MongoDB Inc (MDB) 2024 Q2 法說會逐字稿

Good day, and thank you for standing by. Welcome to the MongoDB Second Quarter Fiscal Year 2024 Earnings Conference Call. (Operator Instructions) Please be advised that today's conference is being recorded.

美好的一天，感謝您的支持。歡迎參加 MongoDB 2024 財年第二季度收益電話會議。 （操作員指示）請注意，今天的會議正在錄製中。

I would now like to turn the conference over to your speaker for today, Mr. Brian Denyeau. Please go ahead, sir. The floor is yours.

現在我想將會議交給今天的發言人布賴恩·丹尼奧先生。請繼續，先生。地板是你的。

Thank you, Lisa. Good afternoon, and thank you for joining us today to review MongoDB's Second Quarter Fiscal 2024 Financial Results, which we announced in our press release issued after the close of the market today. Joining me on the call today are Dev Ittycheria, President and CEO of MongoDB; and Michael Gordon, MongoDB's COO and CFO.

謝謝你，麗莎。下午好，感謝您今天與我們一起回顧 MongoDB 2024 年第二季度財務業績，我們在今天收盤後發布的新聞稿中宣布了這一結果。今天和我一起參加電話會議的是 MongoDB 總裁兼首席執行官 Dev Ittycheria；以及 MongoDB 首席運營官兼首席財務官 Michael Gordon。

During this call, we will make forward-looking statements, including statements related to our market and future growth opportunities. The benefits of our product platform, our competitive landscape, customer behaviors, our financial guidance and our planned investments. These statements are subject to a variety of risks and uncertainties, including the results of operations and financial conditions that can cause actual results to differ materially from our expectations.

在本次電話會議中，我們將做出前瞻性聲明，包括與我們的市場和未來增長機會相關的聲明。我們的產品平台的優勢、競爭格局、客戶行為、財務指導和計劃投資。這些陳述受到各種風險和不確定性的影響，包括可能導致實際結果與我們的預期存在重大差異的運營結果和財務狀況。

For a discussion of the material risks and uncertainties that could affect our actual results, please refer to the risks described in our quarterly report on Form 10-Q for the quarter ended April 30, 2023, filed with the SEC on June 2, 2023.

有關可能影響我們實際業績的重大風險和不確定性的討論，請參閱我們於 2023 年 6 月 2 日向 SEC 提交的截至 2023 年 4 月 30 日的季度 10-Q 表格季度報告中描述的風險。

Any forward-looking statements made on this call reflect our views only as of today, and we undertake no obligation to update them, except as required by law. Additionally, we will discuss non-GAAP financial measures in this conference call. Please refer to the tables in our earnings release on the Investor Relations portion of our website for a reconciliation of these measures to the most directly comparable GAAP financial measure.

本次電話會議中所做的任何前瞻性陳述僅反映我們截至今天的觀點，我們不承擔更新這些陳述的義務，除非法律要求。此外，我們將在本次電話會議中討論非公認會計準則財務指標。請參閱我們網站投資者關係部分的收益發布中的表格，了解這些指標與最直接可比的公認會計準則財務指標的對賬情況。

With that, I'd like to turn the call over to Dev.

有了這個，我想把電話轉給 Dev。

Thank you, Brian, and thank you to everyone for joining us today. I'm pleased to report that we had another exceptional quarter as we continue to execute well despite challenging market conditions. I will start by reviewing our second quarter results before giving you a broader company update. We generated revenue of $424 million, a 40% year-over-year increase and above the high end of our guidance. Atlas revenue grew 38% year-over-year representing 63% of revenue and is now a $1 billion-plus revenue run rate product. We generated non-GAAP operating income of $79 million for a record 19% non-GAAP operating margin. And we had another solid quarter of customer growth, ending the quarter with over 45,000 customers. Overall, we delivered an exceptional Q2. We had a healthy quarter of new business acquisition led by continued strength in new workload acquisition in our existing customers.

謝謝你，布萊恩，也感謝大家今天加入我們。我很高興地向大家報告，我們又度過了一個出色的季度，儘管市場條件充滿挑戰，但我們仍然表現出色。在向您提供更廣泛的公司最新情況之前，我將首先回顧我們第二季度的業績。我們的收入達到 4.24 億美元，同比增長 40%，高於我們指導的上限。 Atlas 收入同比增長 38%，佔收入的 63%，目前是一款收入運行率超過 10 億美元的產品。我們實現了 7,900 萬美元的非 GAAP 營業收入，非 GAAP 營業利潤率為創紀錄的 19%。我們又一個季度的客戶增長強勁，本季度末客戶數量超過 45,000 名。總的來說，我們第二季度的表現非常出色。我們在現有客戶的新工作負載獲取方面持續強勁，帶動了新業務獲取的健康發展。

From a new logo perspective, we added 1,900 new customers in the quarter. Our direct sales team had another strong quarter of enterprise customer additions. Finally, our Enterprise Advanced and other non-Atlas business significantly exceeded our expectations, another indication of our strong product market fit and the appeal of our run-anywhere strategy.

從新標誌的角度來看，我們在本季度增加了 1,900 名新客戶。我們的直銷團隊在企業客戶增加方面又迎來了一個強勁的季度。最後，我們的 Enterprise Advanced 和其他非 Atlas 業務顯著超出了我們的預期，這再次表明了我們強大的產品市場契合度以及我們隨處運行戰略的吸引力。

Moving on to Atlas consumption trends. The quarter played out slightly better than our expectations. Michael will discuss consumption trends in more detail. Finally, retention rates remained strong in Q2, reinforcing the mission criticality of our platform even in a difficult spending environment. As we told you in the past, our market is different from most other software markets because the unit of competition is a workload, not a customer. We started customer relationship by acquiring the first workload, and we grew from there acquiring incremental workloads over time. Over the last few years, we have oriented our entire company around winning more workloads.

接下來是 Atlas 消費趨勢。該季度的表現略好於我們的預期。邁克爾將更詳細地討論消費趨勢。最後，第二季度的保留率仍然很高，即使在困難的支出環境下，也增強了我們平台的使命關鍵性。正如我們過去告訴過您的，我們的市場與大多數其他軟件市場不同，因為競爭的單位是工作負載，而不是客戶。我們通過獲得第一個工作負載開始了客戶關係，並從那裡開始隨著時間的推移獲得增量工作負載。在過去的幾年裡，我們整個公司的定位是贏得更多的工作量。

Starting with product, at our New York user conference held in June, we made a number of product announcements that will position us to capture more workloads faster. We introduced Atlas stream processing, which enables developers to work with streaming data to build sophisticated event-driven applications. The flexibility of the document model and the power of the MongoDB query language provide a compelling and differentiated way to process streaming data compared to alternative approaches. Our early access program has meaningfully oversubscribed as customers realize they can use a familiar and easy approach to work with streaming data and immediately see value.

從產品開始，在 6 月份舉行的紐約用戶會議上，我們發布了許多產品公告，這些產品將使我們能夠更快地捕獲更多工作負載。我們引入了 Atlas 流處理，它使開發人員能夠使用流數據來構建複雜的事件驅動應用程序。與其他方法相比，文檔模型的靈活性和 MongoDB 查詢語言的強大功能提供了一種引人注目的差異化方式來處理流數據。我們的早期訪問計劃已獲得有意義的超額認購，因為客戶意識到他們可以使用熟悉且簡單的方法來處理流數據並立即看到價值。

We announced the general availability of relational migrator, which makes it easier for customers to migrate their existing relational applications to MongoDB. We are seeing increased adoption across industries and geographies. For example, a leading international retailer was able to leverage relational migrator to dramatically accelerate their migration off Oracle. We also announced Atlas Vector Search, which enables developers to store, index and query Vector embeddings, instead of having to bolt on vector search functionality separately, adding yet another point solution and creating a more fragmented developer experience. Developers can aggregate and process the vectorized data they need to build AI applications while also using MongoDB to aggregate and process data and metadata. We are seeing significant interest in our vector search offering from a large and sophisticated enterprise customers even though it's only -- still only in preview.

我們宣布關係遷移器全面上市，這使客戶可以更輕鬆地將現有關係應用程序遷移到 MongoDB。我們看到跨行業和地區的採用率有所增加。例如，一家領先的國際零售商能夠利用關係遷移器顯著加速從 Oracle 的遷移。我們還發布了 Atlas Vector Search，它使開發人員能夠存儲、索引和查詢向量嵌入，而不必單獨添加向量搜索功能，從而添加另一個單點解決方案並創建更加分散的開發人員體驗。開發人員可以聚合和處理構建 AI 應用程序所需的矢量化數據，同時還可以使用 MongoDB 聚合和處理數據和元數據。我們看到大型且成熟的企業客戶對我們的矢量搜索產品表現出極大的興趣，儘管它還只是預覽版。

As one example, a large global management consulting firm is using Atlas Vector Search for internal research applications that allows consultants to semantically search over 1.5 million expert interview transcripts.

舉一個例子，一家大型全球管理諮詢公司正在將 Atlas Vector Search 用於內部研究應用程序，使顧問能夠對超過 150 萬份專家訪談筆錄進行語義搜索。

Over time, AI functionality will make developers more productive to the use of code generation and code assist tools that enable them to build more applications faster. Developers will also be able to enrich applicants with compelling AI experiences by enabling integration with either proprietary or open source large language models to deliver more impact. Now instead of data being used only by data scientists who drive insights, data can be used by developers to build smarter applications that truly transform a business. These AI applications will be exceptionally demanding, requiring a truly modern operational data platform like MongoDB. In fact, we believe MongoDB has even stronger competitive advantage in the world of AI.

隨著時間的推移，人工智能功能將使開發人員更高效地使用代碼生成和代碼輔助工具，從而使他們能夠更快地構建更多應用程序。開發人員還能夠通過與專有或開源大型語言模型集成來豐富申請人的引人注目的人工智能體驗，以產生更大的影響。現在，數據不再僅由推動洞察力的數據科學家使用，而是可以由開發人員使用來構建真正改變業務的更智能的應用程序。這些人工智能應用程序的要求非常高，需要像 MongoDB 這樣真正現代化的操作數據平台。事實上，我們相信MongoDB在人工智能領域具有更強的競爭優勢。

First, the document models inherent in flexibility and versatility renders it a natural fit for AI applications. Developers can easily manage and process various data types all in one place. Second, AI applications require high performance, parallel computations and the ability to scale data processing on an ever-growing base of data. MongoDB supports its features, with features like shorting and auto-scaling. Lastly, it is important to remember AI applications have the same demands as any other type of application: Transactional guarantees, security and privacy requirements, tech search, in-app analytics and more. Our developer data platform that gives developer a unified solution to smarter AI applications.

