數據狗 (DDOG) 2023 Q3 法說會逐字稿

Good day, and thank you for standing by. Welcome to the Third Quarter 2023 Datadog Earnings Conference Call. (Operator Instructions) Please be advised that today's conference is being recorded. I would now like to hand the conference over to your speaker today, Yuka Broderick, Vice President of Investor Relations. Please go ahead.

大家好，感謝您的耐心等待。歡迎參加Datadog 2023年第三季財報電話會議。 （操作員指示）請注意，今天的會議正在錄製中。現在，我想將會議交給今天的發言人，投資者關係副總裁Yuka Broderick。請開始。

Thank you, [Gigi]. Good morning, and thank you for joining us to review Datadog's third quarter 2023 financial results, which we announced in our press release issued this morning. Joining me on the call today are Olivier Pomel, Datadog's Co-Founder and CEO; and David Obstler, Datadog's CFO.

謝謝你，[Gigi]。早安，感謝您加入我們，回顧Datadog 2023年第三季的財務業績，我們在今天上午發布的新聞稿中公佈了這些業績。今天和我一起參加電話會議的還有Datadog聯合創始人兼執行長Olivier Pomel和財務長David Obstler。

During this call, we will make forward-looking statements, including statements related to our future financial performance, our outlook for the fourth quarter and the fiscal year 2023 and related notes and assumptions, our gross margins and operating margins, our product capabilities, our ability to capitalize on market opportunities and usage optimization trends. The words anticipate, believe, continue, estimate, expect, intend, will and similar expressions are intended to identify forward-looking statements or similar indications of future expectations. These statements reflect our views only as of today and are subject to a variety of risks and uncertainties that could cause actual results to differ materially. For a discussion of the material risks and other important factors that could affect our actual results, please refer to our Form 10-Q for the quarter ended June 30, 2023. Additional information will be made available in our upcoming Form 10-Q for the fiscal quarter ended September 30, 2023, and other filings with the SEC. This information is also available on the Investor Relations section of our website, along with a replay of this call.

在本次電話會議中，我們將做出前瞻性陳述，包括與我們未來財務業績相關的陳述、我們對第四季度和2023財年的展望及相關說明和假設、我們的毛利率和營業利潤率、我們的產品能力、我們利用市場機遇的能力以及使用優化趨勢相關的陳述。預期、相信、繼續、估計、期望、打算、將要等字詞及類似表述旨在識別前瞻性陳述或對未來預期的類似表述。這些陳述僅反映我們截至今日的觀點，並受各種風險和不確定性的影響，這些風險和不確定性可能導致實際結果與預期有重大差異。有關可能影響我們實際業績的重大風險和其他重要因素的討論，請參閱我們截至2023年6月30日季度的10-Q表。更多資訊將在我們即將發布的截至2023年9月30日財季的10-Q表以及向美國證券交易委員會提交的其他文件中公佈。這些資訊也可在我們網站的「投資者關係」部分查閱，同時可觀看本次電話會議的重播。

We will also discuss non-GAAP financial measures, which are reconciled to their most directly comparable GAAP financial measures in the tables in our earnings release, which is available at investors.datadoghq.com. With that, I'd like to turn the call over to Olivier.

我們還將討論非公認會計準則 (Non-GAAP) 財務指標，這些指標已與我們業績報告中表格中最直接可比較的公認會計準則 (GAAP) 財務指標進行調節，該報告可在 investors.datadoghq.com 上取得。現在，我想將電話交給 Olivier。

Thanks, Yuka, and thank you all for joining us this morning. We are pleased with our execution in Q3. We delivered another quarter of profitable growth and robust notable bookings and we continue to broaden our platform to have customers become and grow digital businesses.

謝謝 Yuka，也謝謝大家今天早上的到來。我們對第三季的業績表現非常滿意。我們再次實現了獲利成長，訂單量也表現強勁，我們將繼續拓展平台，幫助客戶發展數位業務。

Let me start with a review of our Q3 financial performance. Revenue was $548 million, an increase of 25% year-over-year and above the high end of our guidance range. We ended with about 26,800 customers, up from about 22,200 last year. We ended the quarter with about 3,130 customers with an ARR of $100,000 or more, up from about 2,600 last year. And these customers generated about 86% of our ARR. And we generated free cash flow of $138 million with a free cash flow margin of 25%.

首先，我來回顧一下我們第三季的財務表現。營收為5.48億美元，年增25%，高於我們預期區間的上限。我們最終擁有約26,800名客戶，高於去年的約22,200名。本季末，我們約有3,130名年經常性收入（ARR）達到或超過10萬美元的客戶，高於去年的約2,600名。這些客戶創造了我們約86%的年經常性收入（ARR）。我們創造了1.38億美元的自由現金流，自由現金流利潤率為25%。

Turning to platform adoption. Our platform strategy continues to resonate in the market. As of the end of Q3, 82% of customers were using two or more products, up from 80% a year ago. 46% of customers were using 4 or more products, up from 40% a year ago and 21% of our customers were using 6 or more products, up from 16% last year.

談到平台採用。我們的平台策略持續在市場上引起共鳴。截至第三季末，82% 的客戶使用兩款或兩款以上的產品，高於去年的 80%。 46% 的客戶使用四款或四款以上的產品，高於去年的 40%。 21% 的客戶使用六款或六款以上的產品，高於去年的 16%。

Now let's discuss this quarter's business drivers. In Q3, we saw usage growth we've seen customers improve compared to Q2. Overall growth in Q3 was relatively consistent throughout the quarter and comparable to levels we've seen in Q1. We are seeing signs that the cloud optimization activity from some of our customers may be moderating. As a reminder, last quarter, we discussed a cohort of customers who began optimizing about a year ago and we said that they appear to stabilize their users growth at the end of Q2. That trend has held for the past several months with that cohorts usage remaining stable throughout Q3.

現在讓我們來討論一下本季的業務驅動因素。在第三季度，我們看到使用量成長，客戶數量也比第二季度有所改善。第三季的整體成長在整個季度相對穩定，與第一季的水平相當。我們看到一些客戶的雲端優化活動可能正在放緩。需要提醒的是，上個季度，我們討論了一群大約一年前開始進行最佳化的客戶，我們表示，他們的用戶成長似乎在第二季末趨於穩定。這一趨勢在過去幾個月一直保持，這批客戶的使用量在整個第三季度保持穩定。

Overall, we continue to see impact from optimization in our business. But we believe that the intensity and breadth of optimization we've experienced in recent quarters is moderating. Meanwhile, our new logo activity has remained robust. New logo bookings continue to scale and grow year-over-year. And for the second quarter in a row, we closed a record number of new deals with more than $100,000 in annual commitment.

整體而言，我們業務優化的成效仍在持續顯現。但我們認為，近幾季優化的強度和廣度正在放緩。同時，我們的新標誌推廣活動依然強勁。新標誌預訂量持續擴大，且較去年同期成長。我們連續第二季達成創紀錄的新交易，年度承諾金額超過10萬美元。

With our Land and Expand model, we expect new logos to turn into much larger customers over time as they lead into the cloud and add up more and more products. Finally, regarding customer growth, we are pleased with the new logos, new workloads and new product attaches we added this quarter. We added a number of exciting new customized in Q3, and I'll discuss a couple of examples later.

透過我們的「落地並擴展」模式，我們預計隨著新客戶逐漸轉向雲端並推出更多產品，他們將逐漸發展成為更大的客戶。最後，關於客戶成長，我們對本季新增的新客戶、新工作負載和新產品附件感到滿意。我們在第三季新增了許多令人興奮的客製化服務，稍後我將討論幾個例子。

Note that our total customer count is largely driven by our long tail of very small customers, while our sales motions are more targeted to the middle and high end of our prospects. And as a reflection of our team's strong execution, our net adds of customers over $100,000, saw an increase in Q3 compared to Q2.

請注意，我們的總客戶數量主要由長尾小客戶驅動，而我們的銷售策略則更專注於中高端潛在客戶。此外，由於我們團隊強大的執行力，第三季淨增加超過10萬美元的客戶數量較第二季有所成長。

Despite a more cost-conscious demand environment over the past year, our business has continued to grow across product lines and we are very proud to achieve several key milestones. First, our infrastructure monitoring ARR exceeded $1 billion. Today, our infrastructure products cover monitoring the performance of hosts, networks, containers, Kubernetes deployments, saves functions and other aspects of infrastructure in the cloud as well as a full set of AI and machine learning tools to help our customers separate signal for noise.

儘管過去一年需求環境對成本更加敏感，但我們各產品線的業務仍在持續成長，並取得了多項重要里程碑，對此我們深感自豪。首先，我們的基礎設施監控年平均經常性收入 (ARR) 超過 10 億美元。如今，我們的基礎設施產品涵蓋主機、網路、容器、Kubernetes 部署、儲存功能以及雲端基礎架構其他方面的效能監控，並提供全套人工智慧和機器學習工具，幫助客戶有效識別幹擾因素。

Second, our APM suite, which includes core APM, Synthetics, Real User Monitoring and Continuous profiler, exceeded $500 million in ARR, and we continue to expand our capabilities in APM, most recently with single-step instrumentation, which allows a single engineer to enable APM across all applications without core changes and we ship advances in mobile app monitoring, including mobile application testing and mobilization replay.

其次，我們的 APM 套件（包括核心 APM、合成器、真實用戶監控和連續分析器）的 ARR 已超過 5 億美元，並且我們繼續擴展 APM 功能，最近推出了單步檢測功能，該功能允許單個工程師在所有應用程序中啟用 APM，而無需進行核心更改，並且我們在移動應用程序監控方面取得了進展，包括移動程序和測試進展。

Third, our log management product exceeded $500 million in AR. We also continue to expand our capabilities in load management. And with Flex logs, customers can easily scale storage and compute separately, allowing for new very high-volume logging use cases in a cost-effective manner.

