我是托尼。是的，也許我會兩個都選。關於廣告方面的問題，我們的廣告業務成長非常非常快，我可能應該補充一點，當我想到 DoorDash 在 2025 年創紀錄的一年時，廣告在其中扮演了重要角色——就我們市場的成長而言，它更像是一種衍生性商品或產出。但它今年的表現確實非常出色。

With respect to Symbiosis, I mean, it's, again, kind of similar to some of the other acquisitions that we've made. It's off to a great start. We see that we've doubled the number of advertisers for Symbiosis as well as tripled the spend from those advertisers.

至於 Symbiosis，我的意思是，它和我們之前收購的其他一些公司有點類似。開局非常順利。我們看到，Symbiosis 的廣告客戶數量翻了一番，這些廣告客戶的支出也增加了兩倍。

And so I think that's been kind of the performance kind of speaks for itself. And in terms of kind of the road map this year, I think there's a couple of things. One is just -- I think one of the things that there's a lot of obviously talk in the ecosystem about just agents and agentic commerce and things like this. But one of the biggest agents that actually we've shipped last year was really our Smart Campaigns product in which we are helping restaurants buy on their behalf, always ROI positive ad campaigns effectively.

所以我覺得，這已經說明了一切，表演本身就說明了一切。至於今年的發展路線圖，我認為有兩件事。其一是——我認為生態系統中有很多關於代理商、代理商業等等的討論，這顯然是其中之一。但去年我們推出的最大代理商之一是我們的智慧廣告活動產品，我們透過該產品幫助餐廳購買廣告，始終有效地進行投資回報率為正的廣告活動。

And I think that has been one of our fastest-growing products on the restaurant front. With respect to grocery and retail, Yes, we're definitely earlier to -- on that maturation. But I think that just means that there's a lot more opportunity and runway for us in that space because most of the focus on the ads business thus far has been on the restaurant side.

我認為這是我們在餐飲領域成長最快的產品之一。就食品雜貨和零售而言，是的，我們肯定更早達到——成熟階段。但我認為這只意味著我們在這個領域還有更多的機會和發展空間，因為到目前為止，廣告業務的大部分焦點都集中在餐飲方面。

With respect to kind of coding agents, I think, which is kind of the premise of your second question, this is a topic that is changing literally by the day, maybe by the week. It's been kind of astonishing and almost breathtaking how fast the coding is changing or has changed, is changing and likely will continue to change given the pace and trajectory that we're on.

關於編碼代理的類型，我認為（這算是你第二個問題的前提），這是一個幾乎每天、甚至每週都在變化的話題。程式碼變化之快，或者說已經變化、正在變化、並且鑑於我們目前的發展速度和軌跡，很可能會繼續變化，這著實令人驚訝，幾乎令人嘆為觀止。

I mean we see 90%-plus daily active usage, something like that across all of our engineers when it comes to these coding agents, which certainly has made them productive. The question now is like what is the right new environment for them to kind of keep up that sustained productivity gain. And so that's, I think, the world that we and many others are trying to figure out.

我的意思是，我們看到我們所有工程師的每日活躍使用率都超過 90%，這無疑提高了他們的工作效率。現在的問題是，什麼樣的新環境才能讓他們維持持續的生產力成長？所以，我認為，這就是我們和許多其他人正在努力了解的世界。

Justin Post, BofA.

Justin Post，美國銀行。

Great, thanks. In the release, you talked about unit economics of grocery retail going positive in the second half. What's enabling that? Is that scale? Or are you seeing new efficiencies on that front?

太好了，謝謝。在新聞稿中，您提到食品零售業的單位經濟效益在下半年將轉為正成長。是什麼促成了這種情況？這是比例尺嗎？或者，您是否看到這方面有了新的效率提升？

And then thinking long term, how do you think about grocery retail bottom line profitability relative to US restaurant? Thank you.

從長遠來看，您認為食品零售的利潤率相對於美國餐飲業如何？謝謝。

Justin, I touched on this earlier. Look, I mean, overall, new verticals continues to be doing really well. The profitability side, we've made really good improvements on the unit economics. To me, there's nothing step function or one big thing that we have to go solve, is continued execution across improving logistics efficiency, improving the quality of the product. We talked about the fact that we're seeing basket sizes being bigger, both for existing as well as mature cohorts.