首先，文檔模型固有的靈活性和多功能性使其非常適合人工智能應用程序。開發人員可以在一處輕鬆管理和處理各種數據類型。其次，人工智能應用需要高性能、並行計算以及在不斷增長的數據基礎上擴展數據處理的能力。 MongoDB 支持其功能，包括短路和自動縮放等功能。最後，重要的是要記住人工智能應用程序與任何其他類型的應用程序具有相同的要求：交易保證、安全和隱私要求、技術搜索、應用程序內分析等等。我們的開發者數據平台為開發者提供了更智能的人工智能應用程序的統一解決方案。

We are seeing these applications developed across a wide variety of customer types and use cases. For example, observe.ai is an AI start-up that leverages 40 billion parameter LLM to provide customers with intelligence and coaching that maximize performance of their frontline support and sales teams. Observe.ai processes and run models on millions of support touch points daily to generate insights for their customers. Most of this rich, unstructured data is stored in MongoDB. Observe.ai chose to build on MongoDB because we enable them to quickly innovate, scale to handle large and unpredictable workloads and meet their security requirements of their largest enterprise customers.

我們看到這些應用程序在各種客戶類型和用例中開發。例如，observe.ai 是一家人工智能初創公司，利用 400 億個參數的 LLM 為客戶提供情報和指導，以最大限度地提高其一線支持和銷售團隊的績效。 Observe.ai 每天在數百萬個支持接觸點上處理和運行模型，為客戶生成見解。這些豐富的非結構化數據大部分存儲在 MongoDB 中。 Observe.ai 選擇基於 MongoDB 進行構建，因為我們使他們能夠快速創新、擴展以處理大型且不可預測的工作負載，並滿足最大企業客戶的安全要求。

On the other end of the spectrum is one of the leading industrial equipment suppliers in North America. This company relies on Atlas and Atlas Device sync to deploy AI models at the edge. To their field teams mobile devices to better manage and predict inventory in areas with poor physical network connectivity, they chose MongoDB because of our ability to efficiently handle large quantities of distributed data and to seamlessly integrate between network edge and their back-end systems.

另一方面是北美領先的工業設備供應商之一。該公司依靠 Atlas 和 Atlas Device Sync 在邊緣部署 AI 模型。為了讓他們的現場團隊通過移動設備更好地管理和預測物理網絡連接較差地區的庫存，他們選擇了MongoDB，因為我們能夠高效處理大量分佈式數據，並能夠在網絡邊緣與其後端系統之間無縫集成。

As much as we innovate our products, we also continuously innovate on how we engage with our customers. We are highly focused on reducing friction in the sales process so we can acquire more workloads quickly and cost effectively given the large size of our market opportunity. Historically, the most significant source of friction has been negotiating with customers to secure an upfront Atlas commitment since it can be hard for customers to forecast consumption growth for a new workload. Given our high retention rates and the underlying consumption growth, several years ago, we began reducing the importance of upfront commitments in our go-to-market process to accelerate workload acquisition. This year, we took additional steps in that direction.

在我們創新產品的同時，我們也在與客戶互動的方式上不斷創新。我們高度重視減少銷售流程中的摩擦，以便考慮到我們巨大的市場機會，能夠快速且經濟高效地獲得更多工作量。從歷史上看，最重要的摩擦來源是與客戶談判以確保 Atlas 的前期承諾，因為客戶很難預測新工作負載的消耗增長。鑑於我們的高保留率和潛在的消費增長，幾年前，我們開始降低進入市場過程中前期承諾的重要性，以加速工作負載獲取。今年，我們朝這個方向採取了更多措施。

For example, we no longer incentivize reps to sign customers to 1-year commitments. Obviously, this has short-term impact on our cash flow, but positions us better for the longer term by accelerating workload acquisition. We are pleased with the impact these changes have had in the business in the first half of the year. Specifically, new workload acquisition has accelerated, especially within existing customers. We believe that our efforts to reduce friction are resulting in more efficient growth, and we'll always look for ways to improve our go-to-market approach to make it even easier for customers to bring new workloads onto our platform.

例如，我們不再激勵銷售代表與客戶簽署一年期承諾。顯然，這對我們的現金流有短期影響，但通過加速工作負載獲取，使我們在長期內處於更好的位置。我們對這些變化在今年上半年對業務產生的影響感到高興。具體而言，新工作負載的獲取速度加快，尤其是在現有客戶中。我們相信，減少摩擦的努力將帶來更高效的增長，並且我們將始終尋找改進我們的上市方法的方法，使客戶能夠更輕鬆地將新的工作負載引入我們的平台。

Now I'd like to spend a few minutes reviewing the adoption trends of MongoDB across our customer base. Customers across industries, including Renault, Hootsuite and Ford are running mission-critical projects in MongoDB Atlas, leveraging the full power of our developer data platform. One of the 2023 MongoDB North American innovation award winners is Ford. With a focus on innovation, quality and customer satisfaction, Ford is a leader in the automotive industry and a household name around the world. Ford is committed to developing advanced technologies that enhance the safety, performance and sustainability of its vehicles. Their data explorer and transportation mobility cloud applications aggregate customer vehicle data from 24 different sources at a volume ranging up to 15 terabytes. Since migrating to MongoDB Atlas from their previous solution, Ford has seen a 50% performance improvement and faster read-write times.

現在我想花幾分鐘回顧一下 MongoDB 在我們客戶群中的採用趨勢。包括雷諾、Hootsuite 和福特在內的各行業客戶正在 MongoDB Atlas 中運行關鍵任務項目，充分利用我們開發人員數據平台的全部功能。福特是 2023 年 MongoDB 北美創新獎獲得者之一。福特注重創新、質量和客戶滿意度，是汽車行業的領導者，也是世界家喻戶曉的品牌。福特致力於開發先進技術，以提高其車輛的安全性、性能和可持續性。他們的數據瀏覽器和交通移動雲應用程序匯總了來自 24 個不同來源的客戶車輛數據，數據量高達 15 TB。自從從之前的解決方案遷移到 MongoDB Atlas 以來，Ford 的性能提高了 50%，讀寫時間也更快。

Cathay Pacific, Foot Locker and Market Access are examples of customers turning to MongoDB to free up the developer's time for innovation while achieving significant cost savings. Cathay Pacific, Hong Kong's home airline carrier operating in more than 60 destinations worldwide, turned to MongoDB on their journey to become one of the first airlines to create a truly paperless flight deck. Flight Folder, their application built on MongoDB consolidates dozens of different information sources into one place and includes a digital refueling feature that helps cruise become much more efficient with fueling strategies, saving significant flight time and costs. Since the Flight Folder launch, Cathay Pacific has completed more than 340,000 flights with full digital integration in the flight deck.

國泰航空、Foot Locker 和 Market Access 是客戶轉向 MongoDB 的例子，以釋放開發人員的創新時間，同時實現顯著的成本節省。國泰航空是香港的本土航空公司，在全球 60 多個目的地運營航班，在成為首批創建真正無紙化駕駛艙的航空公司之一的過程中，他們選擇了 MongoDB。 Flight Folder 是他們基於 MongoDB 構建的應用程序，將數十個不同的信息源整合到一個地方，並包含數字加油功能，可幫助郵輪通過加油策略變得更加高效，從而節省大量飛行時間和成本。自推出飛行文件夾以來，國泰航空已完成超過 340,000 次飛行，並在駕駛艙中實現了全數字化集成。

In addition to the greatly improved flight crew experience, flight times have been reduced and the digital refueling has saved 8 minutes on the ground on average. All these efficiencies have helped the company avoid the release of 15,000 tonnes of carbon and save an estimated $12.5 million.

除了極大改善機組人員的體驗外，飛行時間也減少了，數字加油平均節省了8分鐘的地面時間。所有這些效率幫助該公司避免了 15,000 噸碳的排放，並節省了約 1,250 萬美元。

Powerledger, Wells Fargo and System1 are among customers turning to MongoDB to modernize existing applications. System1, a customer and acquisition marketing company acquired MapQuest in 2019. At the time of the acquisition, MapQuest had a fragmented architecture that makes disparate data persistent technologies with third-party services. System1 selected Atlas as a key piece of MapQuest architecture transformation and has realized estimated cost reductions of 75% and performance improvements of 20% over its prior relational database solution. MapQuest is planning a number of future projects that will use Atlas Search and time series collections to improve the user experience and create a feedback loop on location-based relevancy in different cities.

Powerledger、Wells Fargo 和 System1 等客戶轉向 MongoDB 來實現現有應用程序的現代化。 System1 是一家客戶和收購營銷公司，於 2019 年收購了 MapQuest。收購時，MapQuest 擁有分散的架構，可以通過第三方服務實現不同數據的持久化技術。 System1 選擇 Atlas 作為 MapQuest 架構轉型的關鍵部分，與之前的關係數據庫解決方案相比，預計成本降低了 75%，性能提高了 20%。 MapQuest 正在規劃許多未來項目，這些項目將使用 Atlas 搜索和時間序列集合來改善用戶體驗，並針對不同城市基於位置的相關性創建反饋循環。

In summary, I'm incredibly stoked with our second quarter results. Our ability to win new workloads remain strong and our run anywhere strategy is resonating with customers. While it's early days on AI, we continue to see evidence that MongoDB will be a platform of choice for AI applications, just like we are for other modern and demanding applications. We continue to invest to maximize our long-term potential.

總之，我對我們第二季度的業績感到非常興奮。我們贏得新工作負載的能力依然強勁，我們的隨處運行策略引起了客戶的共鳴。雖然人工智能還處於早期階段，但我們不斷看到證據表明 MongoDB 將成為人工智能應用程序的首選平台，就像我們選擇其他現代且要求苛刻的應用程序一樣。我們繼續投資以最大限度地發揮我們的長期潛力。

With that, here's Michael.

邁克爾就是這樣。

Thanks, Dev. As mentioned, we delivered a strong performance in the second quarter, both financially and operationally. I'll begin with a detailed review of our second quarter results and then finish with our outlook for the third quarter and full fiscal year 2024. First, I'll start with our second quarter results. Total revenue in the quarter was $423.8 million, up 40% year-over-year. As Dev mentioned, we continue to see a healthy new business environment especially in terms of acquiring new workloads within existing customers. To us, this is confirmation we remain a top priority for our customers and that our value proposition continues to resonate even in this market.

謝謝，德夫。如前所述，我們在第二季度在財務和運營方面均取得了強勁的業績。我將首先詳細回顧我們的第二季度業績，然後以我們對第三季度和 2024 財年的展望結束。首先，我將從我們的第二季度業績開始。該季度總收入為 4.238 億美元，同比增長 40%。正如 Dev 提到的，我們繼續看到健康的新業務環境，特別是在現有客戶中獲取新工作負載方面。對我們來說，這證明我們仍然是客戶的首要任務，並且我們的價值主張即使在這個市場上也能繼續引起共鳴。

Shifting to our product mix, let's start with Atlas. Atlas grew 38% in the quarter compared to the previous year and represents 63% of total revenue compared to 64% in the second quarter of fiscal 2023 and 65% last quarter. In Q2, Atlas slightly declined as a percentage of revenue due to the exceptionally strong performance of our non-Atlas business, on scoring the demand for MongoDB regardless of where customers are in their cloud adoption journey. As a reminder, we recognize Atlas revenue primarily based on customer consumption of our platform and that consumption is closely related to end-user activity of the application, which can be affected by macroeconomic factors. Let me provide some context on Atlas consumption in the quarter. Consumption growth in Q2 was slightly better than our expectations. As a reminder, we had assumed Atlas would continue to be impacted by the difficult macro environment in Q2, and that is largely how the quarter played out.

