Doordash, Inc. (DASH) 2022 Q2 法說會逐字稿

Good afternoon, and welcome to the DoorDash Quarter 2, 2022 Earnings Call. (Operator Instructions) And finally, I would like to advise all participants that this call is being recorded. Thank you.

午安，歡迎參加 DoorDash 2022 年第二季財報電話會議。（接線員指示）最後，我要提醒所有與會者，本次電話會議將被錄音。謝謝。

I would now like to hand over to Andy Hargreaves, Vice President, Finance and Investor Relations. Andy, please begin your call.

現在我想把時間交給財務與投資人關係副總裁 Andy Hargreaves。Andy，請開始。

Thank you very much. Good afternoon, and thanks, everybody, for joining us on our second quarter 2022 earnings call. I'm pleased to be joined today by Co-Founder, Chair and CEO, Tony Xu; and CFO, Prabir Adarkar. We'll be making forward-looking statements during today's call, including our expectations of our business and the Wolt business following our acquisition the macro academic environment, financial position and operating performance, our market and local commerce opportunities, future financial results and guidance, our strategies and investment approach. Forward-looking statements are subject to risks and uncertainties that could cause actual results to differ materially from those described, and some risks are described in our risk factors including in our SEC filings, including Form 10-K and 10-Qs. You should not rely on our forward-looking statements as predictions of future events. We disclaim any obligation to update any forward-looking statements, except as required by law.

非常感謝。午安，也謝謝各位參加我們 2022 年第二季財報電話會議。今天很高興與我一同出席的有共同創辦人、董事長兼執行長 Tony Xu，以及財務長 Prabir Adarkar。我們在今天的電話會議中將作出前瞻性陳述，包括我們對於本公司業務以及在完成收購後 Wolt 業務的預期、總體經濟環境、財務狀況與營運表現、我們的市場與在地商務機會、未來財務結果與指引、以及我們的策略與投資方法。前瞻性陳述存在風險與不確定性，可能導致實際結果與所述內容有重大差異；部分風險已在我們的風險因素中說明，包括我們向 SEC 提交的文件（包括 Form 10-K 與 10-Q）。您不應將我們的前瞻性陳述視為對未來事件的預測。除法律要求外，我們不承擔更新任何前瞻性陳述之義務。

During this call, we will also discuss certain non-GAAP financial measures. Information regarding our non-GAAP financial results, including a reconciliation of non-GAAP results to the most directly comparable GAAP financial measures may be found in our investor letter, which is available on our IR website. These non-GAAP measures should be considered in addition to our GAAP results and are not intended to be a substitute for our GAAP results.

在本次電話會議中，我們也將討論若干非 GAAP 財務衡量指標。關於我們的非 GAAP 財務結果之資訊（包括將非 GAAP 結果調節至最直接可比之 GAAP 財務衡量指標的對帳），可在我們的投資人信中找到；該信可於我們的 IR 網站取得。這些非 GAAP 指標應與我們的 GAAP 結果一併考量，並非用以取代 GAAP 結果。

Finally, this call in its entirety is being audio webcast on our IR website. An audio replay of the call will be available on our website shortly after the call ends. As in previous quarters, we'll go straight to questions. So with that, operator, please go ahead and take the first question.

最後，本次電話會議全程將在我們的 IR 網站進行音訊網路直播。電話會議結束後不久，網站上將提供音訊重播。如同前幾季，我們將直接進入問答環節。那麼，接線員，請開始進行第一個問題。

(Operator Instructions) And your first question today comes from the line of Deepak Mathivanan from Wolfe Research.

（接線員指示）今天的第一個問題來自 Wolfe Research 的 Deepak Mathivanan。

Maybe one for Tony and one for Prabir. Tony, given the state of the market and kind of competition, it feels like at least some levels of incentives has come out of the system. Maybe perhaps from smaller players in kind of niche categories. Can you talk about whether you're seeing any benefits from favorable operating environment, either on frequency or kind of customer acquisition right now?

我可能各問 Tony 一題、Prabir 一題。Tony，考量市場狀況與競爭態勢，感覺至少某些層面的補貼／激勵已經從系統中退出，或許是一些在利基品類中的小型玩家。你能談談你們是否正從較有利的營運環境中看到任何好處嗎？例如在下單頻率或目前的客戶獲取方面？

And then maybe, Prabir, I know you don't want to guide for 2023 specifically yet. But maybe how should we think conceptually about the 2023 EBITDA band. As your profit ports become larger, do you still see kind of big investment areas that would maybe keep the profit bands at these levels? Or can it gradually edge up?

然後 Prabir，我知道你現在還不想針對 2023 年提供具體指引。但我們應該如何從概念上思考 2023 年的 EBITDA 區間？隨著你們的獲利部位變大，你們是否仍看到一些大型投資領域，可能會讓獲利區間維持在目前水準？還是它可以逐步上移？

Deepak, it's Tony. Yes. Look, on the first question, our business has been always very competitive ever since we founded the company 9 years ago. And I think that what's been very impressive to us is just how resilient our business has been, both in light of competitive activities, both recent and in the years that we've been building DoorDash, as well as just the macro environment as we do see some consumer spending softening, but largely, we've been not impacted by that downward pressure.

Deepak，我是 Tony。是的。首先針對第一個問題，我們的業務自 9 年前創立公司以來一直都非常競爭。我想對我們而言非常令人印象深刻的是，我們的業務展現出很強的韌性——無論是在競爭活動方面（不論是近期或我們打造 DoorDash 這些年來的競爭），以及在總體環境方面：我們確實看到部分消費支出趨緩，但整體而言，我們並未受到那種下行壓力的明顯影響。

I really think that you see a lot of kind of the premise of your question reflected in our results. I mean, this is a quarter in which we beat top line as well as in bottom line, we grew results 25% GOV year-on-year and we beat quite handedly on the bottom line as well. And this is on top of a lot of the types of things that we're investing in. One that I will call out is that while we're not seeing any elevated pressures from certain types of incentives from competitors, what we are doing is we are taking care of our audiences.

我認為你問題中的許多前提，其實都反映在我們的業績之中。這一季我們不僅在營收端超出預期，在獲利端也同樣表現優於預期；我們的 GOV 年增 25%，而且在獲利端也明顯優於預期。這還是在我們投入許多投資項目的情況下。我想特別指出一點：雖然我們沒有看到來自競爭對手在某些激勵措施上的額外壓力，但我們正在做的是照顧好我們的受眾。

We invested over $40 million alone in the second quarter just to make sure that the Dashers who are on the road doing the hard work, can keep the profits that they expect to keep given some of the rising cost of fuel. And so we're not seeing any of the elevated pressures. We're seeing fairly normal activity on that side. And I think, as a result, given our industry-leading retention and order frequency, you continue to see our growth and our share gains.

僅在第二季，我們就投入超過 4,000 萬美元，確保在路上辛勤工作的 Dashers 能在燃油成本上升的情況下，仍能保有他們預期應得的收益。因此我們沒有看到任何升高的壓力；在那方面我們看到的是相當正常的活動。我認為也因此，憑藉我們在業界領先的留存率與下單頻率，你們持續看到我們的成長與市占提升。

Yes. And Deepak, on your second question around 2023, we're not providing quantitative 2023 guidance today. We will provide '23 guidance when we report Q4, which is on our normal cycle, but I will provide some sort of thematic context. The first thing I'd say is going into '23, we do expect our core U.S. restaurant business to grow and increase its contribution profit. At current course and speed, we plan to increase annual EBITDA by a modest amount. But note that this is after absorbing a full year of Wolt. So said differently, core DoorDash Wolt will expand into 2023, even though the consolidated EBITDA won't grow meaningfully. And the only caveat we'll make to this is just a change if we identify attractive growth opportunities, we continue to remain in investment mode and look for these opportunities that benefit our retention and order frequency. But at current course of speed, we would expect annual EBITDA to grow by a modest amount.

是的。Deepak，關於你第二個有關 2023 年的問題，我們今天不提供 2023 年的量化指引。我們會在公布第四季財報時提供 2023 年指引，這符合我們一貫的節奏，但我可以提供一些主題性的背景。首先我要說的是，進入 2023 年，我們預期美國核心餐廳業務將持續成長，並提高其貢獻利潤。以目前的方向與速度來看，我們計畫讓年度 EBITDA 小幅增加。但請注意，這是在吸收 Wolt 全年影響之後。換句話說，核心 DoorDash 與 Wolt 在 2023 年將擴張，即使合併後的 EBITDA 不會有顯著成長。我們唯一要補充的但書是：若我們辨識到具吸引力的成長機會，我們仍會維持投資模式並尋找能提升留存與下單頻率的機會。但以目前的方向與速度，我們預期年度 EBITDA 會小幅成長。

Your next question comes of the line of Lloyd Walmsley from UBS.