賈斯汀，我之前已經提到過這一點。總的來說，新興垂直領域持續表現良好。在獲利能力方面，我們在單位經濟效益方面取得了非常好的進展。對我來說，沒有什麼需要一步一步解決的大問題，只需要持續改善物流效率，提高產品品質。我們討論了這樣一個事實：無論是現有客戶群還是成熟客戶群，他們的購物籃規模都在擴大。

Just continued execution, trying to find pockets of efficiency up and down the P&L, improve the product, which will ultimately get us to being gross profit positive in the second half of the year. Look, I mean, longer-term focus continues to be -- how do we get 30% of our MAUs that order from categories, set of restaurants to be closer to 100%. And if we're able to work on the product, if we make the quality of the product better, I think, this is going to be a large business, which will produce strong free cash flow for us over the long period of time.

繼續執行，努力在損益表的各個環節尋找提高效率的機會，改進產品，最終使我們在下半年實現毛利為正。我的意思是，長期的重點仍然是——我們如何讓從特定類別或餐廳訂購的每月活躍用戶佔比達到 30% 並接近 100%。如果我們能夠改進產品，提高產品質量，我認為這將是一項大生意，將在很長一段時間內為我們帶來強勁的自由現金流。

Thank you.

謝謝。

Mark Mahaney, Evercore ISI.

Mark Mahaney，Evercore ISI。

Thanks. I was just going to ask one question on Deliveroo. You talked about how you accelerated the year-over-year growth in total orders for Deliveroo in the December quarter. That's faster than I would have expected. Was there -- just explain how you were able to do that? And just are there a lot of other things you see that make you think that you can continue to accelerate those orders? Thank you very much.

謝謝。我本來只想問一個關於Deliveroo的問題。您曾談到，您是如何在 12 月季度加快 Deliveroo 總訂單量同比增長的。比我預想的還要快。當時是怎麼做到的？請解釋一下你是如何做到的？那麼，你還看到哪些其他因素讓你認為可以繼續加快這些訂單的交付速度呢？非常感謝。

Yes. Mark, it's Tony. Look, the short answer is we just shipped improvements to the product. And there was no like one thing. And what I've learned about a lot of these businesses is -- it sounds really easy to bring you a burrito on time every time in the condition that you'd expect.

是的。馬克，我是東尼。簡單來說，我們剛剛發布了產品的改進版本。而且沒有一樣東西是相同的。我從許多這類企業了解到的情況是——聽起來好像每次都能按時為你送上一份符合你期望的墨西哥捲餅很容易。

It's actually another much more difficult thing to do it in practice. And it's just shipping a new thing every time we see a problem. And I think one of the great things was just how easy the partnership has been with the Deliveroo team. I mean I really commend our teams working together really well. And just again, shipping, executing to deliver against increasing customer expectations.

實際上，在實踐中做到這一點要困難得多。每次發現問題，我們就得推出新的產品。我認為最棒的一點就是與 Deliveroo 團隊的合作非常順利。我的意思是，我真心讚賞我們團隊之間出色的合作。再次強調，出貨和執行都是為了滿足不斷成長的客戶期望。

We are not done. We have like a ton of work to do. Yes, we had a great kind of start, and I'm really proud of the team. On the flip side, I also see like a ton of opportunity where we have to ship even more things. And so most of it was like just shipping things that we know are broken to also adding things that we've seen work in other parts of our business. We just see a lot more of those opportunities in front of us.

我們還沒完。我們還有一大堆工作要做。是的，我們開局非常棒，我為團隊感到非常自豪。另一方面，我也看到了很多機會，我們需要推出更多的產品。因此，大部分工作就像是把我們知道已經損壞的東西重新發貨，同時也加入了一些我們在業務其他方面看到行之有效的產品。我們看到，擺在我們面前的這樣的機會更多了。

Okay, thank you, Tony.

好的，謝謝你，托尼。

Thank you, everyone. This concludes today's Q&A due to the time, and this also concludes today's call. Thank you for attending. You may now disconnect. Goodbye.

謝謝大家。由於時間關係，今天的問答環節到此結束，今天的電話會議也到此結束。謝謝各位的出席。您現在可以斷開連線了。再見。