轉向我們的產品組合，讓我們從 Atlas 開始。 Atlas 本季度較上年增長 38%，佔總收入的 63%，而 2023 財年第二季度為 64%，上季度為 65%。在第二季度，由於我們的非 Atlas 業務表現異常強勁，無論客戶處於雲採用之旅的哪個階段，Atlas 佔收入的百分比都略有下降，無論客戶對 MongoDB 的需求如何。提醒一下，我們確認 Atlas 收入主要基於我們平台的客戶消費，而消費與應用程序的最終用戶活動密切相關，而最終用戶活動可能受到宏觀經濟因素的影響。讓我提供一些有關本季度 Atlas 消費的背景信息。二季度消費增速略好於我們的預期。提醒一下，我們曾假設 Atlas 將繼續受到第二季度困難的宏觀環境的影響，而這在很大程度上就是該季度的表現。

Turning to non-Atlas revenues. EA significantly exceeded our expectations in the quarter, and we continue to have success selling incremental workloads into our EA customer base. We continue to see that our customers, regardless of their mode of deployment, are launching more workloads at MongoDB and moving towards standardizing on our platform. The EA revenue outperformance was in part a result of more multiyear deals than we had expected. In addition, we had an exceptionally strong quarter in our other licensing revenues. On our last call, we mentioned that we would benefit from a few large multiyear licensing deals, most notably the renewal and extension of our relationship with Alibaba. We also closed some additional multiyear licensing deals in the quarter, which is a meaningful contributor to our outperformance and another sign of popularity of MongoDB and the success of our run anywhere strategy.

轉向非 Atlas 收入。 EA 在本季度大大超出了我們的預期，並且我們繼續成功地將增量工作負載銷售給我們的 EA 客戶群。我們不斷看到我們的客戶，無論採用何種部署模式，都在 MongoDB 上啟動更多工作負載，並在我們的平台上實現標準化。 EA 收入表現出色的部分原因是多年期交易數量超出了我們的預期。此外，我們的其他許可收入季度也異常強勁。在上次電話會議上，我們提到我們將從一些大型的多年期許可交易中受益，最顯著的是我們與阿里巴巴關係的更新和擴展。我們還在本季度完成了一些額外的多年期許可交易，這對我們的優異表現做出了有意義的貢獻，也是 MongoDB 受歡迎以及我們隨處運行戰略成功的另一個標誌。

As a reminder, under ASC 606 for both EA and licensing contracts, the term license component even for multiyear deals, is recognized as upfront revenue.

提醒一下，根據 ASC 606 對於 EA 和許可合同的規定，即使對於多年交易，術語許可部分也被視為預付費用。

Turning to customer growth. During the second quarter, we grew our customer base by approximately 1,900 customers sequentially, bringing our total customer count to over 45,000, which is up from over 37,000 in the year ago period. Of our total customer count, over 6,800 are direct sales customers, which compares to over 5,400 in the year ago period. The growth in our total customer count is being driven primarily by Atlas, which had over 43,500 customers at the end of the quarter compared to over 35,500 in the year ago period. It's important to keep in mind that the growth in our Atlas customer count reflects new customers to MongoDB in addition to existing EA customers adding incremental Atlas workloads.

轉向客戶增長。第二季度，我們的客戶群連續增加了約 1,900 名客戶，使我們的客戶總數超過 45,000 名，高於去年同期的 37,000 名以上。在我們的客戶總數中，超過 6,800 名是直銷客戶，而去年同期為 5,400 多名。我們客戶總數的增長主要是由 Atlas 推動的，截至本季度末，Atlas 擁有超過 43,500 名客戶，而去年同期則超過 35,500 名。請務必記住，Atlas 客戶數量的增長反映了 MongoDB 的新客戶以及現有 EA 客戶增加的 Atlas 工作負載。

Let me double-click into our direct customer count. As Dev mentioned, we're becoming increasingly sophisticated in how we engage our customers. But some of those motions result in the line between our direct sales and our self-service channels becoming more fluid. I thought it'd be helpful to highlight 2 particular interchannel dynamics that impact the channel breakdown of our reported customer counts. While these customer movements represent less than 1% of our ARR, we do expect both of these trends to continue into the future, and so we wanted to make sure you understood how they affect our reported customer counts by channel.

讓我雙擊我們的直接客戶數量。正如 Dev 提到的，我們在吸引客戶方面變得越來越成熟。但其中一些舉措導致我們的直銷和自助服務渠道之間的界限變得更加流暢。我認為強調兩個影響我們報告的客戶數量的渠道細分的特定渠道間動態會有所幫助。雖然這些客戶流動僅占我們的 ARR 的不到 1%，但我們確實預計這兩種趨勢將持續到未來，因此我們希望確保您了解它們如何影響我們按渠道報告的客戶數量。

First, we are having increasing success leveraging cloud provider self-service marketplaces to drive new customer additions. Growth in cloud marketplace volumes is a major secular trend, and we are the only ISV available on all 3 hyperscaler marketplaces. Customers can deploy Atlas in second through cloud provider consoles and can pay for it by drawing down their existing cloud commitments. This further reduces friction as it bypasses the need for our contract altogether. For this reason, our direct sales team has been directing certain new prospects to sign up using self-service marketplaces. We've added several hundred customers using this approach in recent quarters, and these customers show up in our self-serve customer count even though we have a direct sales relationship with them.

首先，我們在利用雲提供商自助服務市場來推動新客戶增加方面取得了越來越大的成功。雲市場容量的增長是一個主要的長期趨勢，我們是唯一一家在所有 3 個超大規模市場上提供服務的 ISV。客戶可以通過雲提供商控制台立即部署 Atlas，並可以通過提取現有的雲承諾來支付費用。這進一步減少了摩擦，因為它完全繞過了我們的合同的需要。因此，我們的直銷團隊一直在指導某些新潛在客戶使用自助服務市場進行註冊。最近幾個季度，我們使用這種方法增加了數百名客戶，即使我們與他們有直接銷售關係，這些客戶也會出現在我們的自助服務客戶數量中。

Second, we continually review and analyze product usage signals to determine the potential of our customers. Because we are focused on velocity and efficiency of new workload acquisition, we're very careful not to deploy our reps on accounts where we don't see significant incremental benefit from sales rep coverage. If we determine that a direct sales customer can be supported more cost effectively in the self-serve channel, we'd prefer to free up the rep's time to focus on winning more new workloads. So far this year, we've moved over 300 small mid-market direct sales customers to the self-service channel.

其次，我們不斷審查和分析產品使用信號，以確定客戶的潛力。因為我們專注於新工作負載獲取的速度和效率，所以我們非常小心，不要將我們的代表部署到我們看不到銷售代表覆蓋範圍帶來的顯著增量收益的客戶上。如果我們確定可以在自助服務渠道中以更具成本效益的方式為直銷客戶提供支持，我們寧願釋放銷售代表的時間來專注於贏得更多新的工作負載。今年到目前為止，我們已將 300 多家中小型直銷客戶轉移到自助服務渠道。

Moving on to ARR. We had another quarter with our net ARR expansion rate above 120%. We ended the quarter with 1,855 customers with at least $100,000 in ARR and annualized MRR, which is up from 1,462 in the year ago period.

繼續討論 ARR。我們又一個季度的淨 ARR 擴張率超過 120%。本季度末，我們有 1,855 名客戶的 ARR 和年化 MRR 至少為 100,000 美元，高於去年同期的 1,462 名。

Moving down the income statement. I'll be discussing our results on a non-GAAP basis unless otherwise noted. Gross profit in the second quarter was $329 million, representing a gross margin of 78%, which is up from 73% in the year ago period. It is important to keep in mind that this quarter, we saw exceptional performance of our EA and licensing revenue, which contains a large upfront license component and very high margins, and therefore, we wouldn't expect to repeat this performance. Our income from operations was $79.1 million or a 19% operating margin for the second quarter compared to a negative 4% margin in the year ago period. Our strong bottom line results demonstrate the significant operating leverage in our model and are a clear indication of the strength in our underlying unit economics. The primary reason for our operating income results versus guidance is our revenue outperformance.

將損益表向下移動。除非另有說明，我將在非公認會計原則的基礎上討論我們的結果。第二季度毛利潤為3.29億美元，毛利率為78%，高於去年同期的73%。重要的是要記住，本季度我們的 EA 和許可收入表現出色，其中包含大量的前期許可組件和非常高的利潤，因此，我們預計不會重複這種表現。第二季度我們的運營收入為 7,910 萬美元，運營利潤率為 19%，而去年同期的利潤率為負 4%。我們強勁的盈利結果證明了我們模型中的顯著運營槓桿，並清楚地表明了我們基礎單位經濟效益的實力。我們的營業收入結果與指導相比的主要原因是我們的收入表現出色。

Net income in the second quarter was $76.7 million or $0.93 per share based on 82.5 million diluted weighted average shares outstanding. This compares to a net loss of $15.6 million or $0.23 per share on 68.3 million base weighted average shares outstanding in the year ago period.

根據 8250 萬股稀釋加權平均流通股計算，第二季度淨利潤為 7670 萬美元，即每股 0.93 美元。相比之下，上年同期淨虧損為 1560 萬美元，即 6830 萬股已發行基本加權平均股的淨虧損為每股 0.23 美元。

Turning to the balance sheet and cash flow. We ended the first quarter with 1 point -- ended the second quarter with $1.9 billion in cash, cash equivalents, short-term investments and restricted cash. Operating cash flow in the second quarter was negative $25.3 million. After taking into consideration approximately $2 million in capital expenditures and principal repayments of finance lease liabilities, free cash flow was negative $27.3 million in the quarter. This compares to negative free cash flow of $48.6 million in the second quarter of fiscal 2023.

轉向資產負債表和現金流量。第一季度結束時，我們以 1 個百分點結束——第二季度結束時，我們擁有 19 億美元的現金、現金等價物、短期投資和限制性現金。第二季度運營現金流為負2530萬美元。考慮到約 200 萬美元的資本支出和融資租賃負債的本金償還，本季度自由現金流為負 2730 萬美元。相比之下，2023 財年第二季度的自由現金流為負 4860 萬美元。

Three things of note on our cash flow performance in the quarter. First, as many of you know, Q2 tends to be our seasonally lowest collections quarter of the year because of low contract volumes in Q1, as evidenced by our Q1 ending accounts receivable balance. Second, while our revenue reflects the ASC 606 treatment of multiyear EA and licensing deals, most multiyear contracts are still billed annually, so there's no equivalent benefit to cash flow. Finally, as Dev mentioned, we continue deemphasizing the value of upfront commitments. So we're seeing fewer of them. In other words, we are intentionally collecting less cash upfront in order to win more workloads more quickly. As evidence of this, we grew Atlas revenue 38% year-over-year, while Atlas dollars committed upfront actually declined by 15% year-over-year.

本季度現金流表現值得注意的三件事。首先，正如你們許多人所知，第二季度往往是我們一年中季節性收款最低的季度，因為第一季度的合同量較低，正如我們第一季度期末應收賬款餘額所證明的那樣。其次，雖然我們的收入反映了 ASC 606 對多年期 EA 和許可交易的處理，但大多數多年期合同仍然按年計費，因此現金流沒有同等的好處。最後，正如 Dev 提到的，我們繼續淡化前期承諾的價值。所以我們看到的人越來越少了。換句話說，我們故意預先收集更少的現金，以便更快地贏得更多工作量。作為這一點的證據，我們的 Atlas 收入同比增長了 38%，而 Atlas 前期投入的資金實際上同比下降了 15%。

Lower upfront commitments only impact the timing of when our customers pay us, not the total payment. But this trend of declining upfront commitments will impact the relationship between our non-GAAP operating income and operating cash flow in the medium term.