下一個問題來自 UBS 的 Lloyd Walmsley。

Two, if I can. In the letter, you guys talked about being a logistics led marketplace, I guess, how much room do you all see over the next few years to reduce cost per order? What are some of the key drivers to kind of get there? And like where is the lowest hanging fruit?

如果可以我有兩個問題。在股東信中，你們談到自己是一個以物流為主導的市集。我想問，未來幾年你們認為每筆訂單成本還有多少下降空間？要達成這點的關鍵驅動因素是什麼？以及最容易先做、也最有立即效果的部分在哪裡？

And then second one would just be you've talked in the past about how you don't try to beat EBITDA, you really try to kind of come in, in the range. And if you beat it, it's because there just weren't attractive investment -- like was there like something changing in the marketing landscape or any reason you guys let it flow through or just a function of some of the kind of inflation coming through as a surprise and you didn't have time to reinvest it. Anything you can share there would be great.

第二個問題是，你們過去談過你們不會刻意去「超越」EBITDA，而是希望落在區間內；如果超出，通常是因為沒有具吸引力的投資機會——那這次是否有行銷環境的變化，或任何原因讓你們讓它直接反映到獲利？還是只是某些通膨因素意外出現，你們沒有時間把它再投入？若能分享一些背景會很有幫助。

Lloyd, it's Tony. I'll take the first question and maybe Prabir can take the second part of the question. With respect to efficiency gains from logistics. I think this has been a hallmark of the DoorDash story and also of the Wolt story now that we've officially closed that partnership, in which if you think about the game that we're playing or the business that we're in, we're really in the game of building a minimum-efficient scale business. That's the type of business when you're talking about a hyper local business where order density is the most important metric. And in order to achieve that, you kind of have to do 2 things. One of the things you talked about in your question, which is really around high-quality logistics efficiency. And the other is high-quality retention and order behavior from customers without discounting and that sort of activity.

Lloyd，我是 Tony。我先回答第一個問題，第二部分可能請 Prabir 回答。關於物流帶來的效率提升，我認為這一直是 DoorDash 故事的核心，也是 Wolt 故事的核心——現在我們已正式完成這項合作。若你思考我們正在玩的這場遊戲或我們所處的業務，本質上是在打造一個達到最低效率規模（minimum-efficient scale）的業務。當你談的是高度在地化（hyper local）的業務時，訂單密度是最重要的指標。為了達成這點，你基本上需要做兩件事：一件是你在問題中提到的，也就是高品質的物流效率；另一件則是在不依賴折扣或類似活動的情況下，維持高品質的客戶留存與下單行為。

And DoorDash has really achieved both. And on the part around gaining extra logistics efficiency, we continue to find areas of opportunity. I mean, we've certainly been leaders in our space up to this point. But I still see massive room to keep increasing the selection on our platform and improve the logistics quality on our platform, the affordability of our service and certainly, our customer service levels. On logistics more specifically, whether it's working on efficiency improvements at the store as well as how we think about how we ought to assign orders, especially now that we're entering multiple categories of deliveries in addition to restaurant deliveries, I think there's a long room to go, and we're seeing that both in our numbers in the most recent quarter as well as in the quarters leading up to this point.

而 DoorDash 其實兩者都做到了。至於進一步提升物流效率的部分，我們仍持續找到機會領域。我的意思是，到目前為止我們確實一直是這個領域的領先者，但我仍看到很大的空間，可以持續增加平台上的選擇、提升平台上的物流品質、提升我們服務的可負擔性，當然也包括提升客戶服務水準。更具體地談物流，不論是改善門店端的效率，或是我們如何思考訂單指派——尤其是現在除了餐廳外送之外，我們也進入多個品類的配送——我認為仍有很長的路要走；而我們在最近一季以及此前幾季的數據中也都看到了這一點。

And Lloyd, on the second part of your question. First, I'll talk about Q2 and then I'll talk about sort of general philosophy. In terms of the Q2 performance and EBITDA, it was really a matter of 2 factors. The first is subtotals as we described in the letter as a result of consumer price inflation, our sub totals were higher because of higher item prices. And that then translates into higher commission dollars as well as higher service fees, both of which benefited revenue and drop through to EBITDA. So that's what drove the upside in Q2 in terms of EBITDA.

Lloyd，關於你問題的第二部分。首先，我會先談 Q2，接著再談比較一般性的理念。就 Q2 的表現與 EBITDA 而言，主要是兩個因素所致。第一個是我們在信中所描述的「小計」：由於消費者價格通膨，因為單品價格較高，我們的小計金額也更高。這也進一步轉化為更高的佣金收入以及更高的服務費，兩者都帶動了營收並流入 EBITDA。因此，這就是 Q2 在 EBITDA 方面優於預期的驅動因素。

In terms of general philosophy, I'd say EBITDA is a function of really 2 things. First, it's the margin expansion that we continue to drive in our U.S. restaurant business. That's one. And the second is the level of investment we make which is discretionary, by the way, the level investment we make in our new verticals and in our international businesses. That level of investment varies from quarter-to-quarter based on the signals we see. And you can see that volatility in our historical trends where some quarters will produce more EBITDA than others because depending on what we see in terms of retention and order frequency, we might invest more or less. And so our EBITDA range is really meant as a guardrail, but where we land within that range based depending on the level of investment.

就一般性的理念而言，我會說 EBITDA 主要取決於兩件事。第一，是我們在美國餐廳業務中持續推動的利潤率擴張。這是一點。第二，是我們在新垂直領域以及國際業務上的投資水準；順帶一提，這些投資是可自行決定的。我們的投資水準會依據所看到的訊號而在每一季有所變動。你也可以從我們過去的趨勢中看到這種波動：有些季度的 EBITDA 會高於其他季度，因為視我們在留存與下單頻率方面看到的情況，我們可能會投資更多或更少。因此，我們的 EBITDA 區間主要是作為一個護欄，而我們落在區間內的哪個位置，則取決於投資水準。

Your next question comes from the line of Youssef Squali from Truist Securities.

下一個問題來自 Truist Securities 的 Youssef Squali。

Congrats on a really impressive quarter, all things considered. My question is around Wolt contribution. Can you maybe speak to the contribution that you are baking in into the 2022 top and bottom line, how do you think about growth there considering kind of what's going on in Europe? And just broadly speaking, maybe Tony, can you address the issue about how do you kind of take kind of a relatively subscale business in Europe that's across multiple geographies and kind of grow it meaningfully for it to become a big part of the business. Your position in the U.S. has been quite the opposite. You've been market leader, market dominant in the single market. So any help there would be great.

在各種因素都納入考量下，恭喜你們交出非常亮眼的一季。我的問題是關於 Wolt 的貢獻。你們能否談談在 2022 年的營收與獲利指引中，已經納入了多少 Wolt 的貢獻？考量到歐洲目前的情況，你們如何看待那邊的成長？另外更廣泛地說，Tony，你能否談談你們如何把一個在歐洲、跨多個地理區域、規模相對較小的業務，實質性地做大，讓它成為業務的重要部分？你們在美國的情況剛好相反：你們是單一市場的領導者、甚至是主導者。若能提供一些看法會很有幫助。

Maybe I'll start and then Tony can chime in. I mean the Wolt business grew over 50% year-on-year, and that's in stark contrast to what we've seen with other companies that operate in the similar geographic region. And really, it all comes back down to 2 things. The fundamentals of the Wolt business in terms of its industry-leading retention and growing order frequency. And what we've seen is as long as you create a product that has high retention and order frequency, ultimately translates into better growth because you're retaining your customers better and that benefit compounds versus alternatives where you end up losing customers because of poor retention.