第三，我們的日誌管理產品在應收帳款 (AR) 方面的收入超過 5 億美元。我們還在持續擴展負載管理功能。借助 Flex 日誌，客戶可以輕鬆地分別擴展儲存和運算能力，從而以經濟高效的方式支援新的高容量日誌記錄用例。

From the very beginning, my cofounder Alexis and I had a vision to create a unified platform that serves end-to-end use cases across datasets, products and SIM boundaries. We believe that these ARR milestones and their balance across the 3 pillars of observability demonstrate that Datadog is unique within the industry in establishing true platform value for customers. And of course, even though these products have become significant in size, we are only just getting started. We will continue to innovate to deliver more solutions for our customers across observability and beyond.

從一開始，我和我的共同創辦人 Alexis 就有一個願景，那就是創建一個統一的平台，服務於跨資料集、產品和 SIM 邊界的端到端用例。我們相信，這些 ARR 里程碑及其在可觀察性三大支柱上的平衡，證明了 Datadog 在為客戶建立真正的平台價值方面在業內獨樹一幟。當然，儘管這些產品已經具備了相當的規模，但我們才剛起步。我們將繼續創新，為客戶提供更多涵蓋可觀察性及其他領域的解決方案。

I will add that we have empathy for our customers in their pain points, in part because we are ourselves users of cloud and next-gen technologies at a meaningful scale and we extensively deploy a user on solution, which is appropriately known as dog footing. As an example, we have extensively relied on our cloud cost management product as we expanded its capabilities this past year.

我想補充一點，我們對客戶的痛點感同身受，部分原因在於我們自己也是雲端運算和下一代技術的大規模用戶，並且我們廣泛部署了「用戶至上」的解決方案，也就是我們常說的「狗腳」。例如，在過去一年中，隨著雲端成本管理產品功能的擴展，我們對其的依賴程度也越來越高。

And the use of our product has played a large role in delivering cost performance and efficiency improvements, optimizing our own cloud usage and ultimately resulting in expansion of our gross margins in recent quarters.

我們的產品的使用在提高成本性能和效率、優化我們自己的雲端使用方面發揮了重要作用，並最終導致我們最近幾季的毛利率擴大。

We also continued to innovate in the depth take-up space. Our recent expansions in cloud security include cloud SIEM investigator, where customers can visualize logs over a long period of time to conduct security investigations. And within our cloud security management product, we have introduced Cloud Infrastructure Entitlement Management, or CIEM, to help customers prevent identity and access management security issues.

我們也在深度應用領域持續創新。我們最近在雲端安全領域的擴展包括雲端SIEM調查器，客戶可以透過它來視覺化長期日誌，從而進行安全調查。此外，我們在雲端安全管理產品中引入了雲端基礎設施授權管理（CIEM），以協助客戶預防身分和存取管理安全問題。

For a few years now, the industry has been talking about the idea of DevSecOps, the breaking down of silos among development, operations and security teams. And we entered the security space on the premise that DevOps and security teams should share the same data in the same platform.

幾年來，業界一直在討論 DevSecOps 的概念，即打破開發、營運和安全團隊之間的各自為政。我們進入安全領域的前提是，DevOps 和安全團隊應該在同一平台上共享相同的資料。

So starting this month, we are making the practice of DevSecOps easy to adopt for all customers by bringing together all the components needed to fully monitor and secure their entire stack with 2 simple packages. First, with infrastructure DevSecOps, our customers can observe and secure their entire cloud environment in one package. With a simple purpose price and a single agent deployed, customers get end-to-end visibility into performance, availability and security issues in one place. And from that one place, teams can also quickly remediate problems using built-in workflows and without any code or contribution changes.

因此，從本月開始，我們將透過兩個簡單的軟體包整合所有必要的元件，讓所有客戶都能輕鬆採用 DevSecOps 實踐，從而全面監控和保護他們的整個堆疊。首先，借助基礎設施 DevSecOps，我們的客戶可以在一個軟體包中監控和保護他們的整個雲端環境。只需支付簡單的費用並部署一個代理，客戶就能在一個地方獲得端到端的效能、可用性和安全問題可見性。而且，團隊還可以從這個地方使用內建工作流程快速修復問題，而無需更改任何程式碼或貢獻內容。

Second, with APM DevSecOps, we take this one step further. Customers can instrument cloud applications for both performance and generality issues in one single package enabled with the same unified agent used for infrastructure DevSecOps. APM DevSecOps complements infrastructure DevSecOps by surfacing open source and code level security viabilities alongside performance issue.

其次，借助 APM DevSecOps，我們更進一步。客戶可以在一個軟體包中，使用與基礎設施 DevSecOps 相同的統一代理，對雲端應用程式的效能和通用性問題進行檢測。 APM DevSecOps 透過在效能問題的同時，提供開源和程式碼級安全可行性，是對基礎設施 DevSecOps 的補充。

Finally, we continue to be excited about the opportunity in generative AI and Large Language Models. First, we believe adopting NextGen AI will require the use of cloud and other modern technologies and drive additional growth in cloud workloads. So we are continuing to invest by integrating with more components at every layer of the new AI stack and by developing our own LLM observability products.

最後，我們持續對生成式人工智慧和大型語言模型領域的機會感到興奮。首先，我們認為採用下一代人工智慧需要使用雲端和其他現代技術，並推動雲端工作負載的進一步成長。因此，我們將繼續投入，在新的人工智慧堆疊的每一層中整合更多組件，並開發我們自己的 LLM 可觀測性產品。

And while we see signs of AI adoption across large parts of our customer base, in the near term, we continue to see AI-related usage manifest itself most accurately with next-gen AI native customers who contributed about 2.5% of our ARR this quarter.

雖然我們看到大部分客戶群都採用了人工智慧，但在短期內，我們仍然會看到與人工智慧相關的使用最準確地體現在下一代人工智慧原生客戶身上，他們本季度貢獻了我們約 2.5% 的 ARR。

In the mid- to long term, we expect customers of all industries and sizes to keep adding value to their products using AI and to get their early explore -- to get from early exploration to development into production, thus driving larger cloud and observability usage across our customer base.

從中長期來看，我們期望所有行業和規模的客戶都能繼續使用人工智慧為其產品增加價值，並進行早期探索——從早期探索到開發再到生產，從而推動整個客戶群中更大規模的雲端和可觀察性的使用。

Besides observing the AI stack, we also expect to keep adding value to our own platform using AI. Datadog's unified platform and purely SaaS model, combined with strong multiproduct adoption by our customers generates a large amount of deep and precise observability data. We believe combining AI capabilities with this broad data set will allow us to deliver differentiated value to customers.

除了觀察 AI 堆疊之外，我們也希望利用 AI 持續提升我們自身平台的價值。 Datadog 的統一平台和純 SaaS 模式，加上客戶對多產品的廣泛採用，能夠產生大量深度精準的可觀察性資料。我們相信，將 AI 能力與這一廣泛的數據集相結合，將使我們能夠為客戶提供差異化的價值。

And we are working to product that is differentiated value through recently announced capabilities such as our bid Bits AI assistant, AI generated synthetic test and AI-led air analysis and resolution, and we expect to deliver many more related innovation to customers over time.

我們正在努力透過最近宣布的功能（例如我們的競標 Bits AI 助理、AI 生成的合成測試和 AI 主導的空氣分析和解決方案）來打造具有差異化價值的產品，並且我們期望隨著時間的推移為客戶提供更多相關創新。

Let's move on to sales and marketing, where we continue to execute on both new logos and existing customers. So let's discuss some of our wins. First, we founded 7-figure land over 5 years with a leading provider of dental care. This company's legacy monitoring just didn't cut and it contributed to delay with their migration to Azure. What's concerning to them was that customers notice poor application performance and were complaining publicly on social media. By adopting 6 Datadog products, they expect to find and fix the vast majority of incidents internally before their customers are affected. And in signing a 5-year deal, this customer showed its confidence in Datadog as a long-term partner in their migration.

接下來是銷售和行銷，我們將繼續在新客戶和現有客戶方面開展工作。讓我們來談談我們的一些成功案例。首先，我們與一家領先的牙科保健服務提供者合作，在五年內建立了七位數的資產。這家公司原有的監控系統效能不佳，導致其遷移到 Azure 的時間延遲。令他們擔憂的是，客戶注意到應用程式效能不佳，並在社群媒體上公開抱怨。透過採用 6 款 Datadog 產品，他們希望在客戶受到影響之前，在內部發現並修復絕大多數問題。在簽署為期五年的協議時，這位客戶表明了對 Datadog 作為其遷移過程中長期合作夥伴的信心。

Next, we signed a 7-figure land with a South American fintech company. By moving from basic built-in cloud monitoring, legacy tooling and open source tools to Datadog, this customer expects to significantly reduce costs by spending less on tooling, reducing time to resolution and giving time back to engineers to innovate on their own products.

接下來，我們與一家南美金融科技公司簽署了一項價值七位數的土地協議。透過從基本的內建雲端監控、傳統工具和開源工具遷移到 Datadog，該客戶希望透過減少工具支出、縮短解決問題的時間，並讓工程師有時間進行產品創新，從而顯著降低成本。

Next, we signed an 8-figure renewal over 3 years with a major American chain of convenience stores. With this expansion, Datadog will bring all aspects of these customers' tech systems into one platform, including their applications, hybrid cloud, network, in-store IoT technology, point-of-sale systems, self-serve kiosks, fuel pumps and corporate infrastructure.

接下來，我們與一家美國大型連鎖便利商店簽署了一份為期三年、金額高達八位數的續約合約。透過此次擴展，Datadog 將把這些客戶技術系統的各個方面整合到一個平台上，包括他們的應用程式、混合雲、網路、店內物聯網技術、銷售點系統、自助服務終端、加油站和企業基礎設施。

This will free up employee time to focus on customer service with expectations to save millions of dollars annually. This customer plans to use 6 Datadog products replacing 3 commercial observability tools. Next, we signed a 7-figure expansion with a major U.S. federal agency.