較低的預付款承諾只會影響客戶向我們付款的時間，而不影響總付款。但這種預付承諾下降的趨勢將在中期內影響我們的非公認會計原則營業收入和營業現金流之間的關係。

I'd now like to turn to our outlook for the third quarter and full fiscal year 2024. For the third quarter, we expect revenue to be in the range of $400 million to $404 million, we expect non-GAAP income from operations to be in the range of $41 million to $44 million and non-GAAP net income per share to be in the range of $0.47 to $0.50 based on 83.5 million estimated diluted weighted average shares outstanding. For the full fiscal year 2024, we expect revenue to be in the range of $1.596 billion to $1.608 billion. For the full fiscal year 2024, we expect non-GAAP income from operations to be in the range of $189 million to $197 million and non-GAAP net income per share to be in the range of $2.27 to $2.35 based on 83 million estimated diluted weighted average shares outstanding. Note that the non-GAAP net income per share guidance for the third quarter and full fiscal year 2024 includes a non-GAAP tax provision of approximately 20%.

我現在想談談我們對 2024 年第三季度和整個財年的展望。對於第三季度，我們預計收入將在 4 億美元至 4.04 億美元之間，我們預計非 GAAP 運營收入為根據估計的8350萬股稀釋加權平均流通股，非GAAP 每股淨利潤將在4100 萬美元至4400 萬美元之間，每股淨利潤在0.47 美元至0.50 美元之間。對於 2024 年整個財年，我們預計收入將在 15.96 億美元至 16.08 億美元之間。對於 2024 年整個財年，我們預計非 GAAP 運營收入將在 1.89 億美元至 1.97 億美元之間，非 GAAP 每股淨利潤將在 2.27 美元至 2.35 美元之間（基於 8300 萬股估計稀釋加權）平均已發行股數。請注意，2024 年第三季度和整個財年的非 GAAP 每股淨利潤指引包括約 20% 的非 GAAP 稅收撥備。

I'll provide some more context on our guidance. First, we have modestly raised our Atlas outlook for the rest of the year, primarily to reflect a slightly stronger Q2 and therefore, a higher starting ARR for the second half. We continue to expect that Atlas consumption growth will be impacted by the difficult macroeconomic environment throughout fiscal '24. Our revised full year revenue guidance continues to assume consumption growth that is, on average, in line with the consumption growth we've experienced since the slowdown began in Q2 of last year, but with a slight seasonal benefit in Q3 and a slowdown in Q4 as observed over the last 2 years.

我將提供有關我們指導的更多背景信息。首先，我們適度上調了 Atlas 對今年剩餘時間的展望，主要是為了反映第二季度的略微強勁，因此下半年的起始 ARR 較高。我們繼續預計 Atlas 消費增長將受到整個 24 財年困難的宏觀經濟環境的影響。我們修訂後的全年收入指引繼續假設消費增長，平均而言，與我們自去年第二季度開始放緩以來所經歷的消費增長一致，但第三季度略有季節性效益，第四季度有所放緩根據過去 2 年的觀察。

Second, we expect to see a significant sequential decline in non-Atlas revenues in Q3 as we simply don't expect similar new business activity, especially when it comes to licensing deals. For that particular line of business, Q2 was extreme positive outlier.

其次，我們預計第三季度非 Atlas 收入將出現大幅連續下降，因為我們根本預計不會出現類似的新業務活動，尤其是在許可交易方面。對於該特定業務領域，第二季度是極端積極的異常值。

Third, we're raising our non-Atlas revenue estimate for the rest of the year, even though we don't expect our exceptional Q2 performance to repeat in the second half, our results in the first half give us incremental confidence in our run anywhere strategy. We continue to expect, however, that the difficult compare in the back half of the year will impact our non-Atlas growth rate.

第三，我們提高了今年剩餘時間的非 Atlas 收入預期，儘管我們預計第二季度的出色表現不會在下半年重演，但上半年的業績讓我們對自己的運營信心增強任何地方的策略。然而，我們仍然預計，今年下半年的困難比較將影響我們的非 Atlas 增長率。

Finally, thanks to strong performance in Q2 and the increased revenue outlook, we're meaningfully increasing our assumption for operating margins in fiscal '24 to 12% at the midpoint of our guidance, an improvement of more than 700 basis points compared to fiscal '23, while continuing to invest to pursue our long-term opportunity. As you update your models, please keep in mind that the majority of our planned fiscal '24 hiring will actually occur in the second half of the year.

最後，由於第二季度的強勁表現和收入前景的增加，我們有意義地將 24 財年營業利潤率的假設提高到指導中值的 12%，與 20 財年相比提高了 700 多個基點23 、同時繼續投資以追求我們的長期機會。當您更新模型時，請記住，我們計劃的 24 財年招聘的大部分實際上將在今年下半年進行。

To summarize, MongoDB delivered excellent second quarter results in a difficult environment. We are pleased with our ability to win new business and are demonstrating the operating leverage inherent in our model. While we continue to monitor the macro environment, we remain incredibly excited about the opportunity ahead, constantly to maximize our long-term value.

總而言之，MongoDB 在困難的環境下交付了出色的第二季度業績。我們對贏得新業務的能力感到滿意，並展示了我們模型中固有的運營槓桿。在我們繼續監控宏觀環境的同時，我們仍然對未來的機會感到無比興奮，不斷實現我們的長期價值最大化。

With that, we'd like to open it up to questions. Operator?

至此，我們願意接受提問。操作員？

(Operator Instructions) And our first question today will be coming from Raimo Lenschow of Barclays.

（操作員說明）今天我們的第一個問題將來自巴克萊銀行的 Raimo Lenschow。

Congrats from me on a great quarter. Two quick questions. First, the new -- or the trends you saw for EA or the numbers you saw for EA this quarter, you called out kind of bigger commitments from existing customers and taking more workloads or returning more workloads towards Mongo. Is that -- could you talk a little bit towards that in terms of is that a new trend? Was it just kind of very special this quarter? What are you seeing there? Because that's kind of in this sort of environment, seems like against what do you see from everyone else. So maybe couple of factors there? And then the second question is on the newer products like the streaming and vector databases, how does that feed into the revenue model for Mongo?

我對這個季度的表現表示祝賀。兩個簡單的問題。首先，新的趨勢，或者您在EA 看到的趨勢，或者您在本季度看到的EA 數字，您呼籲現有客戶做出更大的承諾，並承擔更多工作負載或將更多工作負載返回給Mongo。您能否談談這是一種新趨勢嗎？這個季度是不是有點特別？你在那裡看到什麼？因為在這種環境下，這似乎與你從其他人那裡看到的相反。那麼也許有幾個因素？第二個問題是關於流媒體和矢量數據庫等新產品，它們如何融入 Mongo 的收入模式？

Sure. So Raimo, the trends for EA, I think, is just an indication of our run anywhere strategy. We've been very committed to that strategy since the beginning. As you know, we started with EA and then introduced Atlas. But the whole point is that we give customers choice and we want to meet customers where they are in terms of what deployment model they want to use. And so I think this is just puts and takes of the quarter where we had a number of customers who wanted to double down on EA. And we also had some other non-Atlas business come in, in the quarter, which showed up in our results. But it's really a confirmation of the fact that we give customers choice and customers really appreciate that, and that's what you've seen in the results. With regards to, I think you said streaming and vector, those will show up in the Atlas revenue line as incremental consumption. There won't be a separate SKU. But what it will do is drive -- as those workloads come on, that will drive incremental consumption of Atlas, which will show up in the Atlas revenue line.

當然。所以 Raimo，我認為 EA 的趨勢只是我們隨處運行策略的一個指標。我們從一開始就非常致力於這一戰略。如您所知，我們從 EA 開始，然後推出了 Atlas。但重點是我們為客戶提供選擇，我們希望滿足客戶在他們想要使用的部署模型方面的需求。因此，我認為這只是本季度的情況，我們有許多客戶希望在 EA 上加倍投資。本季度我們還有一些其他非 Atlas 業務，這也反映在我們的業績中。但這確實證實了這樣一個事實：我們為客戶提供了選擇，並且客戶非常欣賞這一點，這就是您在結果中看到的。關於，我想你說的是流媒體和矢量，這些將作為增量消費出現在 Atlas 收入線中。不會有單獨的 SKU。但它要做的是推動——隨著這些工作負載的出現，這將推動 Atlas 的增量消費，這將顯示在 Atlas 收入線中。

Yes. I would just add that also as part of the broader development platform, it gives us the opportunity to win more workloads in the beginning. So you've got both sort of new workload penetration piece, which Dev mentioned, but also sort of the increased Atlas number is where it will show up on a revenue standpoint.

是的。我想補充一點，作為更廣泛的開發平台的一部分，它使我們有機會在一開始就贏得更多的工作負載。因此，你既有 Dev 提到的新的工作負載滲透部分，又有增加的 Atlas 數量，這將在收入的角度顯示出來。

Our next question will be coming from Keith Weiss of Morgan Stanley.

我們的下一個問題將來自摩根士丹利的基思韋斯。

This is Keith Weiss on for Sanjit Singh. One question for Dev and one for Michael. Dev, you guys talked about, I think, last quarter, 1,500 AI companies using MongoDB. You talked a lot about your applicability for AI workloads. I think a question that a lot of investors have is like the time frame for when this actually creates real impacts to where it becomes significant tailwind or just software in general, but more specifically the question to you is, when do you think that becomes a significant tailwind for MongoDB? When will we see that more significantly in Atlas revenues.

我是基思·韋斯 (Keith Weiss)，替桑吉特·辛格 (Sanjit Singh) 發言。一個問題問 Dev，一個問題問 Michael。 Dev，我想你們在上個季度談到了 1,500 家使用 MongoDB 的人工智能公司。您談到了很多關於人工智能工作負載的適用性。我認為很多投資者都有一個問題，就像這實際上對它產生真正影響的時間框架，使其成為重要的順風車或只是一般的軟件，但更具體地說，你的問題是，你認為這什麼時候會成為一個MongoDB 的重要推動力？我們什麼時候才能在 Atlas 收入中看到更顯著的變化。

And then the question for Michael. You talked about the commitments coming down, the Atlas commitments coming down and that being a drag on operating cash flow. Any sense you could give us on how long that drag on OCF persists? Is there any way to size that impact over time?

然後是邁克爾的問題。您談到了承諾的減少，阿特拉斯承諾的減少以及對運營現金流的拖累。您能告訴我們這種對 OCF 的拖累會持續多久嗎？有什麼方法可以衡量隨著時間的推移產生的影響嗎？

So Keith, on AI, obviously, we're really excited about the opportunity that AI presents. We continue to add many more AI customers this quarter. In the short term, we're really excited about some of the use cases we're seeing. We talked about observe.ai, the management consulting company, a more traditional company using MongoDB for a very impactful AI use case. And longer term, we believe our developer data platform value prop will just drive more AI adoption. People want to use one compelling, unified developer experience to address a wide variety of use cases of which AI is just one of them. And we're definitely hearing from customers to being able to do that on one platform versus bolting on a bunch of point solutions is far more the preferable approach. And so we're excited about the opportunity there.