我先回答，然後 Tony 再補充。我是說，Wolt 的業務年增超過 50%，這與我們在同一地理區域內看到其他公司的表現形成鮮明對比。而這一切其實都回到兩件事：Wolt 業務的基本面——其業界領先的留存率，以及持續提升的下單頻率。我們看到的是，只要你打造出一個具有高留存與高下單頻率的產品，最終就會轉化為更好的成長，因為你能更好地留住客戶，而這種效益會複利累積；相較之下，若留存不佳，你最終會因為客戶流失而受影響。

And where we're at in terms of the investment cycle, the Wolt markets are relatively new, both in terms of merchant adoption as well as user adoption. As an example, even in its oldest markets, adoption levels are less than 10% of the population. And that says, there's a lot of room to grow just as you grow your footprint within these countries as well as you go to the merchant selection and grow the consumer base. And so that's driving the investment we're making. We'll continue investing as long as we see strong signal on retention and order frequency.

至於我們目前所處的投資週期，Wolt 的市場相對較新，不論是在商家採用度或用戶採用度方面都是如此。舉例來說，即便在其最早的市場中，採用率也不到人口的 10%。這表示仍有很大的成長空間：你可以在這些國家擴大覆蓋範圍，也可以擴充商家選擇並擴大消費者基礎。因此，這也驅動了我們正在進行的投資。只要我們看到留存與下單頻率的訊號依然強勁，我們就會持續投資。

Yes. And on the second part of the question, I think it's important to start with maybe some historical context and go down memory lane not just a few years ago, DoorDash was certainly not the market leader. And DoorDash especially 5, 6 years ago was quite capital constrained relative to any of its peers by a pretty far margin. And so how was it possible that a "subscale" company was able to rise to market leadership. Well, it really was mastering the order level execution of the business. And again, this is a hyper local business for order density and achieving minimum efficient scale through leading retention order frequency, which is really measured in whether or not you built a superior product of selection quality, price and service.

是的。關於問題的第二部分，我認為重要的是先從一些歷史背景談起，回顧一下不只是幾年前的情況：DoorDash 當時當然不是市場領導者。尤其在 5、6 年前，相較於任何同業，DoorDash 的資本相當受限，而且差距還不小。那麼，一家「規模較小」的公司怎麼可能崛起成為市場領導者？關鍵在於把業務的「單筆訂單層級」執行做到極致。再說一次，這是一個高度在地化的生意，取決於訂單密度，以及透過領先的留存與下單頻率來達到最低有效規模；而這些最終反映在你是否打造出在選擇、品質、價格與服務上更優越的產品。

And then on the other end, whether or not you have the most capital-efficient logistics operations. And I think when I look at the elements that caused DoorDash to rise to market leadership, I find very similar kind of characteristics in the Wolt business, which is what excited us about the business not only today, as Prabir mentioned, far outgrowing some of their European peers. But maybe much more excited are we in their potential? Because when I compare the foundation of what they've built, and I look at the opportunity ahead of them, I mean, even in their oldest markets, Wolt is less than 10% penetrated. On a global basis, both DoorDash and Wolt in our non-U.S. markets are less than 5% of restaurant sales and outside of restaurants, less than 1% of nonfood spend. And so when I compare the foundation that Wolt has built on one hand. And on the other hand, compared it to the opportunity ahead, it's exactly what Prabir said, which is that it's absolutely the right time to invest. And I think you already see evidence of this as Wolt has become a market leader in many of their markets already.

另一方面，還要看你是否擁有資本效率最高的物流營運。我認為，當我回顧促使 DoorDash 成為市場領導者的要素時，我在 Wolt 的業務中看到了非常相似的特質，這也是我們對該業務感到興奮的原因——不僅是因為如 Prabir 所提到的，它目前的成長遠超一些歐洲同業；更讓我們興奮的是它的潛力。因為當我比較他們所建立的基礎，以及他們面前的機會時，我的意思是，即便在其最早的市場，Wolt 的滲透率也不到 10%。從全球角度來看，DoorDash 與 Wolt 在我們的非美國市場中，餐廳銷售的滲透率都不到 5%；而在餐廳以外的領域，非食品支出的滲透率不到 1%。因此，當我一方面看 Wolt 已建立的基礎，另一方面再看前方的機會，正如 Prabir 所說，現在絕對是投資的正確時機。而且你其實已經看到證據：Wolt 已在許多市場成為領導者。

That's great. And Prabir, did you quantify the contribution to that $51 billion to $53 billion in GOV from Wolt.

很好。Prabir，你有量化 Wolt 對於該 $510 億到 $530 億 GOV 指引的貢獻嗎？

We don't break out the guidance between Wolt and DoorDash, but when we do report earnings, we will split out organic versus Wolt's contribution.

我們不會在指引中拆分 Wolt 與 DoorDash 的部分，但在我們公布財報時，會拆分有機成長與 Wolt 的貢獻。

Your next question comes from the line of Bernie McTernan from Needham.

下一個問題來自 Needham 的 Bernie McTernan。

We've noticed a lot of grocers in the marketplace doing delivery and online ordering in-house in addition to being opened up to third-party marketplaces like yourself. So they're a competitor but also a partner. Can you talk about some of the advantages that you have for customer acquisition and retention relative to them? And then also with the 50% growth in Wolt this past quarter for GOV, 50% higher probably than some of the other European delivery operators. Can you just remind us whether it's market growth or just pure market share gains in terms of what's driving this growth?

我們注意到市場上有很多雜貨商除了加入像你們這樣的第三方平台外，也在內部自行做配送與線上下單。因此他們既是競爭者也是合作夥伴。你能談談相較於他們，你們在客戶獲取與留存方面有哪些優勢嗎？另外，Wolt 在上一季的 GOV 成長 50%，可能比其他一些歐洲外送業者高出 50%。你能否提醒我們，這個成長主要來自市場整體成長，還是純粹的市占提升？也就是說，究竟是什麼在驅動這個成長？

Sure. Maybe I'll take the first question, and I'll let Prabir take the second question. So I'm not sure this touches exactly the spirit of your question, but we view all merchants, including grocers as partners. I mean if you think about the mission of the company, is to, on one hand, build the largest local commerce marketplace, where we're driving incremental demand to all of these retailers, whether they be restaurants, grocers, convenience stores, other types of retail stores. And then on the other hand, build the largest local commerce platform where we give the tools, whether it's logistics as a service in the form of DoorDash drive or ordering of the service in the form of DoorDash Storefront all of these retailers so that they can build their own digital operations. So we really view in equal parts, our mission to help grow on one hand, which is to bring that incremental demand. And on the other hand, empower them to do it on their own.

當然。也許我先回答第一個問題，第二個問題我讓 Prabir 來回答。我不確定這是否完全切中你問題的精神，但我們把所有商家（包含雜貨商）都視為合作夥伴。若你想想公司的使命：一方面，是打造最大的在地商務市集，為所有這些零售商帶來增量需求——不論是餐廳、雜貨商、便利商店或其他類型的零售店。另一方面，是打造最大的在地商務平台，提供工具給所有這些零售商，讓他們能建立自己的數位化營運；這些工具包括以 DoorDash Drive 形式提供的「物流即服務」，或以 DoorDash Storefront 形式提供的「下單服務」。因此，我們的使命在兩個面向上同等重要：一方面協助他們成長，也就是帶來增量需求；另一方面賦能他們自行完成這件事。

In fact, we see these customers on these different channels to be quite different. I mean if you think about it, if a customer is quite used to and knows exactly what they want to order from a particular retailer, it probably makes quite a lot of sense for them on that occasion to actually order directly from the retailer. But on the greatest privilege that we have in this business is that people eat 20 to 25 times a week, and they shop even more times than that on top of that when you consider their nonfood spend. And so for other occasions, maybe when they're not exactly sure what it is that they're looking for, but they want to buy something from the neighborhood, that's where the DoorDash marketplace really comes in hand. And so we really view the retailer, the merchants, whether it's a grocer or other type of store as our partner, and we have 2 ways in which we help them.

事實上，我們看到這些客戶在不同渠道上的行為差異相當大。我的意思是，如果你想想看，若一位客戶非常習慣、也非常清楚自己想從某個零售商訂購什麼，那麼在那個情境下，直接向該零售商下單其實很合理。但我們在這個業務中最大的優勢在於，人們每週會吃 20 到 25 次；而且當你把非食品支出也算進去，他們購物的次數甚至更多。因此在其他情境下，也許他們不太確定自己要找的是什麼，但又想從鄰里買點東西，這時 DoorDash 市集就真正派上用場了。所以我們確實把零售商、商家——不論是雜貨商或其他類型的店家——視為合作夥伴，而我們有兩種方式來協助他們。

And Bernie, on your second question around GOVs growth, it's really driven by 2 things. It's MAU growth that was aided by some customer acquisition, but also by just by higher retention compared to the other players in the market. And then second, order frequency growth, as Wolt brings on more selection, both in the restaurant as well as the non-restaurant category. We're seeing some of the same things that we saw in the U.S. where order frequency at a cohort level has continued to grow. And both of those things, particularly in the face of some of the more extreme COVID reversion version that you saw in Europe have led to the 50% year-on-year growth that you cited.