這將使員工有更多時間專注於客戶服務，預計每年可節省數百萬美元。該客戶計劃使用 6 套 Datadog 產品取代 3 套商用可觀測性工具。接下來，我們與美國大型聯邦機構簽署了一份價值七位數的擴展協議。

When we first started working with this customer a year ago, Datadog was approved for a limited subset of programs. But as we have demonstrated value and gain eternal adoption, this customer is now deploying Datadog across the entire agency. They have a added 6 Datadog products, and by doing so, consolidated out of 7 tools.

一年前，我們剛開始與這位客戶合作時，Datadog 只獲準用於有限的幾個專案。但隨著我們展現出的價值並獲得了廣泛的採用，這位客戶現在正在整個機構部署 Datadog。他們新增了 6 款 Datadog 產品，從而整合了 7 種工具。

Next, we signed a 7-figure expansion with a Fortune 500 industrial company. Customer was concerned with out-of-control costs with its legacy log management products and was using a dozen different tools. And when they began using Datadog, they noticed far fewer support tickets submitted to their reliability team. By growing set with Datadog and expanding to 7 products, this customer expects to deliver better service while saving time and reducing cost.

接下來，我們與一家財富500強工業公司簽署了一份價值七位數的擴展合約。客戶擔心其原有日誌管理產品的成本失控，並且使用了十幾種不同的工具。當他們開始使用Datadog時，他們注意到提交給可靠性團隊的支援工單數量大大減少。透過與Datadog合作，並將產品擴展到7種，該客戶希望在節省時間和降低成本的同時，提供更優質的服務。

And last, we signed a 7-figure expansion with a software business that is part of a tech hyperscaler. This long-term customer has used Datadog for infrastruction metrics and will now expand to adopt 7 Datadog products. Datadog will be replacing its commercial APM tool, which was well adopted by its engineers and led to inefficient troubleshooting outages and revenue impact.

最後，我們與一家隸屬於科技超大規模企業的軟體公司簽署了一份價值七位數的擴展協議。這家長期客戶一直使用 Datadog 來監測基礎設施指標，現在將擴展至採用 7 款 Datadog 產品。 Datadog 將取代其商用 APM 工具，該工具先前已被其工程師廣泛採用，並導致故障排除效率低下、中斷服務並影響了收入。

Our support of open telemetry, in particular, was key to their decision to expand with Datadog, that it makes it possible for APM tracing to be democratized and use across their entire DevOps team. And that's it for this quarter's highlights. I'd like to thank our go-to-market teams for their strong execution in Q3.

我們對開放遙測的支援尤其重要，這促使他們決定與 Datadog 合作，這使得 APM 追蹤能夠普及並在整個 DevOps 團隊中使用。本季的亮點就到這裡。我要感謝我們的市場推廣團隊在第三季的出色表現。

Before I turn it over to David for a financial review, let me reiterate our longer-term outlook. As we have said throughout this period of cloud optimization and macro uncertainty, our long-term plans have remained unchanged. We continue to believe digital transformation and cloud migration are long-term secular growth drivers of our business and critical motions for every company to deliver value and competitive advantage.

在交給David進行財務回顧之前，請容我重申一下我們的長期展望。正如我們在雲端優化和宏觀不確定性期間所強調的那樣，我們的長期計劃保持不變。我們始終堅信，數位轉型和雲端遷移是我們業務的長期成長動力，也是每家公司創造價值和競爭優勢的關鍵動力。

So we continue to invest aggressively to better our platform, and we aim to be our customers' mission-critical partners as they move to cloud and to modern DevSecOps With that, I will turn it over to our CFO, David.

因此，我們將繼續積極投資以改善我們的平台，並致力於成為客戶在轉向雲端和現代 DevSecOps 時的關鍵任務合作夥伴。接下來，我將把權力移交給我們的財務長 David。

Thanks, Olivier. Q3 revenue was $548 million, up 25% year-over-year and up 7% quarter-over-quarter. To dive into some of the drivers of this Q3 performance, first, regarding usage growth. We saw an improvement in usage growth in Q3 versus Q2. The Q3 usage growth was more similar to Q1 and relatively steady throughout the quarter.

謝謝，Olivier。第三季營收為5.48億美元，年增25%，季增7%。為了深入探討第三季業績表現的一些驅動因素，首先要說的是使用量成長。我們發現第三季的使用量成長與第二季相比有所改善。第三季的使用量成長與第一季較為相似，並且在整個季度都相對穩定。

We had a very healthy start to Q4 in October. While it is too early in the quarter to know for sure, it will happen in the next couple of months, the trends we see in early Q4 are stronger than they've been for the past year. Regarding usage growth by customer size, we continue to see larger spending customer growth at a slower rate than smaller spending customers, but usage growth improved for all customer sizes in Q3 relative to Q2.

10月份，我們的第四季開局非常健康。雖然現在還無法確定季度的走向，但未來幾個月內就會有所好轉，但我們在第四季初看到的趨勢比去年同期更加強勁。就按客戶規模劃分的使用量增長而言，我們仍然看到，消費金額較大的客戶的增長速度低於消費金額較小的客戶，但第三季度所有規模客戶的使用量增長均較第二季度有所改善。

And as Olivier discussed, we believe there are signs that the optimization activity we've been seeing is moderating. Last quarter, we discussed the cohort of customers who started optimizing about a year ago. This cohort's usage has been stable -- was stable throughout Q3. As we look at our overall customer activity, we continue to see customers optimizing but with less impact than we experienced in Q2, contributing to our usage growth with existing customers improving in Q3 relative to Q2.

正如Olivier所討論的，我們認為有跡象表明，我們之前看到的優化活動正在放緩。上個季度，我們討論了大約一年前開始優化的客戶群。這群客戶的使用情況一直很穩定——整個第三季都很穩定。縱觀我們的整體客戶活動，我們繼續看到客戶在進行優化，但影響不如第二季度那麼大，這促進了我們的使用量增長，現有客戶在第三季度的使用量相對於第二季度有所改善。

While we expect cost management to continue, we believe we are seeing moderation that is still present but -- a moderation is still present, but is less intense and less widespread than we experienced in recent quarters. Geographically, we experienced similar sequential revenue growth in North America and in our international markets. And finally, as regard to retention metrics, our trailing 12-month net revenue retention was in line with our expectations and came in slightly below 120% in Q3. Our trailing 12-month gross revenue retention continues to be stable in the mid- to high 90s, a sign of the mission-critical nature of our platform for our customers.

雖然我們預計成本管理將持續下去，但我們認為目前仍存在放緩的跡象——這種放緩的勢頭和範圍不如最近幾季那麼強烈和廣泛。從地理來看，我們在北美和國際市場的收入環比成長類似。最後，關於留存率指標，我們過去12個月的淨收入留存率符合預期，第三季略低於120%。我們過去12個月的總收入留存率持續穩定在95%左右，這顯示我們的平台對客戶至關重要。

Moving on to our financial results. Billings were $607 million, up 30% year-over-year. Billings duration increased slightly year-over-year. Remaining performance obligations, or RPO, was $1.45 billion, up 54% year-over-year. Current RPO growth was about 30% year-over-year.

接下來是我們的財務表現。帳單金額為6.07億美元，年增30%。帳單期限較去年同期略有增加。剩餘履約義務（RPO）為14.5億美元，年增54%。目前RPO年增約30%。

Over the past couple of quarters, we have seen an increasing preference from our customers to sign multiyear deals, and our weighted average booking duration was up sequentially and year-over-year. We see continued interest in multiyear duration deals in our pipeline as customers seek longer-term strategic partnerships with us.

過去幾個季度，我們發現客戶越來越傾向於簽訂多年期合同，我們的加權平均預訂時長也環比和同比均有所增長。隨著客戶尋求與我們建立更長期的策略合作夥伴關係，我們持續關注著我們未來多年合約的簽訂。

We continue to believe that revenue is a better indicator of our business trends than billings and RPO as those can fluctuate relative to revenues based on the timing of invoice and the duration of customer contracts.

我們仍然相信，收入比帳單和 RPO 更能反映我們的業務趨勢，因為這些指標會根據發票時間和客戶合約期限相對於收入而波動。

Now let's review some of the key income statement results. Unless otherwise noted, all metrics are non-GAAP. We have provided a reconciliation of GAAP to non-GAAP financials in our earnings release. Gross profit in the quarter was $451 million, representing a gross margin of 82.3%. This compares to a gross margin of 81.3% last quarter and 79.7% in the year ago quarter.

現在，讓我們回顧一下一些關鍵的損益表結果。除非另有說明，所有指標均為非公認會計準則 (GAAP)。我們在財報中提供了 GAAP 與非 GAAP 財務資料的對帳。本季毛利為 4.51 億美元，毛利率為 82.3%。相較之下，上季毛利率為 81.3%，去年同期為 79.7%。

As Olivier mentioned, we continued to experience efficiencies in cloud costs reflected in our cost of goods sold in the quarter as our engineering teams pursue cost savings and efficiency projects. Our Q3 OpEx grew 17% year-over-year. This is a decline from 26% year-over-year growth last quarter. We continued to execute on controlling costs given the uncertain environment. And Q3 operating income was $131 million for a 24% operating margin, up from 21% last quarter and above the 17% in the year ago quarter.

正如Olivier所提到的，由於我們的工程團隊致力於成本節約和效率提升項目，我們在本季的銷售成本中體現了雲端成本的持續提升。第三季的營運支出較去年同期成長17%，低於上一季26%的年增。鑑於不確定的環境，我們繼續控製成本。第三季營運利潤為1.31億美元，營運利潤率為24%，高於上一季的21%，也高於去年同期的17%。

Our margins were higher than we expected in Q3 as our organic growth was higher than in Q2, while our internal optimization and cost management efforts were successful. Turning to the balance sheet and cash flow statements. We ended the quarter with $2.3 billion in cash, cash equivalents and marketable securities and cash flow from operations was $153 million in the quarter.

由於有機成長高於第二季度，我們的利潤率在第三季度高於預期，同時我們的內部優化和成本管理工作也取得了成功。談到資產負債表和現金流量表。本季末，我們的現金、現金等價物及有價證券總額為23億美元，經營活動產生的現金流為1.53億美元。

After taking into consideration capital expenditures and capitalized software, free cash flow was $138 million for a free cash flow margin of 25%. And now for our outlook for the fourth quarter and for the full fiscal year 2023. A reminder, our guidance philosophy remains unchanged. We based our guidance on trends observed in recent months and apply conservatism on these growth trends.