基思，關於人工智能，顯然，我們對人工智能帶來的機會感到非常興奮。本季度我們繼續增加更多人工智能客戶。從短期來看，我們對所看到的一些用例感到非常興奮。我們討論了observe.ai，這是一家管理諮詢公司，這是一家使用 MongoDB 來實現非常有影響力的 AI 用例的傳統公司。從長遠來看，我們相信我們的開發者數據平台價值支撐將推動更多人工智能的採用。人們希望使用一種引人注目的、統一的開發人員體驗來解決各種用例，人工智能只是其中之一。我們確實從客戶那裡聽說，能夠在一個平台上做到這一點，而不是使用一堆單點解決方案，這是更可取的方法。因此，我們對那裡的機會感到興奮。

And I think you had some questions -- on the other thing on partners, I do want to say that we're seeing a lot of work and activity with our partner channel on the AI front as well. We're working with Google in the AI start-up program, and there's a lot of excitement. Google had their next conference this week. We're also working with Google to help train Codey, their code generation tool to help people accelerate the development of AI and other applications. And we're seeing a lot of interest in our own AI innovators program. We've had lots of customers apply for that program. So we're super excited about the interest that we're generating.

我認為你們有一些問題——關於合作夥伴的另一件事，我確實想說，我們的合作夥伴渠道在人工智能方面也看到了大量的工作和活動。我們正在與穀歌合作開展人工智能初創項目，令人興奮不已。谷歌本週舉行了下一次會議。我們還與 Google 合作，幫助培訓 Codey，他們的代碼生成工具可以幫助人們加速人工智能和其他應用程序的開發。我們看到人們對我們自己的人工智能創新者計劃很感興趣。我們已經有很多客戶申請該計劃。因此，我們對我們所產生的興趣感到非常興奮。

And on your other question, Keith, it's been a multiyear journey where we've been focused on reducing friction and accelerating new workload adoption. I do think as we called out, we continue to make additional steps and Dev called out some of the specific incremental steps this year. I think it's part of a transition. If you look at the Atlas revenue growth, Atlas grew 38% year-over-year, but dollars collected upfront shrank 15%, right? And so that gives you a sense for the magnitude or the divergence there that's showing up in the op income versus OCF bridge. I think like most things, there'll be a transition time period, but then it will settle into a more normalized level, but we think we've still got like a little bit more transition to go as we kind of work through the balance of the year.

關於你的另一個問題，Keith，這是一個多年的旅程，我們一直致力於減少摩擦並加速新工作負載的採用。我確實認為，正如我們所呼籲的那樣，我們將繼續採取額外的步驟，而開發人員今年也提出了一些具體的增量步驟。我認為這是過渡的一部分。如果你看看 Atlas 的收入增長，Atlas 同比增長 38%，但預收的資金減少了 15%，對吧？因此，這可以讓您了解運營收入與 OCF 橋中出現的規模或差異。我認為像大多數事情一樣，會有一個過渡期，但隨後它將進入一個更加正常化的水平，但我們認為，在我們努力實現平衡的過程中，我們仍然需要更多的過渡今年的。

Our next question will be coming from Kash Rangan of Goldman Sachs.

我們的下一個問題將來自高盛的卡什·蘭根。

I'm sorry, I didn't hear my name. Congrats, Dev and Michael on the quarter. It's hard to put up this kind of operating margin performance being a database company at the scale that you're operating. So kudos on that.

抱歉，我沒有聽到我的名字。恭喜 Dev 和 Michael 在本季度的表現。作為一家數據庫公司，以您所經營的規模，很難維持這種營業利潤率表現。對此表示讚賞。

The relational migrator came off of beta and came into GA this quarter. So I wonder if that had any particular impact on the EA business because you've certainly upsided your modest expectations. I want to get a little bit more detail on how that pipeline of relational migrator beta customers should play out? Is it going to be showing up in Atlas? Or is it going to be showing up in enterprise, the on-prem version.

關係遷移器結束了測試階段，並於本季度進入正式版。所以我想知道這是否對 EA 業務有任何特殊影響，因為你肯定提高了你的適度期望。我想更詳細地了解關係遷移器測試版客戶的管道應該如何發揮作用？它會出現在 Atlas 中嗎？或者它會出現在企業本地版本中。

And on AI, just curious if you can quantify the level of consumption impact in the future to Atlas that you could attribute to the different new things that Mongo is working, whether it's AI or streaming. How should we think about the incremental opportunities for consumption offered by some of the new things you launched at MongoDB line in New York a couple of months back.

在人工智能方面，只是好奇你是否可以量化未來對 Atlas 的消費影響水平，你可以將其歸因於 Mongo 正在開發的不同新事物，無論是人工智能還是流媒體。我們應該如何看待幾個月前在紐約 MongoDB 系列推出的一些新產品所提供的增量消費機會。

Sure. So regarding relational migrators, it's important for investors to know that this is really a high-end enterprise play. That's where the bulk of the legacy relational market is. And relational migrator is designed to help customers reduce the switching cost of migrating off relational databases to MongoDB for both EA and Atlas. So it's a place to -- depending again on the customer's choice of their deployment model but it's really meant to reduce the switching costs. I would say that there was no real impact in terms of revenue of customers using relational migrator because we just only made it generally available in June. But there's a tremendous amount of excitement. We have a large pipeline of customers who are very interested and are actually starting to use relational migrator and projects have begun, but there was no real impact on the quarter.

當然。因此，對於關係型遷移者來說，投資者必須知道這確實是一種高端企業策略。這就是大部分傳統關係市場所在的地方。關係遷移器旨在幫助客戶降低將 EA 和 Atlas 的關係數據庫遷移到 MongoDB 的轉換成本。因此，它是一個地方——再次取決於客戶對其部署模型的選擇，但它的真正目的是降低轉換成本。我想說的是，對於使用關係遷移器的客戶的收入而言，並沒有真正的影響，因為我們只是在 6 月份才將其全面推出。但有很多令人興奮的事情。我們擁有大量非常感興趣並且實際上開始使用關係遷移器的客戶，並且項目已經開始，但對本季度沒有真正的影響。

Regarding your second question about some of the new products and the impact of AI long term, what I would say is we definitely believe it will have a big impact long term. We think that things like Vector Search just make it so much easier to build smarter applications on MongoDB that unified developer experience is a key differentiator. There's a really strong interest in our public preview product. We also see a lot of interest in stream processing. Stream processing is -- it's a use case that's really optimized for MongoDB. The data is typically JSON, the variability of the data lends itself to a document model that's much more flexible. It's obviously very developer-oriented, where all the alternatives are using very rigid schemas and are much more complicated to use, so that plays into our sweet spot. So we think we have a big opportunity there. But -- and so it's hard to quantify what that impact will be in the long term, but I will tell you that we're really excited and the interest level on the new products is incredibly high.

關於你的第二個問題，關於一些新產品和人工智能的長期影響，我想說的是，我們絕對相信它會產生巨大的長期影響。我們認為，像矢量搜索這樣的東西讓在 MongoDB 上構建更智能的應用程序變得更加容易，統一的開發人員體驗是一個關鍵的差異化因素。人們對我們的公共預覽產品非常感興趣。我們還看到人們對流處理很感興趣。流處理是一種真正針對 MongoDB 進行優化的用例。數據通常是 JSON，數據的可變性使其文檔模型更加靈活。顯然，它非常面向開發人員，所有替代方案都使用非常嚴格的模式，並且使用起來更加複雜，因此這正是我們的最佳選擇。所以我們認為我們在那裡有很大的機會。但是，因此很難量化這種長期影響，但我會告訴你，我們真的很興奮，人們對新產品的興趣非常高。

Our next question will be coming from Brad Reback of Stifel.

我們的下一個問題將來自 Stifel 的 Brad Reback。

I'm not sure Dev or Michael, but going back to the commentary on fewer upfront Atlas commits. Oftentimes, when customers sign multiyear deals and pay upfront, they get a better rate. So if we were to think about not having them pay you upfront and make long-term commits, is that a net margin benefit to you guys on the pricing side?

我不確定 Dev 或 Michael，但回到關於較少的前期 Atlas 提交的評論。通常，當客戶簽署多年協議並預付款時，他們會得到更好的價格。因此，如果我們考慮不讓他們預先向您付款並做出長期承諾，這對定價方面的淨利潤是否有利？

Yes. So a couple of things. Thanks Brad for the question. In general, for us, even before the sort of evolution and changes in multiyear deals, typically, they were not all paid upfront. Typically, ours has been annually billed. But yes, to your point, I think as we've reduced upfront commitment, you have a couple of dynamics. The key one is when we are not motivating it or providing an incentive to our sales force, and wind up being customer driven, the leverage in that negotiation shifts. And on the margin, that is helpful for ultimate pricing or discount and it winds up with sort of better pricing for us, less discounting to the customer.

是的。有幾件事。感謝布拉德的提問。總的來說，對我們來說，即使在多年交易的演變和變化之前，通常情況下，它們也不是全部預先付款。通常，我們的費用是按年計費的。但是，是的，就你的觀點而言，我認為隨著我們減少了前期承諾，你就有了一些動力。關鍵是，當我們沒有激勵它或為我們的銷售人員提供激勵，而最終以客戶為導向時，談判的槓桿就會發生變化。在邊際上，這對最終定價或折扣很有幫助，最終為我們帶來更好的定價，為客戶提供更少的折扣。

Excellent. And then on your commentary about second half hiring outpacing first half. Would it be correct to assume that the hiring environment is a little less competitive, so you might actually be able to find people more easily and get better pricing for them as well?

出色的。然後關於你對下半年招聘速度超過上半年的評論。假設招聘環境競爭不那麼激烈，那麼您實際上可能能夠更輕鬆地找到人才並為他們獲得更好的定價，這樣的假設是否正確？

Yes. What I would say, Brad, is that I would say, in general, obviously, the frothiness of a few years ago has abated, but for certain skill sets, there's still significant premium for talent, and we don't want to lower our bar just to optimize on cost. We pride ourselves on recruiting the best of the best in this industry. And we focus on paying market rates. And so while it's a little easier because the market is a little softer, I wouldn't suggest that all of a sudden, we're getting employees at a massive discount.

是的。布拉德，我想說的是，總的來說，幾年前的泡沫顯然已經減弱，但對於某些技能組合來說，人才仍然存在很大的溢價，我們不想降低我們的酒吧只是為了優化成本。我們為能招募到這個行業最優秀的人才而感到自豪。我們專注於支付市場利率。因此，雖然因為市場稍微疲軟而變得容易一些，但我不建議我們突然以大幅折扣招募員工。

Yes. I would think about it as availability rather than cost and then throw in some of the dynamics with some places around different return to office models and other things. I think that sort of incrementally is likely to provide opportunities in the back half of the year. And as they present themselves, we'll certainly pursue those.