Bernie，關於你第二個問題、也就是 GOV 成長的部分，主要是由兩件事驅動。第一是 MAU 成長，這一方面受惠於一些客戶獲取，但也來自於相較於市場上其他玩家更高的留存率。第二是訂單頻次成長，因為 Wolt 持續引入更多選擇，包含餐廳以及非餐廳類別。我們看到一些與美國相同的現象：在同一批次（cohort）層級，訂單頻次持續成長。而這兩點，特別是在你在歐洲看到較為極端的 COVID 回歸（reversion）背景下，促成了你所提到的年增 50%。

Your next question comes from the line of Brian Nowak from Morgan Stanley.

下一個問題來自 Morgan Stanley 的 Brian Nowak。

I have 2. Just the first one, the gross margins in the quarter were a little weaker than we thought, at least. Could you just sort of talk to us about some of the puts and takes on gross margins of the business? And how should we think about the gross margins of the Wolt business sort of going into the back half and maybe even longer term 2023.

我有兩個問題。第一個是，本季的毛利率比我們預期的稍微弱一些。你能否跟我們談談，這項業務毛利率的一些正負因素（puts and takes）？以及我們應該如何看待 Wolt 業務在下半年、甚至更長期到 2023 年的毛利率走勢？

Brian, it's Prabir. Let me start on both of your questions. So first, our cost of sales increase that we saw on a year-on-year basis was only driven by 2 things: the DashMart and insurance costs. on DashMart, as we've launched more and more DashMart, the costs associated with those orders impacts cost of sales. So roughly half of the increase in cost of sales as a percentage of GOV is driven by DashMart. The other half roughly is driven by an increase in insurance costs, which was in line with our expectations. I'll point you to our comments from last quarter where we said, we experienced an increase in insurance reserves due to the outside impact of a few large claims. And we began to take actions to improve the safety on our platform starting last quarter. But it takes a while for these changes to reflect on our claims data. And so as we previously discussed, we expect insurance costs to increase in the near term.

Brian，我是 Prabir。我先回答你這兩個問題。首先，我們看到的銷貨成本（cost of sales）年增，僅由兩個因素驅動：DashMart 與保險成本。就 DashMart 而言，隨著我們推出越來越多 DashMart，與這些訂單相關的成本會影響銷貨成本。因此，銷貨成本占 GOV 比例的上升，大約有一半是由 DashMart 驅動。另一半大致上是由保險成本上升所驅動，這符合我們的預期。我也想提醒你參考我們上季的評論：我們表示，由於少數幾筆大型理賠的外部影響，我們的保險準備金有所增加；並且我們從上季開始採取行動，以提升平台安全性。但這些變更需要一段時間才會反映在理賠數據上。因此如同先前討論，我們預期短期內保險成本將會上升。

On the whole note, both the DashMart cost increase as well as the insurance costs, is reflected in the guidance we've provided. And so it's not incremental to the guidance. So it's already reflected in that. In terms of gross margin for Wolt, really Wolt a combination of, as I view it, so 3 distinct pieces, there's investments that are going into Japan and Germany. These are brand-new markets that we launched less than 2 years ago were to get the flywheel going in terms of selection and quality and price. You've got investments in Wolt markets, which is the equivalent of our DashMart Wolt first-party distribution business. And then you've got the rest, what I call core food delivery. Core food delivery has had consistent increase in gross margin levels. And as we continue driving efficiency and order density at the local level, you start seeing the same type of efficiency in terms of Dashers costs kick in to produce increasing levels of gross margin. Japan and Germany and Wolt markets remain in investment mode and we continue to work on coverage selection quality and price side.

整體而言，DashMart 成本上升以及保險成本，都已反映在我們提供的財測指引中，因此並非指引之外的額外增量；也就是說，已經包含在其中。至於 Wolt 的毛利率，我的看法是，Wolt 其實由三個明確部分組成：第一，是對日本與德國的投資。這些是我們在不到兩年前才推出的全新市場，目標是讓飛輪在選擇、品質與價格等面向運轉起來。第二，是對 Wolt markets 的投資，這相當於我們的 DashMart——也就是 Wolt 的第一方配送（first-party distribution）業務。第三，是其餘部分，我稱之為核心外送（core food delivery）。核心外送的毛利率水準一直在穩定提升。隨著我們持續在本地層級提升效率與訂單密度，你會開始看到類似於 Dasher 成本效率的效果發揮，進而帶動毛利率持續上升。日本、德國以及 Wolt markets 仍處於投資模式，我們會持續在覆蓋、選擇、品質與價格等面向努力。

Your next question comes from the line of Eric Sheridan from Goldman Sachs.

下一個問題來自 Goldman Sachs 的 Eric Sheridan。

Maybe 2, if I can. First, Tony, I know you're talking a lot in the shareholder letter about not seeing anything yet on the consumer spending behavior patterns. But I think one of the questions we all get a lot from investors is, you're seeing some large companies already say how they would run their business differently, if the consumer spending environment did change. Is there any sense you can give us on how you're sort of game planning out different economic scenarios for the business and how you might change investment philosophies or growth philosophies for the company if you did hit a rougher patch in terms of consumer spending and went through the platform? And then maybe question on grocery and convenience at some of the new categories. Can you give us a little bit of sense of folks who use multiple products across the platform, what that might mean in terms of LTV how you're thinking about leaning in and promoting different category expansion on a per customer basis what that might do for the unit economics longer term?

如果可以的話，我也有兩個問題。第一個，Tony，我知道你在致股東信中談了很多，表示目前尚未看到消費支出行為模式有任何變化。但我想投資人常問我們的一個問題是：你已經看到一些大型公司在說，如果消費支出環境真的改變，他們會如何以不同方式經營業務。你能否分享一下，你們如何針對不同經濟情境進行沙盤推演（game planning），以及如果消費支出遇到更艱難的階段並影響到平台，你們可能會如何調整公司的投資理念或成長理念？第二個問題是關於雜貨與便利等新類別。你能否談談在平台上使用多種產品的用戶，這對 LTV 可能意味著什麼？以及你們如何思考在單一客戶層級加大投入並推動不同類別擴張，這對長期單位經濟（unit economics）可能帶來什麼影響？

Eric, it's Prabir. Maybe I'll start with the first one around how we're going to manage the business and EBITDA impact of growth versions and Tony can talk about the impact from some of the new categories. So maybe the first place to start is our core U.S. restaurant business is growing and continues to generate significant cash flow. And historically, we've taken that cash flow and invested the vast majority of it in order to grow our scale in the large and underpenetrated categories in which we operate. Again, as I said earlier, these investments are discretionary. And so that means the reason we're investing is because we continue to see strong signals of product market fit as well as improving unit economics.

Eric，我是 Prabir。我先從第一個問題開始，談談我們將如何管理業務，以及成長版本（growth versions）對 EBITDA 的影響；Tony 可以再談新類別帶來的影響。首先可以從一點開始：我們美國核心餐廳業務正在成長，並且持續產生可觀的現金流。歷史上，我們把這些現金流的絕大部分投入，用以擴大我們在所營運、且滲透率仍偏低的大型類別中的規模。再次強調，如我先前所說，這些投資是可自行決定的（discretionary）。也就是說，我們之所以投資，是因為我們持續看到強烈的產品市場契合（product-market fit）訊號，以及單位經濟正在改善。

Just to give you an example, if I -- on any given day, if I look at the top 10 stores on DoorDash in terms of sales, DashMart shows up there, and that's a strong signal product market fit. To the extent that we don't see continued improvements in terms of unit economics or continued improvements in retention and order frequency, we will alter the pace of our investment. We've been very disciplined in terms of capital allocation so far and we'll continue to do so going forward.