計入資本支出和資本化軟體後，自由現金流為1.38億美元，自由現金流利潤率為25%。現在來看看我們對第四季和2023財年全年的展望。需要提醒的是，我們的指引理念保持不變。我們的指引是基於近幾個月觀察到的趨勢，並對這些成長趨勢採取了保守的立場。

For the fourth quarter, we expect revenues to be in the range of $564 million to $568 million, which represents about 20% to 21% growth year-over-year. Non-GAAP operating income is expected to be in the range of $129 million to $133 million and non-GAAP net income per share is expected to be $0.42 to $0.44 per share based on approximately 355 million weighted average diluted shares outstanding.

我們預計第四季營收將在5.64億美元至5.68億美元之間，年增約20%至21%。非公認會計準則下（Non-GAAP）營業利潤預計在1.29億美元至1.33億美元之間；基於約3.55億股加權平均稀釋流通股，非公認會計準則下每股淨利潤預計在0.42美元至0.44美元之間。

For fiscal year 2023, we expect revenues to be in the range of $2.103 billion to $2.107 billion, which represents 26% year-over-year growth. Non-GAAP operating income is expected to be in the range of $453 million to $457 million and non-GAAP net income per share is expected to be in the range of $1.52 to $1.54 per share based on approximately 351 million weighted average diluted shares outstanding.

我們預計2023財年營收將在21.03億美元至21.07億美元之間，年增26%。非公認會計準則下（Non-GAAP）營業利潤預計在4.53億美元至4.57億美元之間；基於約3.51億股加權平均稀釋流通股，非公認會計準則下每股淨利潤預計在1.52美元至1.54美元之間。

Some additional notes for our guidance. First, we expect net interest income and net interest and other income for fiscal 2023 to be approximately $95 million. We expect tax expense in the fiscal year to be $12 million to $14 million. And finally, we expect capital expenditures and capitalized software together to be in the 3% to 4% of revenue range in fiscal 2023.

一些補充說明，以補充我們的預期。首先，我們預計2023財年的淨利息收入以及淨利息及其他收入約為9,500萬美元。我們預計本財年的稅費支出為1200萬至1400萬美元。最後，我們預計2023財年的資本支出和資本化軟體支出合計將佔營收的3%至4%。

And now regarding 2024. It is too early for us to speak to 2024 revenue growth. We will digest the information we see over the next couple of months and give you our 2024 revenue guidance next quarter. As it relates to non-GAAP profitability, our operating income and margins were a little higher in Q3 than we targeted as usage growth improved from Q2 levels, and we were successful with our cost efficiencies.

現在談2024年。現在談論2024年的營收成長還為時過早。我們將在未來幾個月內消化所看到的訊息，並在下個季度發布2024年的收入預期。就非公認會計準則獲利能力而言，由於使用量成長較第二季度有所改善，並且我們的成本效率也得到了成功提升，因此我們第三季度的營業收入和利潤率略高於我們的目標。

We expect continued strong execution on profitability in Q4. At the same time, we continue to be excited by our numerous long-term growth opportunities and we have no shortage of investments to make and are confident in our ability to execute to strong ROI on those investments.

我們預計第四季度獲利能力將持續強勁成長。同時，我們持續對許多長期成長機會感到興奮，我們並不缺乏投資機會，並且有信心在這些投資上實現強勁的投資報酬率。

As a result, while we are not providing 2024 margin guidance at this point, as always, we will balance our investments in long-term growth with margin discipline. And we will update you on that in more detail next quarter. With that, we will open the call for questions. Operator, let's begin the Q&A.

因此，雖然我們目前不提供2024年的利潤率指引，但一如既往，我們將在長期成長投資和利潤率控制之間取得平衡。我們將在下個季度更詳細地更新這方面的情況。接下來，我們將開始問答環節。接線員，讓我們開始問答環節。

(Operator Instructions) Our first question comes from Mark Murphy of JPMorgan.

（操作員指示）我們的第一個問題來自摩根大通的馬克墨菲。

Congratulations on a very strong performance. Olivier, I'm interested in your mention of 2.5% of ARR being driven by the native AI providers. Should we think of that mostly consisting of open AI, LAMA, anthropic, Cohere, et cetera? Or are you meaning that as a slightly different reference? And can you just help us understand, is that up from close to 0 a year ago? And then I have a quick follow-up.

恭喜您取得如此出色的業績。 Olivier，我對您提到的2.5%的年度經常性收入（ARR）來自原生AI提供者很感興趣。我們是否應該認為這主要包括開放式AI、LAMA、人擇演算法、Cohere等等？還是您指的是一個略有不同的參考？您能否幫我們理解一下，這個數字是否比一年前接近0有所上升？然後我有一個快速的後續問題。

Yes. So it's a number of companies that, without naming anyone, so they tend to be model providers, but not just on the language side, like a providers on the language, image side, like there's a number of different -- or video side, there's a different types of communities there given some good pilot-type companies. These customers all had revenue 1 year ago, but they've been growing a little bit faster than the rest of the customer base recently. And the reason we did that in the crisis -- we -- today, we see the usage growth related directly to AI coming mostly for these customers that provide models to others. Whereas we see broad usage of AI functionality across the customer base, but at low volumes, and it corresponds to the fact that for most customers or most enterprises really, they're still in the early stages of developing and shipping applications. So for now, the usage is concentrated among the model providers.

是的。有很多公司，我不願透露具體名稱，但它們往往是模型提供商，而且不僅僅是語言方面的，還有語言、圖像方面的提供商，比如有很多不同類型的——或者視頻方面的——它們有一些優秀的試點公司，形成了不同類型的社區。這些客戶一年前都有收入，但最近他們的成長速度比其他客戶群略快一些。我們在危機中這樣做的原因——我們——今天，我們看到與人工智慧直接相關的使用量成長主要來自那些為他人提供模型的客戶。雖然我們看到人工智慧功能在整個客戶群中廣泛使用，但使用量較低，這與大多數客戶或大多數企業仍然處於開發和發布應用程式的早期階段這一事實相符。因此，目前，使用量集中在模型提供者身上。

Okay. Yes. That make sense. And Olivier, as a quick follow-up, you mentioned the log management has cost $500 million in ARR. It's quite a milestone. You also mentioned the replacement of some legacy products. I'm curious if you see the acquisition of Splunk or any other acquisition activity in that market as a beneficial development. I'm just wondering if small customers or other companies that have been provided, that have been acquired, excuse me, might be looking for any alternatives out there that are more modern and a more converged platform and if you're seeing that in the pipeline.

好的，是的，很有道理。 Olivier，我想快速跟進一下，您提到日誌管理的年度經常性收入（ARR）達到了5億美元。這是一個里程碑。您也提到了一些舊產品的替換。我很好奇，您是否認為收購Splunk或該市場中的任何其他收購活動是一個有益的發展。我只是想知道，那些已經被收購的小型客戶或其他公司，抱歉，他們是否正在尋找更現代化、更融合的平台，以及您是否正在考慮這些替代方案。

I mean, we've seen that for a while now that customers were looking for more integrated platforms, more model offerings, things that were more cloud first. And that's been one of the reasons of our success in landing largely in brand-new applications, brand new environment, brand new cloud initiatives and then over time, consolidating our customers away from whatever they were using in the legacy. We don't think that it is going to change with the various acquisitions and tech privates that we've seen over the past quarter. So we think we will just see more of that over time.

我的意思是，我們已經看到，客戶一直在尋求更整合的平台、更多的模型產品以及更「雲端優先」的產品。這也是我們成功落地全新應用程式、全新環境和全新雲端計畫的原因之一，隨著時間的推移，我們幫助客戶擺脫了他們之前使用的傳統系統。我們認為，這種情況不會因為過去一個季度發生的各種收購和技術私有化而改變。因此，我們認為隨著時間的推移，我們會看到更多這樣的情況。

One exciting bit on your first question, interesting bit, if you -- as I know many of you are trying to understand what the AI landscape is made of. Interestingly enough, the -- when we look at our cohort of customers that are that we consider to be AI native and built largely on AI in all AI providers, they tend to be on different clouds. What we see is that the majority of those companies actually have a lot of their usage on AWS. Today, the larger part of the usage or the larger of these customers are on Azure. So we see really several different adoption trends there that I think are interesting to the broader market.

關於你的第一個問題，如果你——我知道你們很多人都想了解人工智慧的格局——有一個令人興奮的、有趣的點。有趣的是，當我們觀察那些我們認為是人工智慧原生客戶群，並且主要基於所有人工智慧提供者的人工智慧所建構的客戶群時，他們往往使用不同的雲端。我們發現，這些公司中的大多數實際上在AWS上大量使用了他們的服務。如今，這些客戶中的大部分使用量或更大規模的使用量都在Azure上。因此，我們看到了幾種不同的採用趨勢，我認為這對更廣泛的市場來說很有趣。

And our next question comes from Sanjit Singh of Morgan Stanley.

我們的下一個問題來自摩根士丹利的 Sanjit Singh。

Olivier, the company has been innovating throughout this downturn quite aggressively across core observability, security as well as AI. As we look into 2024, and we think about a potential new product cycle for Datadog, what parts of the portfolio do you think could be contributors either in 2024 -- later in '24 and 2025, what are the things that you face to customers will be most respective to? Just wanted to get a sense of where you think where the sort of the timing of some of these new products that you've been delivering over the past couple of years.

奧利維爾，在整個經濟低迷時期，公司一直在核心可觀察性、安全性以及人工智慧領域積極創新。展望2024年，我們思考Datadog潛在的新產品週期，您認為產品組合中的哪些部分可能在2024年、2024年後期以及2025年做出貢獻？您面對客戶時最關心的是什麼？我想了解一下，您認為過去幾年推出的一些新產品的時機是如何的？

Well, the -- mathematically, the products that will contribute the most to the growth next year are going to be the products that have been here the longest in the core observability products. We mentioned $1 billion in infrastructure, $0.5 billion in APM, $0.5 billion in logs. This is great, but still a small fraction of these products can be at scale, and we are primarily going after that. There's a number of other things we've been investing in and growing and we're fairly happy with the way things are going in security, as I mentioned on the call, we saw new packaging, we've also rolled out. And some of the new initiatives that stand, I would say, a little bit left or right of what we've been doing in observability.

