Q3 2025 Earnings Call Transcript

Good day, and thank you for standing by. Welcome to the MongoDB Third Quarter Fiscal Year 2025 Conference Call. At this time, all participants are in a listen-only mode. After the speakers' presentation, there'll be a question-and-answer session. [Operator Instructions] Please be advised that today’s conference is being recorded.

I would now like to turn the call over to your speaker for today, Brian Denyeau. Please go ahead.

Thank you, Lisa. Good afternoon, and thank you all for joining us today to review MongoDB's third quarter fiscal 2025 financial results, which we announced in our press release issued after the close of the market today. Joining me the call today are Dev Ittycheria, President and CEO of MongoDB; and Michael Gordon, MongoDB's COO and CFO. During this call, we will make forward-looking statements, including statements related to our market and future growth opportunities, our expectations for the macroeconomic environment in fiscal 2025 and the impact of AI, the benefits of our product platform, our competitive landscape, customer behaviors, our financial guidance and our planned investments in growth opportunities in AI.

These statements are subject to a variety of risks and uncertainties, including the results of operations and financial condition that cause actual results to differ materially from our expectations. 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 July 31st, 2024 that we filed with the SEC on August 30th, 2024. 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 on this conference call. Please refer to the tables in our earnings release in the Investor Relations portion of our website for a reconciliation of these measures to the most directly comparable GAAP financial measure. With that, I'd like to turn the call over to Dev. Dev?

Thanks, Brian, and thank you to everyone for joining us today. I'm pleased to report that we had a strong quarter of new business and executed well against our large market opportunity. Let's begin by reviewing our third quarter results before giving you a broader company update. We generated revenue of $529 million, a 22% year-over-year increase and above the high end of our guidance.

Atlas revenue grew 26% year-over-year, representing 68% of total revenue. We generated non-GAAP operating income of $101 million for a 19% non-GAAP operating margin and we ended the quarter with over 52,600 customers. Overall, we were pleased with our performance in the third quarter. We had a strong new business quarter and we're happy with our new workload acquisition on Atlas.

Our non-Atlas business significantly exceeded expectations in part because we benefited from a few large multi-year deals as customers continue to value our run anywhere strategy and want to build a deeper longer-term relationship with MongoDB. Atlas consumption was slightly better than expected in a macro-environment that we would characterize as largely consistent with what we saw in the first half of the year. Michael will cover consumption trends in more detail. Retention rates remained strong in Q3, demonstrating the mission criticality of our platform.

On our Q1 earnings call, we shared with you the three major strategic initiatives that we believe will enable us to maximize our long-term opportunity. I want to give you an update on the progress we're making on those initiatives. First, we are increasing our investment in the enterprise channel since we see the strongest returns in this part of the market. Specifically, we're expanding our strategic account program going to next year, as we see more accounts that will benefit from incremental investment.

In addition, we're investing time and resources to educate developers in large enterprise accounts and uplevel their MongoDB skills. These organizations have thousands of developers and as we penetrate them more deeply, we encounter developers who have historically only built SQL applications and simply do not know how to use MongoDB to its full potential. In our experience, educating these developers on the benefits of MongoDB drives significant incremental adoption of our platform. To fund these up-market investments, we are reallocating a portion of our mid-market investments.

The mid-market remains an attractive opportunity for us, but we believe that prioritizing investment up-market will deliver strong returns in the current environment. We also believe there are additional ways to serve the mid-market more efficiently through our self-serve channel and other scaled technology-enabled sales and customer service motions. Second, we are optimistic about the opportunity to accelerate legacy app modernization using AI and are investing more in this area. As you recall, we ran a few successful pilots earlier this year, demonstrating that AI tooling combined with professional services and our relational migrator product can significantly reduce the time, cost, and risk of migrating legacy applications onto MongoDB.