舉例來說，如果我——在任何一天——去看 DoorDash 上按銷售額排名的前 10 家店，DashMart 都會出現在其中，這就是產品市場契合的強烈訊號。如果我們沒有看到單位經濟持續改善，或沒有看到留存與訂單頻次持續提升，我們就會調整投資節奏。到目前為止，我們在資本配置上一直非常自律，未來也會持續如此。

Yes. And I think if there's add a little bit in terms of just my view on the macro environment. And then I'll hit your second question about the multi-category customers. So obviously, we've been looking, right, in terms of how the tough macroeconomic headwinds that I think is hitting a lot of industries, how that might apply to us. We've been searching for this for many quarters now. And I think so far, the reason why largely we've been less impacted is because, well, one, our product is dynamic. It's been constantly improving. If you look at selection, for instance, in the 12 months leading up to the end of the second quarter, we've added 80,000 net new stores onto the platform. made improvements to many quality metrics in terms of our delivery experiences, whether it's speed or accuracy and other types of improvements. We've made many improvements to the shopping experience, so lower the friction for consumers. And so I think that's one point.

是的。我想我可以補充一點我對總體環境的看法，然後再回答你第二個關於多品類客戶的問題。很明顯地，我們一直在觀察：我認為正衝擊許多產業的嚴峻總體經濟逆風，可能會如何影響我們。我們已經連續好幾個季度在尋找這方面的跡象。到目前為止，我們之所以大致上受到的影響較小，原因在於：第一，我們的產品是動態的，且一直在持續改進。以選擇（selection）為例，在截至第二季末的前 12 個月，我們在平台上淨新增了 80,000 家商店。我們也在配送體驗的多項品質指標上有所改善，不論是速度、準確性或其他類型的提升。我們也對購物體驗做了許多改進，降低消費者的摩擦成本。因此我認為這是一點。

The second point is, I think we still have to remember that relative, especially to other maybe categories of commerce or e-commerce. We are still very early in our penetration even as the market leader just take the U.S. as one example, we're less than 8% of total restaurant industry sales. If you compare that to other categories of commerce, it is much earlier in its evolution. And then the final thing I would just add is just if you study macro, as we've been looking at macroeconomics, I think there's only been a couple of years in history in which food spend has actually declined due to challenging macroeconomic pressures. And I think that because it is less of a discretionary spend relative to other categories of spend. But that said, to Prabir's point, we're equal opportunity growth investors that are very disciplined. DoorDash historically hasn't had a lot of resources. And so we take very seriously every dollar of spend. And as you saw in the second quarter, if we don't think that there is a great investment to be made over the same time period to generate a great return. We're not going to make that investment. And so that's true with all new projects. That's true with marketing investments. That's true with engineering and product investments. That's true with headcount. That's really true for every line item in the P&L.

第二點是，我認為我們仍必須記得，相較於其他可能的商業或電商類別，我們的滲透率仍然非常早期。即使作為市場領導者，以美國為例，我們在整體餐飲產業銷售中的占比仍不到 8%。若與其他商業類別相比，這仍處於更早的演進階段。最後我想補充的是，如果你研究總體面（我們一直在看總體經濟），歷史上真正因為具挑戰性的總體經濟壓力而使食品支出下降的年份其實只有少數幾年。我認為這是因為相較於其他支出類別，食品支出較不屬於可自由裁量的支出。不過話說回來，呼應 Prabir 的觀點，我們是機會均等的成長型投資者，而且非常自律。DoorDash 過去並沒有很多資源，因此我們非常重視每一美元的支出。正如你在第二季看到的，如果我們認為在同一時間期間內沒有值得投入、且能帶來很高回報的投資，我們就不會做那項投資。這適用於所有新專案、行銷投資、工程與產品投資、以及人力編制。基本上，這對損益表（P&L）上的每一個科目都成立。

All right. So moving on to your second question, which I think was about the impact of consumer shopping in multiple categories. Well, one, this is just part of the mission of the company is to make sure we bring everything inside the neighborhood, not just from restaurants. And two, the last disclosure we made was, I think, in the fourth quarter, where we said that about 14% of our customers are now shopping in these non-restaurant categories. And we are seeing higher retention and order frequency activity from these customers who are engaging in multiple categories. I think this makes quite a lot of sense, as we're solving now different jobs and tasks for the customer. But that said, look, we still have to earn every inch. We have a long ways to go in terms of the product experience in each one of these categories before we'd be satisfied with that behavior.

好。接著談你的第二個問題，我理解是關於消費者在多個品類購物的影響。第一，這其實就是公司使命的一部分：確保我們把社區裡的所有東西都帶進來，不只是餐廳。第二，我們上一次揭露是在第四季，當時我們說大約有 14% 的客戶現在會在這些非餐廳品類中購物。我們也看到，這些跨多品類互動的客戶具有更高的留存率與下單頻率。我認為這很合理，因為我們現在是在替客戶解決不同的工作與任務。不過話說回來，我們仍必須一步一腳印地爭取。我們在這些品類各自的產品體驗上，距離讓我們對這種行為感到滿意，還有很長的路要走。

Your next question comes from the line of Andrew Boone from JMP Securities.

你的下一個問題來自 JMP Securities 的 Andrew Boone。

I know you guys talked earlier about the logistics benefits that you guys are running through the platform, but can you double-click on the drivers of Dasher cost savings that you guys highlighted in the letter. And then we haven't talked about drive in a while. Can you provide an update there? Is there any change that you're seeing in terms of enterprise adoption now that we're kind of beyond peak COVID?

我知道你們先前談到平台上正在推進的物流效益，但你能否更深入說明一下你們在信中強調的 Dasher 成本節省的驅動因素？另外，我們有一段時間沒談 Drive 了。你能提供那邊的最新進展嗎？在我們某種程度上已經走過 COVID 高峰之後，你們在企業端採用（enterprise adoption）方面有看到任何變化嗎？

Andrew, maybe I'll start with the Dasher cost question, and then Tony can take the one on our Platform Services. So in terms of Dasher cost, I mean, at the end of the day, right, there's multiple components of this, including as Tony alluded to earlier, how stores operations in terms of how quickly they get the Dasher in and out. So there's opportunities to continue optimizing that there's opportunities to continue dispatching from closer and closer Dashers, which by the way, as the density of your network increases as you get more orders occurring within a certain neighborhood within a certain store, you've got the ability to not just batch, but you've got the ability to get the Dasher to the store quicker than you otherwise might have.

Andrew，也許我先回答 Dasher 成本的問題，然後 Tony 可以回答我們平台服務（Platform Services）的部分。就 Dasher 成本而言，歸根結底，這裡面有多個組成要素，包括 Tony 先前提到的門店營運，例如門店能多快讓 Dasher 進出取餐。這方面仍有持續優化的空間；也有機會把派單（dispatching）做得更精準，讓更近的 Dasher 接單。順帶一提，當你的網路密度提高、在某個社區、某家店周邊有更多訂單發生時，你不僅能做併單（batch），也能讓 Dasher 比原本更快抵達門店。

And so there's really -- if you think about the sources of opportunities really reducing the amount of time it takes for Dasher to get to the store and reducing the amount of time that Dasher spends in the store, and we continue to work both those levers and that's what resulted in the improvement in terms of Dasher costs that we're seeing. The second thing I'll say, which is really just to clarify, last year was anomalous in terms of Dasher cost. Because we were operating in a very expensive labor environment that's fueled by fiscal stimulus. So in some ways, Dasher cost this year just at least in my mind, normalizing back to historical levels was ending incremental. So last year was elevated this year isn't.

因此，如果你思考機會來源，核心就是：降低 Dasher 抵達門店所需的時間，以及降低 Dasher 在店內停留的時間。我們持續同時拉動這兩個槓桿，這也帶來了我們目前看到的 Dasher 成本改善。第二點我想補充、也算是釐清：去年在 Dasher 成本上屬於異常（anomalous）。因為我們是在一個非常昂貴的勞動力環境下運作，而那是由財政刺激所推動。所以某種程度上，今年的 Dasher 成本在我看來只是回歸到歷史水準的正常化，而不是額外的增量改善。也就是說，去年偏高，今年並沒有。

Yes. And with respect to the second question around our platform services products such as Drive, we see continued excitement for the stores. But I mean, in many ways, our Platform Services business experiences similar seasonality in our marketplace business where the second and third quarters are generally more muted in activity. And that's mostly because as customers are back out, many of whom are taking advantage of the good weather as well as perhaps lost vacations from the 2 years of COVID and now eating out again or visiting retail stores again that these stores have to make sure that their in-store activity is protected and taken care of and the customer service levels are exceptional before they invest aggressively in their off-premise business.