嗯，從數學上講，明年對成長貢獻最大的產品將是核心可觀察性產品中歷史最悠久的產品。我們提到了10億美元的基礎建設投資、5億美元的APM投資和5億美元的日誌投資。這很棒，但這些產品中仍然只有一小部分能夠實現規模化，而我們主要關注的是這些產品。我們也正在投資和發展許多其他產品，我們對安全領域的進展相當滿意，正如我在電話會議上提到的，我們看到了新的產品組合，我們也已經推出了一些產品。我想說，一些新的舉措與我們在可觀察性領域的舉措略有不同。

This year, as you mentioned, was a year of innovation for us. I think it was also a year of cost optimization for customers. It's not necessarily the best year to get products to very quick -- extremely quick revenue growth, but we've planted a lot of seeds that we think are going to deliver in the next couple of years.

正如您所說，今年對我們來說是創新的一年。我認為，對客戶來說，這也是成本優化的一年。今年不一定是產品收入快速成長的最佳年份，但我們已經播下了許多種子，相信這些種子將在未來幾年結出碩果。

That's great. And then I had a sort of a follow-up question on the sort of new packaging for the DevSecOps, the new package. I was wondering if you could just give us a little bit of color around why you sort of what was the packaging approach and what you're trying to solve for? Is it sort of trying to adopt the capabilities in a single integrated capability? Or is it also about sort of consolidated pricing being potentially 1 SKU, 1 SKU price or to consume all these capabilities. So I just love some detail around -- on the motivation for these new packaging.

太好了。然後我有一個關於 DevSecOps 新包裝的後續問題。我想知道您能否稍微解釋一下為什麼採用這種包裝方法，以及您想解決什麼問題？是想把所有功能整合成單一的整合功能嗎？還是說是採用某種統一的定價方式，例如 1 個 SKU，1 個 SKU 的價格，或使用所有這些功能？所以我很想了解這些新包裝動機的細節。

Yes. A couple of things. So the first one is we security products have reached a certain level of maturity. And so we think they can be brought into the conversation with a larger set of our customers as opposed to being something that our customers sell select to which is how we started and how we start with most products really. But also we're trying to bring those products in the same conversation as the initial adoption of DevOps basically as opposed to having to branch that conversation into, oh, hey, you're doing operations and applications and can I interest you in some security with that, which would be a different conversation. So we -- so far, the signs for this are encouraging. And again, we think it goes with the broader market trends, the adoption of DevSecOps and what customers actually want to do, and we think is going to help them deliver better outcomes in security. But obviously, it's still -- we just rolled that out. So it's still too early to tell whether we got it right or whether we still need to tweak it a little bit.

是的，有幾件事。首先，我們的安全產品已經達到了一定的成熟度。因此，我們認為它們可以與更廣泛的客戶群進行對話，而不是像我們最初那樣，只向客戶銷售部分產品，而大多數產品實際上都是這樣起步的。同時，我們也試圖將這些產品與 DevOps 的初始採用放在一起討論，而不是像之前那樣，把話題延伸到“嘿，你在做運維和應用程序，我能不能讓你對一些相關的安全問題感興趣？”，那將是另一個話題。所以，到目前為止，這方面的跡象令人鼓舞。而且，我們認為這與更廣泛的市場趨勢、DevSecOps 的採用以及客戶的實際需求相符，我們認為這將有助於他們在安全方面取得更好的成果。但顯然，這仍然只是——我們才剛剛推出。所以現在判斷我們是否做對了，或者是否還需要進行一些調整還為時過早。

And our next question comes from Raimo Lenschow of Barclays.

我們的下一個問題來自巴克萊銀行的 Raimo Lenschow。

Congrats from me as well. Olivier, like we're almost a year into this kind of current situation and you saw Q2, obviously, saw the digital natives that you commented, just kind of having extra savings, but we're now back to kind of Q1 usage pattern. What do you see in terms of changing behavior on customers, not thinking into market. More like how do you think about observability and how that potentially would change the world as we think about '24, '25 coming out of this in terms of kind of vendor consolidation, how to kind of build observability, et cetera? And then I have one follow-up for David.

我也祝賀你。 Olivier，我們進入這種現狀已經快一年了，你在第二季度看到了你提到的數位原生代，他們確實節省了不少成本，但現在我們又回到了第一季度的使用模式。你認為顧客行為會有什麼變化，而不是市場層面。更確切地說，你如何看待可觀察性？在我們考慮2024年、2025年擺脫困境時，可觀察性會如何潛在地改變世界？例如，在供應商整合、如何建構可觀察性等方面。然後我還有後續問題想問David。

So we think the trends of vendor consolidation will continue. So basically, customers are getting more of it -- more mature in their needs, they're getting further into the cloud. And they're -- as part of that, they will want to add lesser integrators. And if they can use one platform instead of 12 different products, that's something that they all react very positively to and we see that again and again as we expand into our customers.

所以我們認為供應商整合的趨勢將會持續下去。基本上，客戶的需求越來越成熟，他們正在進一步深入雲端。作為其中的一部分，他們會希望添加一些規模較小的整合商。如果他們可以使用一個平台而不是12種不同的產品，他們都會非常積極地做出反應，隨著我們不斷拓展客戶群，這種情況屢見不鮮。

In terms of the broader trends, I think we -- it's too early to tell exactly what the next couple of quarters are going to be made of. We said -- it looks like we've hit an inflection point. It looks like there's a lot less overhang now in terms of what needs to be optimized or could be optimized by customers. It looks like also optimization is more -- is less intense and less widespread across the customer base. So all that is positive. Now there's still quite a bit of uncertainty in the macro environment. So I don't think we should get ahead of ourselves either and declare that it's the end of it for the foreseeable future. So we feel positive about things. But it's still hard to know exactly what's going to happen in a couple of quarters from now.

就更廣泛的趨勢而言，我認為現在預測未來幾季的走勢還為時過早。我們之前說過，看起來我們已經到達一個轉捩點了。現在看來，客戶需要優化或可以優化的地方已經少很多了。而且，優化的強度和覆蓋範圍似乎也更小了。所以這一切都是正面的。目前宏觀環境仍存在相當多的不確定性。所以我認為我們也不應該操之過急，斷言在可預見的未來，這種情況已經結束。我們對未來持樂觀態度。但目前還很難確切知道幾季後會發生什麼事。

From a buying behavior perspective, we've never seen customers really slow down in their intent to move to the cloud and direct adoption of new platforms, new products from us. We've done well, we scaled the New Logo acquisition with scale, the new product Hitachi. So that has been a constant throughout this optimization phase. In terms of the user growth and customers scaling new workloads, I don't think we're back to the exibirant times of 2021. But maybe we are reverting back to the mean of what was happening before that.

從購買行為的角度來看，我們從未見過客戶真正放緩遷移到雲端以及直接採用我們新平台和新產品的意願。我們做得很好，我們擴大了對New Logo的收購，以及新產品Hitachi的規模。因此，在整個最佳化階段，這始終是不變的。就用戶成長和客戶擴展新工作負載而言，我認為我們還沒有回到2021年的輝煌時期。但也許我們正在回歸到之前的平均值。

Okay. And if I can squeeze one quick one and David, we talked about OpEx and OpEx growth and this quarter was lower than others. Should I read your comments about next year. And clearly, there is a big investment opportunity that OpEx growth like maybe wasn't quite where you wanted it to be? Just any comment there? And I know you just can't guide.

好的。如果我可以快速提一個問題，David，我們討論了營運支出和營運支出成長，本季的營運支出低於其他季度。我應該看看你對明年的評論嗎？顯然，這是一個巨大的投資機會，但營運支出成長可能沒有達到你的預期？有什麼評論嗎？我知道你無法提供指導。

Yes. I think as we talked about the movement of the top line because of consumption moves more quickly than we can adjust resources. So we've taken more of an optimization and cost prioritization this year. But we offset that, we do think there's a very large opportunity. So we are expecting to increase the level of investment. But as we say that, we've always been looking at the balance between maximizing the top line growth with producing profit and are going to continue to operate on that taking advantage of the long-term opportunities.

是的。正如我們之前提到的，我認為由於消費變化速度快於我們調整資源的速度，導致營收變動。因此，我們今年更加重視優化和成本優先排序。但我們抵消了這些影響，認為這是一個巨大的機會。因此，我們預計將增加投資。但正如我們所說，我們一直在尋找營收成長最大化與獲利之間的平衡，並將繼續以此為基礎，抓住長期機會。

And our next question comes from Karl Keirstead of UBS.

我們的下一個問題來自瑞銀的 Karl Keirstead。

Maybe David, I'll direct this to. I was intrigued by your comment that the fourth quarter, the month of October was off to a healthy start understanding that that's just a month. But just curious if you could unpack that a little bit, largely because investors on this call, we are picking up signals from other tech firms that would suggest a still very tough macro environment or maybe even slightly tougher. So I'm just curious where you might be seeing pockets of strength, if you could add a little more color.

也許可以，大衛，我會把這個問題轉給你們。我對你說第四季度，也就是十月開局良好，這話很感興趣，但我知道這只是一個月的數據。我好奇你能否稍微解釋一下，主要是因為參加這次電話會議的投資者，我們從其他科技公司那裡聽到了一些信號，這些信號表明宏觀環境仍然非常艱難，甚至可能略有惡化。所以我很好奇，你認為哪些方面可能表現強勁，能否再補充一些細節？

Yes. I think it's just essentially what we try to do in the last couple of quarters is to cushion everybody that we still expect that there will be optimization and cost management, but give everyone a flavor for the direction. And what we're seeing, as I think Olivier and I mentioned, we're seeing the continuation of that but at a more moderated level and that was across the customer base. So we're seeing essentially clients leaning a little more into growth, again, early in the quarter, too early to call it, but the trends seem to be a moderation of the previous cost management and optimization, although it's still continuing.