While it's early days, we have observed a more than 50% reduction in the cost to modernize. On the back of these strong early results, additional customer interest is exceeding our expectations. Large enterprises in every industry and geography are experiencing acute pain from their legacy infrastructure and are eager for more agile, performant, and cost-effective solutions. Not only are customers excited to engage with us, they also want to focus on some of the most important applications in their enterprise, further demonstrating the level of interest and size of the long-term opportunity.

As relational applications encompass a wide variety of database types, programming languages, versions, and other customer-specific variables, we expect modernization projects to continue to include meaningful service engagements in the short and medium term. Consequently, we're increasing our professional services delivery capabilities, both directly and through partners. In the long run, we expect to automate and simplify large parts of the modernization process. To that end, we are leveraging the learnings from early service engagement to develop new tools to accelerate future modernization efforts.

Although it's early days and scaling our legacy app modernization capabilities will take time, we have increased conviction that this motion will significantly add to our growth in the long term. Third, we are investing to capitalize on our inherent technical advantages as a key component of the emerging AI tech stack. As a reminder, MongoDB is uniquely equipped to query rich and complex data structures typical of AI applications. The ability of a database to query rich and complex data structures is crucial because AI applications often rely on highly detailed, interrelated, and nuanced data to make accurate predictions and decisions.

For example, a recommendation system doesn't just analyze a single customer's purchase but also considers their browsing history, peer group behavior, and product categories requiring a database that can query and interlink these complex data structures. In addition, MongoDB's architecture unifies source data, metadata, operational data, and vector data in an all-in-one platform, outdating the need for multiple database systems and complex back-end architectures. This enables a more compelling developer experience than any other alternative. From what we see in the AI market today, most customers are still in the experimental stage as they work to understand the effectiveness of the underlying tech stack and build early proof-of-concept applications.

However, we are seeing an increasing number of AI apps in production. Today, we have thousands of AI apps on our platform. What we don't yet see is many of these apps actually achieving meaningful product market fit and therefore significant traction. In fact, as you take a step back and look at the entire universe of AI apps, a very small percentage of them have achieved the type of scale that we commonly see with enterprise-specific applications.

We do have some AI apps that are growing quickly, including one that is already a seven-figure workload that has grown 10x since the beginning of the year. Similar to prior platform shifts as the usefulness of AI tech improves and becomes more cost-effective, we will see the emergence of many more AI apps that do nail product market fit, but it's difficult to predict when that will happen more broadly. We remain confident that we will capture our fair share of these successful AI applications as we see that our platform is popular with developers building more sophisticated AI use cases. We continue investing in our product capabilities, including enterprise-grade Atlas Vector Search functionality to build on this momentum and even better position MongoDB to capture the AI opportunity.

In addition, as previously announced, we are bringing search and vector service to our community and EA offerings, leveraging our run anywhere competitive advantage in the world of AI. Finally, we are expanding our MongoDB AI applications program or MAAP, which helps enterprise customers build and bring AI applications into production by providing them with reference architectures, integrations with leading tech providers, and coordinated services and support. Last week, we announced a new cohort of partners including McKinsey, Confluent, Capgemini, and Unstructured, as well as the collaboration with Meta to enable developers to build AI-enriched applications on MongoDB using Llama.

Next, I'd like to provide you with a brief product update. At our dot local developer conference in London in October, we announced the general availability of MongoDB 8.0, the fastest and most performant version of MongoDB ever. MongoDB 8.0 performs 20% to 60% better against common industry benchmarks compared to our prior version and is built to exceed our customers' most stringent security, resiliency, availability, and performance requirements. To best serve our customers, we regularly review and reprioritize investments in our product portfolio to ensure we're allocating our resources to products with the highest demand from our customers.