是的。至於第二個關於我們平台服務產品（例如 Drive）的問題，我們看到門店端仍持續感到興奮。不過在很多方面，我們的平台服務業務也呈現與市集業務（marketplace business）相似的季節性：第二季與第三季的活動通常較為平淡。這主要是因為客戶回到戶外活動，許多人利用好天氣，也可能補回過去兩年 COVID 期間錯過的假期，現在又開始外出用餐或再次造訪零售門店。這些門店在更積極投資其店外（off-premise）業務之前，必須先確保店內活動能被妥善維持與照顧，並且客戶服務水準要非常出色。

So while there's some seasonal kinds of activity happening in that business, I think the COVID highs in terms of the excitement to invest and continue to accelerate the momentum behind e-commerce for all of these retailers across any category remains just as high as ever. And so that's something that we expect to continue to help grow our Platform Services business, whether they're large merchants like some of the ones you mentioned who participate with products like DoorDash Drive or smaller merchants that really need help getting online for the first time and growing their kind of same-store sales off-premise with products like DoorDash Storefront. And so our focus right now is making sure that those products can be easier to use and that we can build products to help teach the playbooks that we've used to build a successful digital marketplace in our own right such that these businesses can do it other out.

因此，雖然該業務存在一些季節性的活動變化，但我認為，COVID 期間那種對投資、以及持續加速各類零售商電商動能的熱度，依然一如既往地高。我們預期這將持續推動我們的平台服務業務成長：不論是你提到的那類大型商戶，透過 DoorDash Drive 等產品參與；或是需要首次上線、並透過 DoorDash Storefront 等產品來成長其店外同店銷售（same-store sales off-premise）的小型商戶。我們目前的重點是確保這些產品更容易使用，並且打造產品來傳授我們用以建立成功數位市集的作戰手冊（playbooks），讓這些企業也能在自身場域把它做起來。

Your next question comes from the line of Ron Josey from Citi.

你的下一個問題來自 Citi 的 Ron Josey。

Maybe 2, please. The chart in the note in the letter that talks about existing consumer order rates. 2022 is trending higher than previous years and suggesting that these users in 2022 are more engaged. I'm sure -- can you just talk about the drivers here? Is it greater repeat rates, adoption of newer verticals, impacted DashPass, I'm sure it's all of the above. And so maybe the bigger question is just are these newer users, this newer cohort of users, talk about how that compares to the prior cohorts? Are they just doing more?

我想請教兩個問題。信中附註的那張圖表談到既有消費者的下單率。2022 年的趨勢高於前幾年，顯示這些 2022 年的用戶更投入。我確定——你能談談背後的驅動因素嗎？是更高的回購率、較新的垂直品類採用、DashPass 的影響——我想應該都是。那更大的問題是：這些新用戶、這個新的用戶世代（cohort）與先前的 cohorts 相比如何？他們就是做得更多嗎？

And then maybe Prabir as a quick as a quick follow-up. You mentioned just a strong cash flow generated from the U.S. restaurant business. Any way to provide some guideposts or insight in terms of that profitability of that U.S. restaurant business?

另外也許請 Prabir 快速追問一下。你提到美國餐廳業務帶來強勁的現金流。是否能提供一些參考指標或洞見，說明美國餐廳業務的獲利能力？

Ron. So on -- let me start with the second one first. On the U.S. restaurant profitability, we're not actually -- we haven't provided any disclosure to break that out other than to say that continues generating more contribution profit and has improved both in terms of its net revenue margins as well as its contribution profit as a percentage of GOV on a year-on-year basis. So we're happy with the progress there.

Ron，我先從第二個問題開始。關於美國餐廳業務的獲利能力，我們其實沒有提供任何拆分揭露；我們僅表示它持續產生更多的貢獻利潤（contribution profit），並且其淨營收利潤率（net revenue margins）以及以 GOV 計的貢獻利潤占比（contribution profit as a percentage of GOV）都在年對年基礎上有所改善。我們對這方面的進展感到滿意。

And as we've said previously, it's a valuable funding source that we used to make investments in these new categories that we're growing.

而且如同我們先前所說，這是一個很有價值的資金來源，我們用它來投資於正在成長的這些新類別。

In terms of your question on the order rate chart, really by the way, the purpose of that chart was to demonstrate the fact that the order rate trends were similar to what we've seen historically. And to make the point that increased inflation as well as potentially weakening consumer spending has not had an adverse impact relative to old cohort. And so it's hard for us to untangle the impact of consumer spending from ordinary cohort seasonality. But taking a step back in terms of the 2022 cohorts, we're happy with the quality so far. Remember, we run our sales and marketing to a payback period, and we've been operating within the same payback period for quite some time now. So what's driving that is not just enhanced order frequency, which has continued growing on a year-on-year basis as a result of improvements in selection, quality and price as well as the new categories that we're adding, but also improvements in terms of the core margin structure of the product. And so the core product has gotten more profitable, in part, due to sub totals, but also in part because of improvements in terms of Dasher cost the quality of our orders, which improves our customer support costs and so on. And that, coupled with the order frequencies contributed to better LTVs and roughly the same payback.

至於你對下單率圖表的問題，順帶一提，那張圖表的目的其實是要展示：下單率的趨勢與我們過去歷史所見相似，並指出通膨上升以及可能走弱的消費支出，並未相較於舊 cohort 造成不利影響。因此，我們很難把消費支出的影響與一般 cohort 的季節性拆解開來。不過退一步看 2022 年的 cohorts，我們目前對其品質感到滿意。請記得，我們的銷售與行銷是以回收期（payback period）為目標來運作，而且我們在相同的回收期範圍內已經運作了相當長一段時間。因此，驅動因素不僅是下單頻率提升——這在年對年基礎上持續成長，原因包括選擇（selection）、品質與價格的改善，以及我們新增的新類別——也包括產品核心毛利結構（core margin structure）的改善。核心產品變得更有利潤，部分來自小計（sub totals），也部分來自 Dasher 成本的改善、訂單品質提升（進而降低客服成本）等等。這些因素再加上下單頻率，共同帶來更好的 LTV，以及大致相同的回收期。

Your next question comes from the line of Ross Sandler from Barclays.

您的下一個問題來自巴克萊（Barclays）的 Ross Sandler。

Two for me. Prabir, the 30% retention stat for Wolt, is that comparable to the 48% for core DoorDash? I know they don't have a subscription business built out yet and some of their markets are a little bit younger, so could you just walk us through that? And how does that 30% compare to some of the EU competitors that they face in their home markets? And do you think the gap is as wide between that 30% and the competition as it is between the 48% and your U.S. competition?

我有兩個問題。Prabir，Wolt 的 30% 留存率數據，是否可與核心 DoorDash 的 48% 相比？我知道他們尚未建立訂閱業務，而且部分市場相對較新，能否請您帶我們梳理一下？另外，這個 30% 與他們在本土市場面對的一些歐盟競爭對手相比如何？您認為 30% 與競爭對手之間的差距，是否像 48% 與你們美國競爭對手之間的差距那麼大？

And then the second question is just you guys gave a good overview of how the consumer is holding up. I'm just curious, is the frequency for DashPass kind of stand-alone holding up as well as non-DashPass? I know there's a mix shift towards DashPass, which is driving up frequency. But could you just talk about those 2 kind of separately.

第二個問題是，你們對消費者韌性的概況說明得很清楚。我想了解的是，DashPass 的使用頻率（frequency）本身是否也像非 DashPass 一樣維持得不錯？我知道訂單結構正朝 DashPass 傾斜，這會推升整體頻率。但能否把這兩者分開談談。

Yes. Ross, maybe I'll start and Tony can chime in. So first, on the 30% retention, I'm not sure what the 48% is that you're referring to. The 30% is month 12 retention. And if you actually compare that to DoorDash versus Wolt on an apples-to-apples basis, it's pretty comparable depending on some markets, it's even higher. So we feel good about the retention stats certainly compared to our data. And when you look at our data using third-party sources, we understand that our retention is better than everybody else in the category.

好的。Ross，也許我先回答，Tony 也可以補充。首先，關於 30% 留存率，我不確定你所指的 48% 是什麼。這個 30% 是第 12 個月的留存率（month 12 retention）。如果你把 DoorDash 與 Wolt 放在同樣口徑下做可比（apples-to-apples）比較，其實相當接近；在某些市場甚至更高。因此，就我們的數據來看，我們對留存率指標很有信心。而且當我們使用第三方來源來看數據時，我們理解到我們的留存率優於該類別中的所有其他業者。

In terms of the European competitors, there isn't an exact apples-to-apples source right that compares market to market. But based on what we're seeing in terms of competitors in the U.K. and other markets that are covered, the 30% actually compares favorably.