是的。我認為，我們在過去幾季所做的本質上就是向所有人傳達我們仍然預期會進行最佳化和成本管理的預期，但同時也讓每個人都能大致了解未來的方向。正如我和奧利維爾所提到的，我們看到這種情況仍在繼續，但幅度會有所緩和，而且是在整個客戶群中都如此。因此，我們看到客戶實際上更傾向於成長，雖然這在本季度初還為時過早，但趨勢似乎是對先前成本管理和優化的緩和，儘管這種趨勢仍在持續。

Yes, those color on the -- so yes, we had a healthy start to Q4. We we see trends that are as strong as they've been for the past year in terms of what happened early in the quarter. That being said, Q4 is a tough quarter to call because it has fairly high seasonality. There's a typically a drop of usage in the -- at the very end of the quarter with the holidays. And that drop in different years has been -- can be more or less pronounced. Last year, in particular, it was very pronounced. So we're -- we've given guidance with all that in mind, basically.

是的，這些顏色……所以是的，我們第四季開局都很好。從季度初的情況來看，我們看到的趨勢與去年一樣強勁。話雖如此，第四季的走勢難以預測，因為它的季節性相當強。通常情況下，在季度末假期期間，使用量會下降。而且不同年份的下降幅度可能或多或少。尤其是去年，下降幅度非常明顯。所以我們基本上是在考慮所有這些因素後給出的指導。

Let me just add because this question has been talked about. Like last quarter, we didn't take the strength of October account, we took the exact same guidance approach, which was to take the weighted average historical trends and discount apply conservatism. So like last quarter, we had said that the first quarter looked a little more stable. We didn't take that into consideration on our guidance. And we have stuck to the exact same guidance methodology, which is to act with that conservatism.

我補充一下，因為這個問題已經有人討論過了。和上個季度一樣，我們沒有考慮10月份的強勁表現，而是採用了完全相同的指引方法，即採用加權平均歷史趨勢，並應用保守主義進行折現。所以和上個季度一樣，我們之前說過第一季看起來比較穩定一些。但我們在製定指引時並沒有考慮到這一點。我們一直堅持完全相同的指引方法，即採取保守主義的行動。

One last thing to -- just to comment on that. Because I know a lot of you are trying to understand how we fit with respect to the Large cap providers and core trends correspond to theirs. Remember that we are -- while our trends are similar in the long run, we, in the short term, there can be differences of timing in terms of when we're going to see certain effects where they're going to see them. We also have a different mix where the mix of cap provider is not exactly the same as the border market and a different mix of customers and geographies than the individual cloud providers as well. So things are not exactly one to one there.

最後，我想評論一下。因為我知道你們很多人都想了解我們與大型市值提供者的契合度，以及我們的核心趨勢與他們之間的對應關係。請記住，雖然我們的長期趨勢相似，但短期來看，我們和大型市值提供者在何時看到某些影響方面可能會存在時間差異。我們的組合也有所不同，其中市值提供者的組合與邊界市場並不完全相同，客戶和地理的組合也與各個雲端供應商不同。所以，情況並非完全一一對應。

Our next question comes from Matt Hedberg of RBC.

我們的下一個問題來自 RBC 的 Matt Hedberg。

And I'll offer my congrats as well. Oli, I wanted to double-click on some of the improved usage trends. Can you provide us with an overview on how some of your large sort of strategic customers think about optimization is really part of an ongoing IT spending strategy, coupled with what's driving some of these increased uses. I guess I'm wondering have IT executives from your perspective changed their view on the level of monitoring needed with new levels of workload?

我也要向你表示祝賀。 Oli，我想再次強調一些改進的使用趨勢。您能否概述一下，您的一些大型策略客戶如何看待優化作為持續IT支出策略的一部分，以及推動這些使用量成長的因素。我想知道，從您的角度來看，IT主管是否改變了他們對新工作負載所需監控等級的看法？

No. So they didn't change if you on the level of monitoring needed. I think they try to save money wherever they could save it. By far, the biggest area they can save in their cloud infrastructure is their cloud deal itself. As a reminder, when customers pay $1 to us, they pay $10 or $20 to their cloud provider. So there's a lot more sales to be had there. But these savings flow down to us. We charge commercial rate to the size of our customers infrastructure. They also tried to save what they could in observability specifically. And usually, there's always a bit of that you can cut. It can always sample certain things a bit more, you can retain your loss a bit less. You can remove some of the debug logging like things like that, that can drive your cost up, but don't necessarily generate a ton of value. And that's the behavior by the way, that we see all the time. Like we see it for most customers, once a year, maybe twice a year, sometimes usually before contract renegotiations and things like that, why they try to understand what they're going to need for the next 2 or 3 years. The big difference over the past year has been that everybody has been doing that at once and multiple times. It was really an environment where everyone was feeling very uncertain about the economy and needed to save money very quickly. So we expect optimization to continue as part of this macro trend in the near future. And in perpetuity after that, we'll have that continuous cycle of customers optimizing reducing what they can reduce and then growing workloads and maybe creating a little bit more mass over time as well. And then optimizing again on a regular basis, not everyone at the same time.

不。所以，就所需的監控等級而言，他們沒有改變。我認為他們盡可能地節省成本。到目前為止，他們在雲端基礎設施方面能夠節省的最大部分就是雲端交易本身。提醒一下，當客戶向我們支付 1 美元時，他們會向他們的雲端供應商支付 10 美元或 20 美元。因此，這方面可以帶來更多的銷售。但這些節省最終會流向我們。我們根據客戶基礎設施的規模收取商業費率。他們也特別在可觀察性方面盡可能地節省成本。通常情況下，總是有一些可以削減的地方。它可以多採樣一些特定內容，從而減少損失。你可以刪除一些調試日誌之類的東西，這些東西可能會增加成本，但不一定能產生很大的價值。順便說一句，我們經常看到這種行為。就像我們看到的大多數客戶一樣，他們每年一次，也許兩次，有時通常是在合約重新談判之類的事情之前，試圖了解未來兩三年的需求。過去一年最大的不同是，每個人都一次多次這樣做。當時的環境確實讓每個人都對經濟感到非常不確定，需要迅速節省開支。因此，我們預計優化將在不久的將來作為宏觀趨勢的一部分繼續下去。在那之後，我們將永遠看到客戶不斷優化，減少他們能減少的部分，然後增加工作負載，並可能隨著時間的推移創造更多的規模。然後定期再次進行優化，而不是同時對所有人進行優化。

And our next question comes from Fatima Boolani of Citi.

我們的下一個問題來自花旗銀行的 Fatima Boolani。

One for Oli and one for Dave, if I may. Olivier, the packaging and pricing motions that you discussed for the DevSecOps solution set. I wanted to zoom out generally a little bit in knowing what you know about how buying behavior and procurement behavior for a lot of your customers has changed over the course of this past year. I was wondering if you can kind of shed light on how you're thinking about an ELA or EAA-type selling motion. I know it's something that historically you've been averse to, but I'm curious how you're thinking about it in sort of the current day and age, if you will, in terms of how your customers have changed the way they're buying and deploying. And then for David, just wanted to get a sense of what expectations on net retention rates are built into your guidance? And in other words, should this be the quarter that we see a trough in extension rates or net retention rates as you're thinking about it in the guidance?

如果可以的話，我想分別向 Oli 和 Dave 提一個問題。 Olivier，您之前討論過 DevSecOps 解決方案集的打包和定價方案。我想稍微詳細地談談您對過去一年中許多客戶的購買行為和採購行為變化的看法。我想知道您能否解釋一下您對 ELA 或 EAA 類型的銷售方案有何看法。我知道您過去一直反對這種方案，但我很好奇，在當今時代，考慮到您的客戶購買和部署方式的變化，您是如何看待它的。然後，對於 David，我想了解一下，您的指導中對淨留存率的預期是如何的？換句話說，本季續訂率或淨留存率是否應該像您在指導中預期的那樣，出現低谷？

Yes. So on the ELAs, so in general, look, we're always very open to new approaches in packaging. We try to see core things are consumed from the customer side and what makes the most sense to them. Now ELAs and things that are very difficult or very inappropriate for a business like ours. One because we are fully SaaS-based and there's a very, very large volume dimension to absolutely everything we do for our customers. So it's very, very hard to provide one price it all -- for them. We also philosophically, we like to have -- to get signal from what customers are willing are not willing to pay for. And that drives a lot of our product innovation. We had a lot of good news this way because customers want to buy products and they scale them a lot. We'll also get bad news. We get customers who don't find there is enough value in a given product or think it should be doing more or I think you should be doing things differently or seems that the packaging doesn't make sense. And we like to have that. The reason why we simplified the pricing there, and we created those new SKUs is really to try and change the motion a little bit and integrate further the observability in the security side and make it easier to bring security into the conversation for these customers. So far, we have some early evidence that it's it seems to reason it, but it's way where we're too weary to call it. We'll need 2, 3 quarters of that to understand the implications of the new packaging.