And to do that, we also deprecate products that are not showing results we desired. Consequently, we made the decision to consolidate our Atlas serverless offerings with our smallest dedicated tiers to create Atlas Flex customers, a new offering with a simpler architecture that provides the elasticity features akin to serverless. We will begin migrating effective customers to the single, simple, entry-level solution in Q4. We also decided to deprecate Atlas Device Sync and other capabilities not widely adopted in order to focus our engineering resources on the core platform. While these reprioritization decisions are not made lightly, they allow us to deliver the most value to the largest number of customers, reinforcing our commitment to being the best modern database and helping us to grow faster.

Now, I'd like to spend a few minutes reviewing the adoption trends of MongoDB across our customer base. Customers across industries and around the world are running mission-critical projects in MongoDB Atlas, leveraging the full power of our developer data platform, including Financial Times, CarGurus, and Victoria's Secret. As part of the digital transformation journey, global specialty retailer Victoria's Secret & Company migrated its e-commerce platform to MongoDB Atlas. As a fully managed platform, MongoDB Atlas allowed the company to simplify its architecture and improve performance, supporting the retailer to provide a resilient, secure, and fast web and mobile e-commerce experience for their millions of customers around the world.

Allianz, Alfamart, Swiss Post, and Paylocity are turning to MongoDB to modernize applications. Paylocity, a leading provider of cloud-based payroll and human capital management software, selected MongoDB to power proprietary application aimed at fostering employee connections and engagement. When traffic increased and the original SQL-based solution was unable to keep up with the required performance metrics, Paylocity migrated to MongoDB Atlas to take advantage of the flexible schema architecture, performance, and scalability. MongoDB costs five times less than the previous SQL database solution and the company's developers can now create an application within minutes, something that used to take weeks.

Mature companies and startups alike are using MongoDB to help deliver the next wave of AI-powered application to customers, including NerdWallet, Cisco, and Tealbook. Tealbook, a supplier intelligence platform, migrated from Postgres, PG Vector, and Elastic Search to MongoDB to eliminate technical debt and consolidate their tech stack. The company experienced workload isolation and scalability issues in PG Vector and were concerned with the search index inconsistencies, which were all resolved with the migration to MongoDB. With Atlas Vector search and dedicated search nodes, Tealbook has realized improved cost-efficiency and increased scalability for the supplier data platform, an application that uses GenAI to collect, verify, and enrich supplier data across various sources.

In summary, we had a healthy Q3 with both Atlas and EA exceeding expectations. We saw a strong new business quarter and we remain confident in our ability to become an increasing strategic provider in our large and growing market. Looking forward, we see a great opportunity to grow our adoption in the enterprise through new workloads, modernizing legacy applications, and winning the next generation of AI-powered applications. I would like to finish by providing an update on our senior leadership.

First, as we announced early in the press release, after nearly 10 years, Michael Gordon has made the decision to leave MongoDB. Michael has been instrumental in MongoDB's success over the past decade, leading our successful IPO, helping us grow our revenue nearly 50-fold, and scaling -- and successfully scaling our business model to generate meaningful operating leverage. He has been a trusted advisor and business partner to the Board and me over the years and also has become a personal friend. Michael is excited to take a well-deserved break.

We have commenced the search for Michael's replacement and will be evaluating both internal and external candidates. One of Michael's proudest compliments -- accomplishments has been building a world-class finance team under his leadership, and I'm confident that we will not miss a beat during this transition. Michael will continue to serve as CFO through January 31st to help us finish the fiscal year and then will transition to an Advisor to the company to ensure a seamless process. If we have not named Michael's successor by fiscal year-end, Serge Tanjga, SVP of Finance, will serve as Interim CFO, beginning on February 1st.

Second, we are promoting Cedric Pech, currently our Chief Revenue Officer to the newly created role of President Worldwide Field operations. In this new position, Cedric will oversee all our field-based customer-facing and go-to-market enablement teams, including professional services. We believe this org structure will best enable us to execute on some of the key strategic initiatives I discussed earlier, in particular, our increased focus on up-market and the app monetization opportunity. I would like to congratulate Cedric on this well-deserved promotion.