至於歐洲競爭對手，目前沒有一個完全可比（apples-to-apples）的來源能逐一市場對市場地比較。不過，根據我們在英國及其他涵蓋市場看到的競爭情況，30% 的表現其實相對有利。

On the question of DashPass, DashPass frequency as well as all the figures on our classic product, you really need to look at it on a cohort level. Yes, you've got a mix shift between as you get more DashPass subscribers, the blended order frequency grows. But even on a cohort level, underlying order frequency has continued to grow. And what's driving that is these new categories. We're seeing people adopt new categories, Tony referred to the 14% adoption rate at the end of Q4, which has continued to grow in Q1 and Q2, and that's driving order frequency growth at the cohort level.

關於 DashPass 的問題，DashPass 的頻率以及我們經典產品（classic product）的所有指標，你確實需要以同批次用戶（cohort）層級來看。是的，隨著 DashPass 訂閱者增加，產品組合（mix）會發生轉移，進而使加權後的整體下單頻率（blended order frequency）上升。但即便在 cohort 層級，底層的下單頻率也持續成長。推動因素是這些新類別。我們看到用戶開始採用新類別；Tony 提到在第四季末採用率為 14%，而且在第一季與第二季持續成長，這正在 cohort 層級推動下單頻率成長。

Your next question comes from the line of Jason Helfstein from Oppenheimer.

您的下一個問題來自 Oppenheimer 的 Jason Helfstein。

I just want to dig a bit more into DashPass. With the economy reopening and kind of return to work progressing in a certain direction and increased use of ridesharing. Is it making it customer acquisition of DashPass, any less efficient or more challenging?

我想再深入一點談 DashPass。隨著經濟重啟、復工在某種方向上推進，以及共乘（ridesharing）使用增加，這是否讓 DashPass 的客戶獲取變得更沒有效率或更具挑戰？

And then secondly, I think you said that you pointed out Dashers costs are down this year versus last year, but specifically and perhaps this month and last month, are you seeing the cost to acquire new Dashers come down even further?

第二點，我想你提到今年相較去年 Dasher 成本下降，但更具體地說，尤其是本月與上月，你們是否看到獲取新 Dasher 的成本進一步下降？

Jason, so first on the DashPass front, what I should say is would not observe any impact on our DashPass signups, as a result of people going back to the office or increased ridesharing or anything like that. In fact, our DashPass subs have continued to grow both on a year-on-year basis as well as on a quarter-on-quarter basis to a record high. So we feel good that DashPass is a key component of driving better affordability for our consumers and the growth has been consistent and reliable. So that's great.

Jason，首先就 DashPass 而言，我要說的是，我們並未觀察到因為人們回到辦公室、共乘使用增加或任何類似因素，而對 DashPass 註冊造成影響。事實上，我們的 DashPass 訂閱數不僅年增，也季增，並創下歷史新高。因此我們很有信心，DashPass 是提升消費者可負擔性（affordability）的關鍵組成，而其成長一直穩定且可靠。這非常好。

In terms of Dasher costs, the reduction in Dasher was really driven by 2 things. There's the Dasher cost per order, which I mentioned earlier was really a normalization from the elevated levels last year because of increased fiscal stimulus. There also was, if you recall, last year, increased advertising costs because everyone is competing for Dashers. And that environment has gotten a little bit better, which has driven lower Dasher cost in Q2, and those dasher cost continue to remain where they were in the last couple of months of the quarter.

至於 Dasher 成本，Dasher 成本下降主要由兩個因素驅動。一是每單 Dasher 成本（Dasher cost per order），我先前提到，這主要是從去年因財政刺激增加而偏高的水準回歸常態。二是如果你還記得，去年因為大家都在競爭 Dasher，廣告成本上升。而這樣的環境已稍有改善，帶動第二季 Dasher 成本下降，而且這些 Dasher 成本在該季最後幾個月也持續維持在相近水準。

Your next question comes from the line of Douglas Anmuth from JPMorgan.

您的下一個問題來自摩根大通（JPMorgan）的 Douglas Anmuth。

I just wanted to circle back on profitability and just the EBITDA guide for the full year. Should we think about the increase at the low end there. Should we think about that as the same dynamics that you suggested in 2Q in terms of driven mostly by inflation? And is that all coming on the Dash side of the business? Or is -- are we seeing any improvement perhaps in your outlook there for Wolt?

我想回到獲利能力，以及全年 EBITDA 指引。對於低端上調的部分，我們是否應該把它理解為你們在第二季所提到的相同動態，也就是主要由通膨驅動？而這是否全部來自 Dash 端的業務？或者——你們對 Wolt 的展望是否也有任何改善？

Doug, can you clarify, are you talking about Q3 or Q4?

Doug，你能澄清一下嗎？你是在問第三季還是第四季？

The full year outlook for EBITDA.

全年 EBITDA 的展望。

Yes. The full year outlook of EBITDA, I mean, we're bringing it up because we've made progress here in the business, both in terms of overall scale has increased relative to our expectations at the beginning of the year. If you remember our guidance at the beginning of the year was -- I think it was [$49 billion or $50 billion] at the top end. So that's gone up. Second, we are seeing positive benefits in terms of sub totals, which we talked about earlier, which results in more revenue and EBITDA. Third, Dasher costs are trending slightly better, and that's been driven by 2 things as product changes that have helped drive increased retention of our Dashers as well as some of the macro factors such as rising cost of living and declining consumer savings that are helping increase the retention of our existing fleet. So it's these various factors that have given us a little more confidence on EBITDA, as a result of which, we have increased the low end of our guidance. And it's really a commitment to say that we're going to stay above breakeven.

好的。關於全年 EBITDA 展望，我們之所以上調，是因為我們在業務上取得了進展：首先，整體規模相較年初預期有所提升。如果你記得我們年初的指引——我想上限大約是 490 億或 500 億美元——現在已經上調。第二，我們在小計（subtotals）方面看到正向效益，我們先前也談過，這帶來更多營收與 EBITDA。第三，Dasher 成本走勢略優，這由兩個因素驅動：一是產品變更有助於提升 Dasher 留存；二是一些總體因素，例如生活成本上升與消費者儲蓄下降，這些都有助於提高我們既有車隊（fleet）的留存。正是這些不同因素讓我們對 EBITDA 更有信心，因此我們上調了指引的低端。這也代表我們承諾將維持在損益兩平以上。

Your next question comes from the line of Nikhil Devnani.

您的下一個問題來自 Nikhil Devnani。

Just on the Q3 guide, if we take Wolt out, it looks like GOV is set to kind of step down sequentially for core DoorDash. Just wondering if you get there with the assumption that the July trends kind of hold roughly flattish through the quarter? Or is there some further kind of softness embedded in that outlook?

關於第三季指引，如果把 Wolt 拿掉，看起來核心 DoorDash 的 GOV 似乎會較前一季下滑。我想了解，你們在達到該指引時，是假設 7 月的趨勢在整個季度大致持平延續？還是展望中其實已經納入了進一步走弱？

And then as a second question, just given all the macro concerns, have you seen any indications that consumers are trading down here in the types of restaurants that they order from? And any kind of color you can provide on the demographics of the customer base would be helpful.

第二個問題是，考量各種總體疑慮，你們是否看到消費者在下單餐廳類型上出現「降級消費」（trading down）的跡象？另外，若能提供一些客群人口統計（demographics）的資訊也會很有幫助。

Sure. Maybe I'll take a crack at that. So the first thing I'd say is our consumer metrics remain healthy. So if you look at what happened in Q2, our MAUs grew by double digits year-on-year, order frequency grew, our pace of consumer acquisition continued to remain healthy. And these signals provide a solid foundation for growth in the long term.

好的。我來試著回答。首先我要說的是，我們的消費者指標仍然健康。如果看第二季的表現，我們的 MAU 年增達雙位數、下單頻率成長、消費者獲取的速度也持續維持健康。這些訊號為長期成長提供了穩固基礎。

With respect to the second half specifically, we expect normal course seasonality this year. So in general, we've experienced growth in Q4 and Q1. Sequential growth in Q2 and Q3 is usually muted as a result of summer seasonality. In fact, if you look at our Q2 to Q3 growth last year, you'll see that our GOV dipped a little bit, and we're baking that in into the guide.