是的。我會將其視為可用性而不是成本，然後在某些地方圍繞不同的重返辦公室模式和其他事物引入一些動態。我認為這種增量可能會在今年下半年提供機會。當他們展現自己時，我們一定會追求這些。

Our next question will be coming from Karl Keirstead of UBS.

我們的下一個問題將來自瑞銀集團 (UBS) 的卡爾·凱斯特德 (Karl Keirstead)。

Okay. Great. Maybe this one to Mike. Michael, I wouldn't normally ask about the other segment, but it's such an outlier if I could ask a 2-parter. First is, what surprised on the upside there? Was the Alibaba deal much larger than you thought? Or did you grab a few others, maybe you could unpack that? And secondly, you did tell us that the second half guidance assumes a significant decline in the non-Atlas business. Is it fair to assume that this other category might return back to the levels it was at pre the July quarter?

好的。偉大的。也許這是給邁克的。邁克爾，我通常不會詢問其他部分，但如果我可以詢問兩人，那就太離群了。首先，有什麼令人驚訝的好處？阿里巴巴的交易比你想像的要大得多嗎？或者你抓了一些其他的，也許你可以打開包裝？其次，您確實告訴我們，下半年的指導假設非阿特拉斯業務將大幅下降。假設這個其他類別可能會回到七月份季度之前的水平是否公平？

Yes. Thanks, Karl. No, other deals, not Alibaba. Alibaba was baked at the time of the last guidance call. So it was sort of incremental deals that surprised us to the upside there. And yes, obviously, it's a volatile or variable, especially given the 606 and the nature of it where it goes, given the lumpiness of the term license revenue and things like that. And so, yes, I think that this is not repeatable performance, and I think it should settle back down to a more -- to a lower and more normalized level.

是的。謝謝，卡爾。不，是其他交易，不是阿里巴巴。阿里巴巴在上次指導電話會議時就已經出爐了。因此，增量交易讓我們對那裡的上漲感到驚訝。是的，顯然，它是一個不穩定或變量，特別是考慮到 606 及其發展方向的性質，考慮到許可收入一詞的波動性以及類似的事情。所以，是的，我認為這不是可重複的表現，我認為它應該回到一個更低、更規範的水平。

Okay. And then if I could ask a follow-up, Mike, you did a good job explaining the changes in the model and the licensing on cash flow, but it's not a metric you often talk about, but your deferred revenue balance was actually down year-over-year, highly unusual. Is this basically the same explanation that would be impacting DR?

好的。然後，如果我可以問後續問題，邁克，你很好地解釋了模型的變化和現金流許可，但這不是你經常談論的指標，但你的遞延收入餘額實際上是下降的一年- 與去年同期相比，非常不尋常。這與影響 DR 的解釋基本上相同嗎？

Yes. I think it's the same explanation or discussion overlaid with our recurring discussion around billings, and that's sort of not a metric that we focused on and that we sort of discourage people from using and that we're focused sort of on those workloads and winning new workloads rather than large upfront commitments. But one of the ways that plays out is absolutely in deferred and for anyone still doing defer -- I mean, calculated billings. Calculations that will affect that there as well, yes.

是的。我認為這是與我們圍繞賬單反复討論重疊的相同解釋或討論，這不是我們關注的指標，我們不鼓勵人們使用，我們專注於這些工作負載並贏得新的工作量而不是大量的前期承諾。但發揮作用的方式之一絕對是延期，對於任何仍在延期的人來說——我的意思是，計算賬單。是的，計算也會影響那裡。

One moment for the next question. Our next question is coming from Rishi Jaluria of RBC.

請稍等一下，回答下一個問題。我們的下一個問題來自 RBC 的 Rishi Jaluria。

This is Rich Poland on for Rishi Jaluria today. So I guess if we look at the workloads you have in front of you with Vector Search, relational migrator and streaming and could even throw an application search, which was more of a driver, let's call it, last year. But if we had to stack rank each of those workloads in terms of your positioning to win and your overall opportunity in each of those use cases. How would you go about doing that?

今天是 Rishi Jaluria 的富裕波蘭節目。因此，我想如果我們通過矢量搜索、關係遷移器和流媒體來查看擺在您面前的工作負載，甚至可以拋出應用程序搜索，這更像是一個驅動程序，我們稱之為去年的驅動程序。但是，如果我們必鬚根據您的獲勝定位以及在每個用例中的總體機會對每個工作負載進行堆疊排名。你會怎樣做呢？

Yes. So Rich, thanks for your question. I would say, obviously, the general operational workload of what you call the OLTP workload is still our bread and butter workloads that people come to us and relational migrator would just be more of that because of the migrating operational workloads of relational databases to MongoDB. And then the other use case really, other products really function of the use cases that customers are really interested in. For example, Atlas Device Sync, which is really focused on the enterprise mobility play, for example, point-of-sale devices for the retail industry in automotive, connected car in manufacturing, instrumenting the factory floor.

是的。非常豐富，謝謝你的提問。我想說，顯然，所謂的 OLTP 工作負載的一般操作工作負載仍然是人們來找我們的麵包和黃油工作負載，而關係遷移器會更多，因為關係數據庫的操作工作負載遷移到 MongoDB。然後是其他用例，其他產品確實能夠滿足客戶真正感興趣的用例。例如，Atlas Device Sync，它真正專注於企業移動應用，例如，銷售點設備汽車零售業、製造業互聯汽車、工廠車間儀表。

So it really depends on the use cases. In application search, we're really seeing an acceleration of large workloads in that -- for that product. So we're really excited about the size of some of the business that we're seeing there. Obviously, Vector is still in public preview. So we hope to have a GA sometime next year, but we're really excited about the early and high interest from enterprises. And obviously, some customers are already deploying it in production, even though it's a public preview product.

所以這實際上取決於用例。在應用程序搜索中，我們確實看到該產品的大型工作負載正在加速。因此，我們對在那裡看到的一些業務的規模感到非常興奮。顯然，Vector 仍處於公共預覽階段。因此，我們希望明年某個時候舉行一次 GA，但我們對企業的早期和高興趣感到非常興奮。顯然，一些客戶已經在生產中部署它，儘管它是公共預覽產品。

Streaming is something that we're super excited about. This is more for event-driven real-time applications. It's just very suitable for MongoDB due to the -- most of the data is in JSON and the flexible data model makes it a very compelling play. And so I would say that, it really depends on the customer's use case. What it really does is it just enables us to go after more workloads more quickly, and it really positions us as a truly strategic supplier to large enterprises and obviously, a critical supplier for early-stage companies, and that's our strategy is to get customers to use MongoDB for a variety of use cases across a variety of deployment models.

流媒體是我們非常興奮的事情。這更適合事件驅動的實時應用程序。它非常適合 MongoDB，因為大多數數據都是 JSON 格式，靈活的數據模型使其成為一個非常引人注目的遊戲。所以我想說，這實際上取決於客戶的用例。它真正的作用是讓我們能夠更快地處理更多的工作負載，它真正使我們成為大型企業的真正戰略供應商，顯然，也是早期公司的關鍵供應商，這就是我們的戰略是吸引客戶將MongoDB 用於跨各種部署模型的各種用例。

The other thing that I'd add, which is probably implicit and comes across that, but I think it's important enough to make explicit is one thing that, that sort of, Rich, in your question, sort of the slice-by-slice view misses is the aggregate benefit of delivering the whole platform, right, in delivering a common integrated unified experience to developers so they don't have to use a bunch of point solutions. And I think that's really a key part of the strategy.

我要補充的另一件事，可能是隱含的，並且會遇到這一點，但我認為重要的是要明確的是，在你的問題中，Rich，有點像切片。視圖失誤是提供整個平台的總體好處，對吧，為開發人員提供通用的集成統一體驗，這樣他們就不必使用一堆單點解決方案。我認為這確實是該戰略的關鍵部分。

And our next question will be coming from Brent Bracelin of Piper Sandler.

我們的下一個問題將來自 Piper Sandler 的 Brent Bracelin。

This is Brent. I believe I was next there. Dev, I wanted to talk a little bit about AI. And Mongo has been at the leading powering new apps for the better part of the decade. We're all trying to figure out what this AI first world looks like, given your purview as a new app enabler, what's your sense in the next 3 to 4 years, how many of these new apps are going to layer in large language models? And what is the net result on the database?

這是布倫特。我相信我就是下一個。 Dev，我想談談人工智能。在過去十年的大部分時間裡，Mongo 一直處於領先地位，為新應用程序提供動力。我們都在試圖弄清楚這個人工智能第一世界是什麼樣子，考慮到你作為新應用程序推動者的權限，你對未來3 到4 年的感覺是什麼，有多少新應用程序將分層在大型語言模型中？數據庫的最終結果是什麼？

Yes. So Brent, thanks for your question. I firmly believe that we, as an industry, tend to overestimate the impact of a new technology in the short term and underestimate the impact in the long term. So as you may know, there's a lot of hype in the market right now, in the industry right around AI and in some of the early stage companies in the space, have the valuations to the roof. In some cases, almost -- it's hard to see how people can make money because the risk reward doesn't seem to be sized appropriately. So there's a lot of hype in the space. But I do think that AI will be a big impact for the industry and for us long term. I believe that almost every application, both new and existing, will have some AI functionality embedded into the application over the next -- in your horizon 3 to 5 years.

是的。布倫特，謝謝你的提問。我堅信，作為一個行業，我們往往會高估新技術的短期影響，而低估長期影響。如您所知，目前市場上、人工智能行業以及該領域的一些早期公司的估值已經達到頂峰。在某些情況下，幾乎很難看出人們如何賺錢，因為風險回報的大小似乎不合適。所以這個領域有很多炒作。但我確實認為人工智能將對行業和我們產生長期的重大影響。我相信，在接下來的 3 到 5 年內，幾乎每個應用程序（無論是新的還是現有的）都將在應用程序中嵌入一些人工智能功能。

And let me just remind people why we see -- where we see AI impacting our business. One, developers will become far more productive with the use of code generation and code assist tools. What that will mean is like that will lead to more applications, which means they need more databases and more data platforms. Second, developers will use things like generative AI to just build smarter and more intelligent applications. One, they don't want to use point solutions because it's clear developers want one platform to process and analyze data and metadata and vectorized data and MongoDB is that platform. So which is why we are seeing so much interest. And Vector Search, I believe, is really ultimately a feature, not a product. And essentially, it's basically enabling people to marry private data with public data to really offer a compelling experience and there's so much interest in our public preview right now.

讓我提醒人們為什麼我們會看到人工智能在哪裡影響我們的業務。第一，通過使用代碼生成和代碼輔助工具，開發人員的工作效率將大大提高。這意味著將會帶來更多的應用程序，這意味著他們需要更多的數據庫和更多的數據平台。其次，開發人員將使用生成式人工智能等技術來構建更智能的應用程序。第一，他們不想使用單點解決方案，因為很明顯，開發人員希望有一個平台來處理和分析數據、元數據和矢量化數據，而 MongoDB 就是這個平台。這就是為什麼我們看到如此多的興趣。我相信，矢量搜索最終實際上是一個功能，而不是一個產品。從本質上講，它基本上使人們能夠將私人數據與公共數據結合起來，真正提供引人入勝的體驗，而且現在人們對我們的公共預覽非常感興趣。

And as we -- as I mentioned earlier, we've continued to add many more AI customers this quarter and we think the impact will be big. But in the short term, these are still smaller workloads, but they're going to grow over time. Some of the use cases are really interesting, but the fact is that we're really well positioned because what generative AI does is really instantiate AI in front of -- in software, which means developers play a bigger role rather than data scientists, and that's where you'll really see the business impact. And I think that impact will be large over the next 3 to 5 years.