是的。關於ELA，總的來說，我們始終對新的包裝方法持開放態度。我們努力從客戶的角度了解核心產品的消費情況，並思考什麼對他們來說最有意義。現在，ELA和其他一些對我們這樣的企業來說非常困難或不合適的產品。首先，因為我們完全基於SaaS，我們為客戶所做的一切都涉及非常非常廣泛的業務量。因此，很難為他們提供統一的價格。從理念上講，我們也希望了解客戶願意為哪些產品付費，不願意為哪些產品付費。這推動了我們的許多產品創新。透過這種方式，我們收到了很多好消息，因為客戶想要購買產品，而且他們會大幅擴展產品規模。我們也會收到壞消息。有些顧客認為某個產品的價值不夠，或認為它應該發揮更多作用，或是我認為應該採取不同的做法，或是覺得包裝不合理。我們喜歡看到這種情況。我們簡化定價並創建這些新SKU的原因，實際上是為了嘗試稍微改變流程，進一步整合安全方面的可觀察性，讓客戶更容易將安全納入討論之中。到目前為止，我們有一些早期證據表明，這似乎是合理的，但目前我們還無法確定。我們需要2到3個季度的時間來了解新包裝的影響。

Yes. And as it relates to net retention, we do not provide guidance, but just the way that we think about it and some of the drivers, we don't provide guidance, sorry, on net retention per se. But essentially, what we said was that Q3 organic growth was similar to what it was in Q1. And we have said previously that in Q2 and Q3 of last year, we had lower than before but higher than the Q1 -- Q4, Q1 and now Q3. So it really is a matter of lapping those comps. The comps have gotten increasingly easy to lap. And we will let everybody know if we do produce an organic growth that's higher this Q4 than it was in Q4 last year, we will have, in this period, trough in net retention and it will begin to head up. So that's sort of how we think about it. In terms of our guidance, that's a different story. As we said, we provide conservatism. So what's implicit in the guidance is something worse than we're experiencing, which would apply a lower organic growth. But in terms of the business trends themselves, it's really a matter of do we lap the Q4 and produce a higher organic growth in this -- in the Q4 coming up than we had in the Q4 last year.

是的。至於淨留存率，我們不提供指引，只是就我們對此的看法以及一些驅動因素而言，抱歉，我們不會就淨留存率本身提供指引。但本質上，我們所說的是，第三季的有機成長與第一季相似。我們之前也說過，去年第二季和第三季的有機成長低於之前，但高於第一季——第四季、第一季度，以及現在的第三季。所以，這實際上是一個與可比較數據進行比較的問題。可比數據越來越容易進行比較。我們會告訴大家，如果我們今年第四季的有機成長確實高於去年第四季度，那麼在此期間，我們的淨留存率將觸底，然後開始回升。這就是我們的想法。至於我們的指引，情況就不同了。正如我們所說，我們提供的是保守的指導。因此，指引中隱含的是比我們目前經歷的更糟糕的情況，這將導致有機成長率下降。但就業務趨勢本身而言，真正的問題是我們能否在第四季度取得比去年第四季更高的有機成長。

And our next question comes from Brent Thill of Jefferies.

我們的下一個問題來自 Jefferies 的 Brent Thill。

David, I was curious if you could focus on the enterprise traction and what you're seeing and maybe for Olivier. David, I guess, on the customer add, you're 700 and that's lower than your normal accruals out of 1,000-plus per quarter. Can you just talk about is that number being misled because the enterprise traction is higher and that number is, therefore, going to be lower? Could you just drive into that a little bit?

大衛，我很好奇您能否重點談談企業級業務的牽引力以及您所看到的情況，或許可以談談奧利維爾的情況。大衛，我想，在客戶成長方面，您的客戶成長是700個，這低於您每季1000多個的正常成長。您能否談談這個數字是否被誤導了，因為企業級業務的牽引力更高，因此這個數字會更低？能稍微解釋一下嗎？

Yes. As a reminder, we have a broad range of customers in the long tail. And similar to what we discussed last quarter, the gross number of ads, so the accumulation and in the enterprise, it was quite strong, was very similar to what we've experienced. And we have a very small tail that has a larger attrition rate doesn't have a lot of dollars attached to it. So the trends continue, which is very strong, as Olivier mentioned, new logo accumulation both in terms of number of new logos and ARR, and that was true when Olivier talk about the enterprise side of it. offset by this tail that has very little dollars associated with it.

是的。提醒一下，我們在長尾領域擁有廣泛的客戶群。與我們上季度討論的類似，廣告總量，也就是企業級廣告的積累，表現相當強勁，與我們之前經歷的非常相似。我們的長尾業務規模很小，流失率較高，但帶來的效益並不大。因此，正如Olivier所提到的，這種趨勢仍在持續，而且非常強勁。新標誌的積累，無論是新標誌數量還是年度經常性收入（ARR），Olivier在談到企業級廣告時也提到了這一點。但這種成長被與它相關的收益很少的長尾業務所抵消。

Yes. And just to comment on that, remember, the bottom half of our customer represents around 1% of our revenue. and the lower increase in customer number comes from the very, very low end, which is customers that pay us in the tens of dollars a month. And those customers we're getting a little bit less of those to start with. I think that's part of the economical environment. and the churn is a little bit higher than it used to be there, too. So that's why these numbers are bit depressed right now. That being said, we had a record number of new logos over $100,000. We're doing very well in the enterprise. We're doing very well in the new market also. We're also doing very well at the high end of the SMB. So we are very happy with all of the segments we're targeting with our sales and marketing motions today. On the enterprise side, we actually mentioned a few of those very exciting contracts on the core. We get really, really excited when we see very traditional enterprises moving to the cloud and adopting us and consolidate in world to all. David, is -- we consider in the room, David's smiling because he's excited when he sees the dental care company is excited when you see the network of convenience stores where we even instrument the fuel pumps. So these are great developments for us, and we're leaning hard into that.

是的。關於這一點，我想說一下，請記住，我們客戶群中排名靠後的一半大約占我們收入的1%。客戶數量成長放緩主要來自那些非常非常低端的客戶，也就是每月支付給我們幾十美元的客戶。而這些客戶一開始就少了一些。我認為這是經濟環境的一部分，而且客戶流失率也比以前高了一點。所以，這些數據目前略顯低迷。話雖如此，我們超過10萬美元的新標誌數量創下了紀錄。我們在企業市場表現非常出色。我們在新興市場也表現良好。我們在中小企業的高端市場也表現良好。因此，我們對今天銷售和行銷活動所瞄準的所有細分市場都非常滿意。在企業市場方面，我們實際上提到了一些非常令人興奮的核心合約。當我們看到非常傳統的企業遷移到雲端、採用我們的技術，並在全球範圍內實現整合時，我們感到非常非常興奮。大衛，我們想到在場的各位，他之所以微笑，是因為他看到牙科保健公司時感到興奮；看到我們甚至在便利店網絡（其中的燃油泵也配備了儀表）時也感到興奮。所以，這些對我們來說都是巨大的進步，我們正全力以赴地推進。

And our next question comes from Kash Rangan of Goldman Sachs.

下一個問題來自高盛的卡什·蘭根 (Kash Rangan)。

I'm sure that you're all smiling at your results. Two things. One is -- with respect to LLM monitoring, which was demo the DASH conference, which was I thought absolutely fascinating. I know you quantified, Oli, 2.5 percentage points of growth coming from general AI workloads. How do we think about the revenue opportunity for LL monitoring that it's at an early stage? And I also have a second question, slightly more controversial that when and if we run it to customers that think they're spending their bills for Datadog are getting to be a little bit on the larger side, that is a sign of success. But how do you ensure that, that success does not work against the company that it opens up the door for price competition from others?

我相信你們都對成果感到滿意。有兩點。一是關於LLM監控，它在DASH大會上進行了演示，我認為這非常令人著迷。我知道你量化了2.5個百分點的成長來自通用AI工作負載。我們如何看待LL監控的早期收入機會？我還有第二個問題，這個問題爭議性更大：如果我們把它推廣給那些認為Datadog帳單金額略高的客戶，這是否是一個成功的標誌？但是，你如何確保這種成功不會對公司不利，從而為其他公司的價格競爭打開大門？

Yes. So on the DevSecOps side, I think it's too early to tell how much the revenue opportunity there is in the tooling specific lab there. When you think of the whole spectrum of tools, the closer you get to the developer side to how are is to monetize and the further you get towards operations and infrastructure, the easier it is to monetize. You can ship things that are very useful and very accretive to our platform because they get you a lot of users, a lot of attention and a lot of stickiness that are harder to monetize. So we'll see where on the spectrum that is. What we know, though, is that broader Generative AI up and down the stack from the components themselves, the GPUs all the way up to the models and the various things that are used to orchestrate them and store the data and move the data around all of that is going to generate a lot of opportunity for us. We said right now, it's conciliated among the AI native largely model providers. But we see that it's going to broaden and concern a lot more of our customers down the road. And your second question was, so when we grow a lot, we are very successful with some customers, how do we not create a long-term issue where they spend a lot of money and that (inaudible). Look, it's a great situation to be in to have customers spend a lot of money on you and have to justify that value over time. I think it's very healthy. I think that, again, that would drive innovation and great product development. And our role there is to make sure we have a healthy partnership with customers every single step of the way. And we charge them an order of money to less than what they spend for their cloud infrastructure. Maybe two others are many to less than would spend on their R&D. And so we think we should be in a position of leverage where if we do our jobs right, we show a lot of value for our customers. We send them a lot of money. We make them a lot faster, and we have them generate a lot more revenue. So that's how we see things and we hold ourselves to.

是的。所以在 DevSecOps 方面，我認為現在判斷特定工具實驗室的獲利機會還為時過早。當你考慮整個工具範圍時，你越接近開發者端，就越容易實現獲利，越接近營運和基礎設施，就越容易獲利。你可以將非常有用且極具增值潛力的產品發佈到我們的平台上，因為它們能為你帶來大量用戶、大量關注和大量用戶黏性，而這些很難獲利。所以我們會看看它處於哪個範圍。不過，我們知道的是，更廣泛的生成式人工智慧，從元件本身、GPU 一直到模型，以及用於編排、儲存和行動數據的各種工具，都將為我們創造大量機會。我們目前說過，它主要在 AI 原生模型提供者之間進行協調。但我們認為，未來它會進一步擴展，並吸引更多客戶。你的第二個問題是，當我們快速發展，在某些客戶身上取得了巨大成功時，我們如何避免造成一個長期問題，讓他們花費大量資金，而且（聽不清楚）。你看，讓客戶在你身上花費大量資金，並且必須隨著時間的推移證明這種價值，這是一個很好的情況。我認為這非常健康。我認為這將推動創新和偉大的產品開發。我們的職責是確保我們與客戶在每一步中保持健康的合作關係。我們向他們收取的費用低於他們在雲端基礎設施上花費的費用。也許還有兩位客戶的費用遠低於他們在研發上的支出。因此，我們認為我們應該處於一種優勢地位，如果我們做得好，就能為客戶帶來巨大的價值。我們為他們帶來很多收入。我們讓他們更快完成工作，並讓他們創造更多的收入。這就是我們看待事物和堅持的態度。

And our next question comes from Alex Zukin of Wolfe Research.