就下半年而言，我們預期今年會呈現正常的季節性。一般來說，我們在第四季與第一季會經歷成長；而第二季與第三季的季增通常會因夏季季節性而較為溫和。事實上，如果你看去年第二季到第三季的成長，你會看到我們的 GOV 略有下滑，而我們已在指引中納入這點。

The second point I'll make is the macro environment continues to remain uncertain. And so we haven't seen any impact from weakening consumer spend, at least so far in Q2 but we're accounting for that uncertainty because it could have an impact on normal course seasonality in the bounce back that you see in Q4. And so that's what we're baking into our guidance. And then can you repeat the question on macro?

我要提出的第二點是，總體宏觀環境仍然充滿不確定性。因此，至少到目前為止在第二季我們尚未看到消費支出走弱帶來的任何影響，但我們已將這種不確定性納入考量，因為它可能會影響你在第四季看到的反彈所對應的正常季節性。因此，這就是我們在指引中所納入的假設。另外，你可以重複一下關於宏觀的問題嗎？

Yes. Just any indication that consumers might be trading down in the types of restaurants they order from? And just a reminder what your demographics look like for the customer base.

是的。只是想問，有沒有任何跡象顯示消費者可能在下調他們下單的餐廳類型？另外也請提醒一下，你們客戶基礎的人口統計概況是什麼？

Yes. What I'll say on that point is at least based on the data we've seen so far, we haven't seen anything that would point to a particular income peer or people shifting from certain types of restaurants to another. In fact, I think we said in the letter, the sub totals increased because of higher item prices. But that was offset by fewer items per order as consumers responded to inflation. So it's less that they're shifting away and adjusting their buying behavior from other restaurants. They just -- they're spending the same amount, just buying a few things. And that just goes to show you how resilient the category is. Convenience and delivery and dining in has become a stable part of people's lives and you're seeing that in the data.

是的。就這一點而言，我想說的是，至少根據我們目前看到的數據，我們沒有看到任何跡象指向特定的收入族群，或是人們從某些類型的餐廳轉向另一種類型。事實上，我想我們在信中提到，小計金額上升是因為單品價格更高。但隨著消費者對通膨做出反應，每筆訂單的品項數變少，抵消了這個影響。所以與其說他們在轉移、從其他餐廳調整購買行為，不如說他們——花的金額差不多，只是買的東西少了一些。這也顯示這個品類有多具韌性。便利性、外送以及外出用餐已成為人們生活中穩定的一部分，你也能從數據中看到。

Your next question comes from the line of Michael McGovern from Bank of America.

下一個問題來自美國銀行（Bank of America）的 Michael McGovern。

I wanted to dig a little bit into the gas rewards program. just curious, I noticed you listed gas the program as one of the contributing factors to Dasher cost per order actually being down. So I was wondering, is there a potential impact on Dasher supply or Dasher cost per order when the gas rewards program ends?

我想再深入問一下加油回饋計畫。我只是好奇，我注意到你們把加油計畫列為每單 Dasher 成本實際下降的因素之一。所以我想問，當加油回饋計畫結束時，是否可能對 Dasher 供給或每單 Dasher 成本造成影響？

And then secondly, is there a chance that, that could be extended beyond August 31, if there is a potential supply impact from that? And are the savings from that ending in August currently being baked into guidance for Q3 and for the full year?

第二個問題是，如果它可能對供給造成影響，這個計畫有沒有可能延長到 8 月 31 日之後？另外，目前你們對第三季以及全年指引，是否已把 8 月結束後所帶來的節省納入考量？

Michael, it's Prabir. Maybe I'll take a your question. Yes, we spent over $40 million on the gas savings program, and we see very positive feedback from Dashers. You are the commentary I think you're referring to is it's not a Dasher cost per order reduced, but what ended up happening is the retention of our existing fleet improved. And that the improvement in that retention of the existing fleet was driven by 3 things, is the product changes that will need in order to help attention, it's the gas rewards program that frankly displays the increase in fuel costs for Dashers and third, the macro factors that we talked about around declining consumer savings and rising inflation. And so it's the combination of these 3 things that we believe improved the retention of our existing fleet that then resulted in us needing fewer new Dashers, and therefore, acquiring a greater portion of our Dashers organically than we otherwise would. That then translate into lower [tax] costs. In terms of renewing the program, it's too early to say right now. We have assume that we will -- we continue to operate the current dasher program to the end of August, and we'll make a call depending on where gas prices are high as well as other things. At the end of the day, it might be a cheaper way to obtain Dashers, what we'll see as we need progress here in the quarter.

Michael，我是 Prabir。我來回答你的問題。是的，我們在加油節省計畫上花了超過 4,000 萬美元，而且我們看到 Dashers 的回饋非常正面。你提到的那段評論，我想你指的是：並不是每單 Dasher 成本下降，而是我們既有車隊的留任率提升了。而既有車隊留任率的提升是由三件事驅動的：第一，是我們為了提升留任而做的產品變更；第二，是加油回饋計畫，坦白說它抵消了 Dashers 的燃油成本上升；第三，是我們談到的宏觀因素，包括消費者儲蓄下降與通膨上升。因此，我們認為正是這三者的組合提升了既有車隊的留任率，進而使我們需要招募的新 Dashers 變少，因此相較於原本情況，我們能以更高比例透過自然方式獲得 Dashers。這就轉化為較低的 [tax] 成本。至於是否續辦該計畫，現在說還太早。我們假設會——我們會把目前的 Dasher 計畫運作到 8 月底，之後會視油價是否仍然偏高以及其他因素再做決定。歸根結底，這可能是一種更便宜的方式來取得 Dashers；我們會在本季進展中再觀察。

Your next question comes from the line of Mark Mahaney from Evercore ISI.

下一個問題來自 Evercore ISI 的 Mark Mahaney。

I wanted to ask about whether you're seeing any pressure from competitors in nonurban markets, either on the restaurant supply, acquisition side or consumer incentives? And since the Amazon-Grubhub deal was sort of announced, does that show up in any of the metrics that you track?

我想問的是，你們是否在非都會市場看到競爭對手帶來的任何壓力——不論是在餐廳供給、拓展/招募端，或是消費者補貼/誘因方面？另外，自從 Amazon-Grubhub 的交易大致宣布之後，這是否反映在你們追蹤的任何指標上？

I can that lead off and then feel free to chime in here, Prabir. So far, we haven't seen impact whether it's recent competitor announcements or moves to invest in certain types of geographies versus other types of geographies in the numbers, but obviously, we're taking stock of what's happening. But I think it's important to remember, regardless of what's happening in the external environment, that our focus is making sure that we continue building the best products. I mean, at the end of the day, the customer, whether it's for our platform or someone else's, they're going to judge all of us on the combination of the selection of places we deliver from the quality of that delivery experience in terms of speed, timeliness and accuracy, the affordability of the platform and the customer service level.

我可以先回答，Prabir 也歡迎補充。到目前為止，不論是近期競爭對手的公告，或是他們投資某些地區而非其他地區的動作，我們在數據上都還沒有看到影響，但顯然我們正在密切關注正在發生的事情。不過我認為重要的是要記住，無論外部環境如何變化，我們的重點都是確保持續打造最好的產品。我的意思是，歸根結底，客戶不管是在我們的平台或別人的平台，他們會以幾個面向的組合來評價我們所有人：我們可配送的商家選擇、配送體驗的品質（速度、準時性與準確性）、平台的可負擔性，以及客戶服務水準。

And it's that combination, I think, that so far has been evidenced or performance with leading retention and order frequency that has separated us from the pack. And we have to just keep making sure that we stay ahead on that dimension. At the end of the day, it's that combination that ultimately is going to judge us or anyone else, and even if you offer the product for free, if that combination isn't where it needs to be for the customer, I'm not sure it's going to matter.

而我認為，正是這樣的組合，迄今已透過我們在留存率與下單頻率方面的領先表現得到證明，讓我們從同業中脫穎而出。我們必須持續確保在這個維度上保持領先。歸根結底，最終評判我們或任何其他人的，就是這樣的組合；即使你免費提供產品，如果這個組合未達到客戶所需要的水準，我不確定那是否還會有影響。

Thank you. This concludes today's conference call. I would like to thank our speakers, and thank you all for joining us today. This now concludes the call. You may now disconnect.

謝謝。今天的電話會議到此結束。我要感謝我們的講者，也感謝各位今天的參與。本次會議至此結束。您現在可以掛線。