正如我之前提到的，本季度我們繼續增加更多的人工智能客戶，我們認為影響將會很大。但從短期來看，這些工作負載仍然較小，但會隨著時間的推移而增長。有些用例確實很有趣，但事實是我們確實處於有利位置，因為生成式人工智能所做的實際上是在軟件中實例化人工智能，這意味著開發人員比數據科學家發揮更大的作用，並且在那裡您將真正看到業務影響。我認為未來 3 到 5 年影響將會很大。

Super helpful color there. And a quick follow-up for Mike. 3-year annual growth rate for EA is over 20%. It slipped below 10% in Q1, and spiked now above 30% here in Q2, excluding the licensing multiyear deals. If you continue to see enterprise workload migrations happen, why can't you continue to see strength in EA?

那裡的顏色超級有用。邁克的快速跟進。 EA的3年年增長率超過20%。第一季度下滑至 10% 以下，第二季度則飆升至 30% 以上（不包括多年許可交易）。如果您繼續看到企業工作負載遷移發生，為什麼不能繼續看到 EA 的優勢？

Yes. So a couple of things. I think as Dev mentioned, relational migrations, with the destination of those, whether it's Enterprise Advanced or Atlas will depend on the customers cloud strategy and their overall approach from like an IT strategy standpoint. Certainly, some of that could benefit EA, and we've continued to see adoption and adoption of new workloads within that EA customer base. I think the key thing, at least thinking about the back half of this year, and as you start thinking about next year, it's just we have had very strong results from EA. And so we think about EA on a compare basis, I just think it's really important to keep in mind.

是的。有幾件事。我認為，正如 Dev 提到的，關係遷移及其目的地，無論是 Enterprise Advanced 還是 Atlas，都將取決於客戶的雲戰略以及從 IT 戰略角度來看的整體方法。當然，其中一些可能會讓 EA 受益，而且我們不斷看到 EA 客戶群中不斷採用新的工作負載。我認為關鍵的是，至少考慮到今年下半年，當你開始考慮明年時，我們從 EA 獲得了非常強勁的業績。因此，我們在比較的基礎上考慮 EA，我認為記住這一點非常重要。

And our next question will be coming from Jason Ader of William Blair.

我們的下一個問題將來自威廉·布萊爾的傑森·阿德爾。

Just wanted to get a sense on EA. You talked about doing a good job of existing customers adding incremental workloads. What's the main driver there? Is there something you're doing differently? Or do you think it's just maturity and customers getting more comfortable with you for more workloads?

只是想了解一下 EA。您談到了做好現有客戶增加增量工作負載的工作。那裡的主要驅動力是什麼？你有什麼不同的做法嗎？或者您認為這只是成熟和客戶對您承擔更多工作負載感到更加滿意的結果？

Yes. I think it's really a function of people really recognizing that MongoDB is truly a standard. It's a platform they can bet on to run the most mission-critical use cases and the flexibility of the deployment models means that they can start on-prem, but they can always migrate to the cloud. And so that optional -- that built-in optionality also makes going to EA that much more comforting because it's not like they'll be locked into an on-prem solution or locked into some proprietary cloud solutions. So that that's why I believe, just given our maturity given how we're really becoming a standard in so many organizations, people are much more comfortable doubling down on EA.

是的。我認為這實際上是人們真正認識到 MongoDB 確實是一個標準的結果。他們可以押注這個平台來運行最關鍵的任務用例，部署模型的靈活性意味著他們可以在本地啟動，但也可以隨時遷移到雲端。因此，這種可選性——內置的可選性也讓使用 EA 更加令人欣慰，因為他們不會被鎖定在本地解決方案或鎖定在某些專有云解決方案中。因此，這就是為什麼我相信，考慮到我們的成熟度以及我們如何真正成為許多組織的標準，人們會更加放心地在 EA 上加倍投入。

Got you. And then just to follow up on the EA question. It seems like -- I mean, I don't want to put words in your mouth, but it seems like maybe you've been a little bit surprised, at the strength of EA relative to Atlas over last year. I mean Atlas has been really strong, too, but EA has, I think, surprised you more to the upside. What does that say, I guess, about kind of on-prem versus cloud or self-managed versus fully managed. Any comments on that, Dev?

明白你了。然後繼續跟進 EA 問題。看起來——我的意思是，我不想把話放在你嘴裡，但你似乎對去年 EA 相對於 Atlas 的實力感到有點驚訝。我的意思是 Atlas 也非常強大，但我認為 EA 的優勢更讓你驚訝。我想，這對於本地部署與雲、自我管理與完全管理意味著什麼。對此有什麼評論嗎，Dev？

Yes. What I would say is it really is a reinforcement that customers still will run workloads on-prem that they still will run workload they want to manage themselves versus use a managed service like Atlas. And so -- and I think customers value choice. The customers value the ability to have different deployment models, but also value the fact that if they want to migrate from one deployment model to another, it's easiest to do so using MongoDB. So I think what we're really seeing is customers valuing choice and to run anywhere strategy, really resonating with customers.

是的。我想說的是，這確實是一種強化，客戶仍然會在本地運行工作負載，他們仍然會運行他們想要自行管理的工作負載，而不是使用 Atlas 等託管服務。所以——我認為客戶重視選擇。客戶重視擁有不同部署模型的能力，但也重視這樣一個事實：如果他們想要從一種部署模型遷移到另一種部署模型，那麼使用 MongoDB 是最簡單的。因此，我認為我們真正看到的是客戶重視選擇並在任何地方運行策略，真正與客戶產生共鳴。

And I would just add Jason, that the premise of your question is correct. We have been very surprised, obviously, pleasantly so, with the performance of EA. It's been terrific to see, but it's definitely been surprising to the upside.

我想補充一下傑森，你的問題的前提是正確的。顯然，我們對 EA 的表現感到非常驚訝和愉快。這真是太棒了，但它的好處絕對令人驚訝。

And the next question will be coming from Tyler Radke of Citi.

下一個問題將由花旗銀行的泰勒·拉德克（Tyler Radke）提出。

Yes. So Atlas revenue grew by almost $30 million quarter-over-quarter, which is the highest you've ever seen it. Certainly, that performance is better than any of the other consumption models you're seeing. It seems like the commentary at least on Atlas, consumption was pretty consistent with your expectations and still a bit below where it had been pre some of the macro challenges. So could you just kind of unpack what's driving that strength and revision back to kind of record high levels of sequential dollar adds? Is it better pricing just given some of the sales changes you made or perhaps maybe it's the new AI use cases that you talked about. If you could just help us understand that a bit better.

是的。因此，Atlas 的收入環比增長了近 3000 萬美元，這是你見過的最高收入。當然，這種性能比您看到的任何其他消費模型都要好。至少在 Atlas 上的評論似乎是，消費與您的預期非常一致，但仍略低於一些宏觀挑戰之前的水平。那麼，您能否解釋一下推動美元走強和修正回到歷史新高水平的原因是什麼？考慮到您所做的一些銷售更改，定價是否會更好，或者可能是您談到的新的人工智能用例。如果您能幫助我們更好地理解這一點就好了。

Yes. Thanks for the question, Tyler. A few things. If you're looking at an absolute dollars basis, obviously, the business is much larger. So let's start with that. Secondly, if we're looking at the sequential from Q1 to Q2, remember, Q1 has fewer days. And so that's obviously part of the dynamic, and you'll see that historically as well. And then third, we had talked about the Q1 consumption being better than planned, and therefore, the starting ARR in Q2 being better. We saw the consumption itself kind of once we had that -- once you kind of adjust for that higher starting base, broadly in line with our expectations. So slightly better, but not big upside there. And when we look at it in the back half of the year, that's what we're continuing to assume is that same trend line, obviously, seasonally adjusted based on those emerging seasonal patterns. But that's really how to kind of tie and square all the numbers.

是的。謝謝你的提問，泰勒。一些東西。如果你考慮的是絕對美元基礎，顯然，這個業務要大得多。那麼讓我們從這個開始吧。其次，如果我們查看從第一季度到第二季度的順序，請記住，第一季度的天數較少。所以這顯然是動態的一部分，你也會在歷史上看到這一點。第三，我們談到第一季度的消費好於計劃，因此第二季度的起始 ARR 更好。一旦我們有了這樣的消費，我們就看到了消費本身——一旦你對更高的起始基礎進行了調整，大致符合我們的預期。所以稍微好一點，但沒有太大的上升空間。當我們在今年下半年審視它時，我們繼續假設同樣的趨勢線，顯然是根據這些新興的季節性模式進行季節性調整的。但這確實是對所有數字進行平局和平方的方法。

Okay. That's helpful. And then a follow-up question, just in terms of the excitement, obviously, out in the industry around generative AI. But I guess I'm curious specifically how internally you're using generative AI in products like relational migrator to automate a lot of the re-architecture process? And secondly, are you seeing a greater appetite from customers to modernize kind of legacy transactional applications. And is that starting to pick up just given the excitement around Gen AI?

好的。這很有幫助。然後是一個後續問題，顯然是圍繞生成人工智能的行業的興奮程度。但我想我特別好奇的是，你們在內部如何在關係遷移器等產品中使用生成式人工智能來自動化大量的重新架構過程？其次，您是否看到客戶對現代化遺留事務應用程序的興趣更大了？鑑於 Gen AI 的興奮點，這種情況是否開始回升？

Yes. So with regards to Gen AI, I mean, we do see opportunities essentially, the reason when you migrate off using relational migrator, there's really 3 things you have to focus on. One is mapping the schema from the old relational database to the MongoDB platform, moving the data appropriately and then also rewriting some, if not all, of the application code. Historically, that last component has been the most manually intensive part of the migration, obviously, with the advance of cogeneration tools. These opportunities to automate the rewriting of the application code. I think we're still in the very early days. You'll see us continue to add new functionality to relational migrator to help again reduce the switching costs of doing so. And that's obviously an area that we're going to focus. So that's in some ways, a big opportunity for us. And Tyler, there was the second part to your question, which I...

是的。因此，關於 Gen AI，我的意思是，我們本質上確實看到了機會，因為當您使用關係遷移器進行遷移時，您確實需要關注 3 件事。一種是將模式從舊的關係數據庫映射到 MongoDB 平台，適當地移動數據，然後重寫部分（如果不是全部）應用程序代碼。從歷史上看，隨著熱電聯產工具的進步，最後一個組件顯然是遷移過程中人工最密集的部分。這些機會可以自動重寫應用程序代碼。我認為我們仍處於早期階段。您將看到我們繼續向關係遷移器添加新功能，以幫助再次降低這樣做的轉換成本。這顯然是我們要關注的一個領域。所以從某種程度上來說，這對我們來說是一個巨大的機會。泰勒，你問題的第二部分，我......

Yes. Yes, it was just around the customer appetite, like is that the frenzy around Gen AI, is that causing an acceleration in the pace in which customers want to take on these modernization projects?