我們的下一個問題來自 Wolfe Research 的 Alex Zukin。

Congrats on a great quarter. Maybe just two quick ones for me. Olivier, you mentioned the federal opportunity or you mentioned the federal activity in the quarter with a deal that was pretty interesting. That's something that we picked up is an area of excitement for you guys. Can you maybe just talk about what the opportunity there is over the next 12 months and beyond, maybe stack-ranked as a priority for you guys. And then I've got a quick follow-up.

恭喜您本季取得了優異的成績。我只想快速問兩個問題。奧利維爾，您提到了聯邦業務機會，或者說您在本季度的聯邦業務活動中達成了一筆非常有趣的交易。我們發現這是一個令你們感到興奮的領域。您能否談談未來12個月及以後有哪些機會，或者可以將其列為您的優先事項。然後我還有一個簡短的後續問題。

So I missed the domain you're talking about, sorry. Feds. So that's definitely an area of investment for us. It's -- to look, we're happy with 2 things happening with the fact that we're moving further and further into the various level of certifications needed. We're happy with the early success with some agencies where we are spreading and in those government agencies and getting to go some of those world to all. But we are only scratching the surface of what we can do inside and there's a level we need to do. Some of it on the certification side, on the product side, and some of it on the go-to-market side and making sure we have all the parts of the motion working. So we -- I expect that to be one of the main areas of investment on the go-to-market side. Next year in terms of new markets we're going after.

抱歉，我漏掉了你提到的領域。聯邦調查局。所以這絕對是我們投資的一個領域。我們對兩件事感到高興，一是我們在所需的各種認證級別方面正在不斷取得進展。我們對一些機構的早期成功感到滿意，這些機構包括政府機構，我們正在將這些認證推廣到世界各地。但我們目前只觸及了我們內部工作的皮毛，我們還需要做一些其他的工作。有些是在認證方面，有些是在產品方面，有些是在市場推廣方面，確保我們所有環節都正常運作。所以我預計這將成為市場推廣的主要投資領域之一。明年我們將開拓新市場。

Perfect. And then maybe on the -- back on the AI question. I guess maybe just we drill a little bit deeper within those AI native companies, the criticality of their database users, meaning are you seeing something different where -- in a world where these applications become more prevalent, there's the opportunity to kind of expand wallet share of observability becomes even more important. And how should we think about the growth opportunities from those types of workloads in either 2024 or 2025?

太好了。然後回到人工智慧的問題。我想，也許我們可以更深入地研究一下那些人工智慧原生公司，看看他們的資料庫使用者的重要性。這意味著，在這些應用程式變得越來越普遍的世界中，擴大可觀察性錢包份額的機會變得更加重要。我們該如何看待2024年或2025年這些類型的工作負載所帶來的成長機會？

Yes. So in general, the more complexity there is, the more useful observability, the more you see his value from writing code to actually understanding it and observing it. So to caricature if you -- if you spend a whole year writing 5 lines of code that are really very deep, you actually know those 5 lines pretty well, maybe you don't observe because you'll see you understand exactly how they work and what's going on with them. On the other hand, if thanks to all the major advances of technology and all of the very super source AI and you can just very quickly generate thousands of lines of code, ship them and start operating them, you actually have no idea hwo these work what they do. And you need a lot of tooling observability to actually understand that and keep driving that and secure it and do everything you need to do with it over time. So we think that overall, this increases in productivity are going to favor observability.. In terms of the future growth of AI, look, I think like everyone, we're trying to guess how transformative it's going to be. It looks like it's going to be pretty is, if you judge from just internally, how much of that technology we are adopting a how much is the productivity impact, it seems to be having. So again, today, we only see a tiny bit of it, which is early adoption by model providers and a lot of companies that are trying to scale up and experiment and figure out who it applies to their businesses and what they can ship to use the technology. But we think it's going to drive a lot of growth in the years to come.

是的。所以總的來說，複雜度越高，可觀察性就越有用，你就越能看到它從編寫程式碼到真正理解和觀察程式碼的價值。所以，舉個例子——如果你花了一整年的時間寫了5行非常深奧的程式碼，你對這5行程式碼非常了解，那麼你可能不會觀察，因為你會發現你完全理解了它們的工作原理和它們內部發生了什麼。另一方面，如果由於所有重大技術進步和所有超級人工智慧來源，你可以非常快速地產生數千行程式碼，發布並開始運行它們，那麼你實際上根本不知道它們是如何運作的。你需要大量的可觀察性工具來真正理解它，並持續驅動它，保護它，並隨著時間的推移完成你需要的一切。所以我們認為，總的來說，生產力的提升將有利於可觀察性。就人工智慧的未來發展而言，我想和所有人一樣，我們都在試圖猜測它將帶來多大的改變。看起來前景會很不錯。如果僅從內部來看，我們採用了多少這樣的技術，以及它對生產力的影響有多大。所以，今天我們只看到了一小部分，這些技術只是模型提供者和許多試圖擴大規模、進行實驗、探索哪些技術適合其業務、以及哪些產品可以使用該技術的公司的早期採用。但我們認為，它將在未來幾年推動巨大的成長。

And our next question comes from Jake Roberge of William Blair.

我們的下一個問題來自 William Blair 的 Jake Roberge。

Congrats on the great results. Olivier, you called out the 2.5 points from AI native customers a few times, but you've also said that the broader customer base should start adding AI workloads to our platform over time. When do you think that actually takes place and the broader customer base starts to impact that AI growth in more earnest?

恭喜您取得的優異成績。 Olivier，您多次提到AI原生客戶應該獲得2.5分，但您也說過，更廣泛的客戶群應該隨著時間的推移開始在我們的平台上添加AI工作負載。您認為什麼時候才能真正實現這一點，更廣泛的客戶群才能開始更積極地影響AI的成長？

We don't know. And I think it's too early to tell. For one part, there's some uncertainty in terms of -- these customers are being to figure out what it is they are going to ship to their own customers. I think everybody is trying to learn that right now and experiment it. And -- but the other part is also that right now, the innovation is largely concentrated among the model providers. And so it's rational right now for most customers to rely on those instead of they're deploying their own infrastructure. Again, we think it's slightly going to change. We see a lot of demand in interest in other ways to host models and run models and customers and all those things like that. But today, that's the -- these are the trends of the market today basically.

我們不知道。而且我認為現在下結論還太早。一方面，存在一些不確定性——這些客戶正在弄清楚他們要向自己的客戶交付什麼。我認為現在每個人都在努力學習和嘗試。另一方面，目前創新主要集中在模型提供者。因此，現在大多數客戶依賴這些模型提供者而不是部署自己的基礎設施是合理的。同樣，我們認為這種情況會略有改變。我們看到很多人對其他託管模型、運行模型、客戶等等的方式感興趣。但今天，這基本上就是當今市場的趨勢。

Okay. Helpful. And then just a follow-up on the optimization front. It sounds like the early optimizers took about a year to complete those initiatives. But what type of time lines are you seeing from kind of the second or third layer of customers that started their optimizations later in the game. Have those also started to stabilize given they weren't as large as the early optimizers? Just trying to kind of parse out the customer base there.

好的。很有幫助。然後是關於優化方面的後續問題。聽起來早期的優化人員花了大約一年的時間才完成這些計畫。但是，您從後期開始優化的第二層或第三層客戶那裡看到了什麼樣的時間軸？考慮到他們的規模不如早期的優化人員那麼大，他們現在是否也開始穩定下來了？只是想分析一下那裡的客戶群。

Yes. So we don't really know for sure. That's also why we are careful not to call an end to it forever. I would say that for customers that are not part of this initial cohort, there's less of an overhang. So the customers that were the early optimizers and that had the most acute optimization tended to be cloud native, so all in on the cloud, very heavy on IT spend in general and the -- substantially all of their IT being in the cloud. They tended to be also companies that were slightly high growth but low profitability that needed to pivot their financials over a fairly short amount of time. I think if you look at the rest of the customer base, they are mostly not in that situation. So we expect the behavior to be different.

是的。所以我們也不確定。這也是為什麼我們謹慎地避免永遠停止這項服務。我想說，對於那些不屬於這群初始客戶群的客戶來說，壓力較小。因此，那些早期優化且優化最徹底的客戶往往是雲端原生客戶，他們把所有資源都投入到雲端，整體上IT支出非常大，而且幾乎所有的IT業務都部署在雲端。這些客戶往往也是成長速度略快但獲利能力較低的公司，需要在相當短的時間內調整財務狀況。我認為，如果你看看其他客戶群，他們大多沒有這種情況。所以我們預期他們的行為會有所不同。

Thank you. This concludes the question-and-answer session. I would now like to turn it back to CEO, Olivier Pomel for closing remarks.

謝謝。問答環節到此結束。現在請執行長奧利維爾·波梅爾（Olivier Pomel）致閉幕詞。

Thank you very much. I want to thank everyone for attending the call today. I also want to take the opportunity to thank our customers for their trust. We know these are trying times with all the macro uncertainty and we thank them for their trust. I also want to thank our employees, all the Datadogs for a quarter of hard work and great successes. And on these good words, we'll all get back to work and get busy for the end of the year. So thank you very much.

非常感謝！我要感謝今天參加電話會議的各位。我也想藉此機會感謝客戶的信任。我們知道，現在是充滿宏觀不確定性的艱難時期，我們感謝他們的信任。我還要感謝我們的員工，感謝所有Datadogs團隊，感謝他們本季的辛勤工作和取得的巨大成功。帶著這些美好的祝愿，我們將重返工作崗位，為年底的業績而努力。非常感謝大家。

This concludes today's conference call. Thank you for participating, and you may now disconnect.

今天的電話會議到此結束。感謝您的參與，現在可以掛斷電話了。