是的。是的，這只是為了滿足客戶的需求，就像圍繞 Gen AI 的狂熱一樣，這是否會導致客戶想要承擔這些現代化項目的步伐加快？

Yes. So I would say that the recent MongoDB is well suited for these new modernization projects was one, obviously, the data that's trapped in these legacy platforms is incredibly important if you want to leverage that proprietary data for a competitive advantage; two, is that the performance requirements of these new modern applications require a new modern platform; and three, because it's such an iterative area where people are -- is changing so quickly, you all see the platform that's inherently flexible. So that's driving people to move to MongoDB and to more modern platforms more quickly. So unlike the old lift and shift where people are just trying to say, avoid paying the Oracle tax. Now people are being much more thoughtful about not just lifting and shifting, but modernizing and going off relational to MongoDB. And that's definitely a trend that's increasing.

是的。因此，我想說，最近的 MongoDB 非常適合這些新的現代化項目，顯然，如果您想利用專有數據獲得競爭優勢，這些遺留平台中的數據非常重要；二是這些新的現代應用程序的性能要求需要新的現代平台；第三，因為這是一個迭代領域，人們的變化如此之快，所以你們都看到了這個平臺本質上是靈活的。因此，這促使人們更快地轉向 MongoDB 和更現代的平台。因此，與人們只是想說的舊式提升和轉變不同，避免繳納甲骨文稅。現在，人們不僅在考慮提升和轉移，而且還在考慮與 MongoDB 相關的現代化和脫離。這絕對是一個不斷增長的趨勢。

And our next question will be coming from Patrick Walravens of JMP Securities.

我們的下一個問題將來自 JMP 證券的帕特里克·沃爾拉文斯 (Patrick Walravens)。

This is Owen for Pat. Congrats on the strong quarter. Just a quick one for me. What will the pricing structure for some of the new features like Vector Search and stream processing be?

這是帕特的歐文。祝賀季度表現強勁。對我來說只是一個快速的。矢量搜索和流處理等一些新功能的定價結構是什麼？

So the pricing will be a function of the -- obviously, the consumption of the back-end infrastructure that supports those new capabilities. So they'll show up as essentially more consumption of Atlas clusters or increases of clusters depending on the load of the application, that will show up and the Atlas revenue line.

因此，定價將取決於支持這些新功能的後端基礎設施的消耗。因此，它們將顯示為 Atlas 集群的更多消耗或集群的增加，具體取決於應用程序的負載，這將顯示出來以及 Atlas 收入線。

And our next question will be coming from Michael Turits of KeyBanc.

我們的下一個問題將來自 KeyBanc 的 Michael Turits。

My name got swelled here, I wasn't sure it was me. Quick one for you, Mike, and then one for Dev. Very quick. Mike, are you able to comment on the linearity in the quarter relative to those consumption growth trends and how we exited. And then, Dev, for you, you just said that you think that Vector Search is a feature, not a product. There are 2 databases deliberately out there, explicitly out there in the market. And then, you, as well as others, who don't have vector databases, including Google, with Alloy the other day, are talking about the applicability of their databases for vector embedding. So can you talk about how that's playing out as far as you can tell with customers in terms of their receptivity of looking for something besides a dedicated vector database for this? So linear question and then that one.

我的名字在這裡被誇大了，我不確定那是我。邁克，快給你一份，然後給戴夫一份。很快。邁克，您能否評論一下本季度相對於這些消費增長趨勢的線性以及我們如何退出。然後，Dev，對於你來說，你剛才說你認為矢量搜索是一個功能，而不是一個產品。市場上故意有兩個數據庫。然後，您以及其他沒有矢量數據庫的人，包括前幾天與 Alloy 合作的 Google，正在談論他們的數據庫對於矢量嵌入的適用性。那麼，您能談談客戶對除專用矢量數據庫之外的其他東西的接受程度嗎？那麼線性問題，然後是那個問題。

Sure. yes, just click in linear. I don't think there's anything particularly point out of note there.

當然。是的，只需單擊線性即可。我不認為那裡有什麼特別值得注意的地方。

And so regarding Vector Search, and I've shared this with other -- talked about this previously, Vector Search is really a reverse index. So it's like an index that's built into all databases. I believe, over time, Vector Search functionality will be built into all databases or data platforms in the future. There are some point products that are just focused solely on Vector Search. But essentially, it's a point product that still needs to be used with other technologies like MongoDB to store the metadata, the data to be able to process and analyze all that information. So developers have spoken loudly that having a unified and elegant developer experience is a key differentiator. It removes friction in how they work. It's much easier to build and innovate on one platform versus learning and supporting multiple technologies. And so my strong belief is that, ultimately, Vector Search will be embedded in many platforms and our differentiation will be a -- like it always has been a very compelling and elegant developer experience.

關於矢量搜索，我已經與其他人分享過——之前談到過，矢量搜索實際上是一個反向索引。所以它就像一個內置於所有數據庫中的索引。我相信，隨著時間的推移，矢量搜索功能將被內置到未來的所有數據庫或數據平台中。有一些重點產品只專注於矢量搜索。但本質上，它是一個單點產品，仍然需要與 MongoDB 等其他技術一起使用來存儲元數據，以便能夠處理和分析所有信息。因此，開發人員大聲疾呼，擁有統一且優雅的開發體驗是一個關鍵的差異化因素。它消除了他們工作方式中的摩擦。與學習和支持多種技術相比，在一個平台上構建和創新要容易得多。因此，我堅信，矢量搜索最終將嵌入到許多平台中，而我們的差異化將是——就像它一直以來一樣，是一種非常引人注目和優雅的開發人員體驗。

Our next question will be coming from Mike Cikos of Needham.

我們的下一個問題將來自尼達姆的邁克·西科斯。

I'm sorry. I apologize, the operator tuned out when you mentioned me. If I could just follow up on Dev, your comments there in response to Michael on the Vector Search. I know that we're talking about the developers and how they -- they're voting here because they want the data in a unified platform, a unified database that preserves all that metadata, right? But I would think there's probably also a benefit to having it all in a single platform as well just because you're lowering the TCO for your customers as well, right? They're not paying a tax for the movement or duplication of all that data between different vendors. Is that also a fair assumption when I'm thinking about the potential that you guys bring versus maybe some of those more point features or databases out there?

對不起。抱歉，當你提到我時，接線員沒有接聽。如果我可以跟進 Dev，您對 Michael 關於矢量搜索的評論。我知道我們正在談論開發人員以及他們如何在這裡投票，因為他們希望數據位於一個統一的平台、一個保留所有元數據的統一數據庫中，對嗎？但我認為將所有內容都集中在一個平台中也可能有好處，因為您也可以降低客戶的總體擁有成本，對嗎？他們不會為不同供應商之間所有數據的移動或複制繳稅。當我考慮你們帶來的潛力以及可能存在的一些更重要的功能或數據庫時，這也是一個公平的假設嗎？

Well, with vectors, vectors are really a mathematical representation of different types of data, so there is not a ton of data, unlike application search, where there's a profound benefits by storing everything on one platform versus having an operational database and a search database and some glue to keep the data in sync. That's not as much the case with Vector because you're talking about storing essentially an elegant index. And so it's more about the user experience and the development workflow that really matters. And what we believe is that offering the same taxonomy in the same way they know how to use MongoDB to also be able to enable Vector Search functionality is a much more compelling differentiation than a developer have to bolt on a separate vector solution and have to provision, configure and manage that solution along with all the other things they have to do.

嗯，對於向量來說，向量實際上是不同類型數據的數學表示，因此與應用程序搜索不同，沒有大量數據，將所有內容存儲在一個平台上與擁有操作數據庫和搜索數據庫相比具有深遠的好處以及一些保持數據同步的膠水。 Vector 的情況並非如此，因為您談論的是本質上存儲一個優雅的索引。因此，真正重要的是用戶體驗和開發工作流程。我們相信，與開發人員必須安裝單獨的向量解決方案並必須配置相比，以他們知道如何使用 MongoDB 的相同方式提供相同的分類法來啟用向量搜索功能更具吸引力。 、配置和管理該解決方案以及他們必須做的所有其他事情。

Got it. And then just a quick follow-up for Michael. Michael, I've gotten a couple of inbounds just trying to unpack the Q3 revenue guide here, specifically as it pertains to Atlas. And I think what people are looking at is, I think Atlas embed slower sequential growth on a daily consumption basis. The questions I'm getting are, was there anything onetime 2Q that would cause us to think that the daily Atlas consumption growth would decelerate going into 3Q? Or anything else you can tease out there while we have you?

知道了。然後是邁克爾的快速跟進。邁克爾，我收到了一些消息，只是想在這裡解開第三季度的收入指南，特別是與 Atlas 相關的內容。我認為人們關注的是，我認為 Atlas 在日常消費的基礎上嵌入了較慢的連續增長。我收到的問題是，第二季度是否有任何事情會讓我們認為每日 Atlas 消費增長會在進入第三季度時減速？或者當我們有你的時候你可以在那裡取笑什麼？

No. I think if you look at what we said is the real underlying reason for the increase, obviously, we don't guide by product, but we are trying to give you a much of color around it, the real increase in the Atlas number is given the slight outperformance in Q2, we have a starter hiring ARR in Q2, and that's at the beginning of Q3. And that's really what's flowing through in the numbers and in the guidance.

不。我認為，如果你看看我們所說的增長的真正根本原因，顯然，我們不按產品進行指導，但我們試圖為你提供圍繞它的大量色彩，即 Atlas 中的真正增長考慮到第二季度的表現略有出色，我們在第二季度開始招聘ARR，那是在第三季度初。這確實是數字和指導中所體現的內容。

Thank you. This concludes the Q&A session for today. And I would now like to turn the call back over to Dev Ittycheria, CEO, for closing remarks. Please go ahead.

謝謝。今天的問答環節到此結束。現在我想將電話轉回給首席執行官 Dev Ittycheria，讓其發表結束語。請繼續。

Thank you, everyone, for joining us today. I just want to reinforce that we had another strong quarter of new business performance, which really validates our value proposition and our run anywhere strategy, Again, we remain focused on our North Star, which is acquiring new workloads, both from new workloads -- the new customers and existing customers. And we're innovating both on the product and go-to-market dimensions to accelerate workload acquisition. And while it's early days, we believe that with the rise of AI, MongoDB will be beneficiary of -- as AI becomes more prominent. Thank you very much, and I appreciate all your time. Take care.

謝謝大家今天加入我們。我只是想強調，我們的新業務業績又一個強勁的季度，這確實驗證了我們的價值主張和我們的隨處運行戰略。同樣，我們仍然專注於我們的北極星，它正在從新的工作負載中獲取新的工作負載 -新客戶和現有客戶。我們正在產品和上市層面進行創新，以加速工作負載的獲取。雖然現在還處於早期階段，但我們相信，隨著人工智能的興起，MongoDB 將受益於人工智能變得更加突出。非常感謝您，也感謝您花費所有時間。小心。

This concludes today's conference call. Thank you all for participating and enjoy the rest of your evening. You may now disconnect.

今天的電話會議到此結束。感謝大家的參與並祝您度過愉快的夜晚。您現在可以斷開連接。