With that, let me turn the call over to Michael.

Thanks, Dev, and thanks for the kind words and our incredible partnership over the past decade. The past 10 years have been the most rewarding in my professional career and I'm extremely proud of what we've achieved together and of course, to the whole MongoDB team. With as much success as we had, I still believe that MongoDB is in the early stages of realizing its full potential as it continues to take share in one of the largest markets in software. Now, turning to the results for the quarter.

I'll begin with a detailed review of our third-quarter results and then finish with our outlook for the fourth quarter and full fiscal year 2025. First, I'll start with our third quarter results. Total revenue in the quarter was $529.4 million, up 22% year-over-year and above the high end of our guidance. Shifting to our product mix, Atlas grew 26% in the quarter compared to the previous year and now represents 68% of total revenue compared to 66% in the third quarter of fiscal 2024 and 71% last quarter. We recognized Atlas revenue primarily based on customer consumption of our platform and that consumption is closely tied to end user activity of their applications.

Let me provide some context on Atlas consumption in the quarter. In Q3, consumption was slightly ahead of our expectations. This year's Q3 seasonal improvement was more muted than in years past as expected. On a year-over-year basis, consumption growth remains below that of prior year period.

Turning to non-Atlas revenue. Non-Atlas came in significantly ahead of our expectations. As Dev mentioned, EA new business was strong and we continue to have success selling incremental workloads into our existing customer base. In addition, our Q3 non-Atlas revenue benefited from a few large multi-year deals.

As you know, due to ASC 606, we recognized the entire term license component of a multi-year contract at the start of that contract. Compared to Q3 of last year, the multi-year license component of non-Atlas revenues was over $15 million higher. Turning to customer growth. During the third quarter, we grew our customer base by approximately 1,900 customers sequentially, bringing our total customer count to over 52,600, which is up from over 46,400 in the year-ago period.

Of our total customer count, over 7,400 are direct sales customers, which compares to over 6,900 in the year-ago period. The growth in our total customer count is being driven primarily by Atlas, which had over 51,100 customers at the end of the quarter compared to over 44,900 in the year-ago period. It is 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 their first Atlas workload. Continuing on, in Q3, our net ARR expansion rate was approximately 120%.

We ended the quarter with 2,314 customers with at least $100,000 in ARR and annualized MRR, up from $1,972 in the year-ago period. Moving down the income statement, I'll be discussing our results on a non-GAAP basis unless otherwise noted. Gross profit in the third quarter was $405.7 million, representing a gross margin of 77%, which is flat versus the year-ago period. Our income from operations was $101.5 million or 19% operating margin for the third quarter compared to an 18% operating margin in the year-ago period.

The primary reason for more favorable operating income results versus guidance is our revenue outperformance, including the very high margin multi-year license revenue benefit. Net income in the third quarter was $98.1 million or $1.16 per share, based on 84.2 million diluted weighted average shares outstanding. This compares to a net income of $79.1 million or $0.96 per share on 83.7 million diluted weighted-average shares outstanding in the year-ago period. Turning to the balance sheet and cash flow, we ended the third quarter with $2.3 billion in cash, cash equivalents, short-term investments, and restricted cash.

Operating cash flow in the third quarter was $37.4 million. After taking into consideration approximately $2.9 million in capital expenditures and principal repayments of finance lease liabilities, free cash flow was $34.6 million in the quarter. This compares to free cash flow of $35 million in the year-ago period. In Q3, we did not incur capital expenditures to purchase IPV4 addresses as we previously expected, but we did start making those purchases in November and still expect a total outlay of $20 million to $25 million this fiscal year as we previously communicated.

I'd now like to turn to our outlook for the fourth quarter and full fiscal year 2025. For the fourth quarter, we expect revenue to be in the range of $515 million to $519 million. We expect non-GAAP income from operations to be in the range of $55 million to $58 million and non-GAAP net income per share to be in the range of $0.62 to $0.65 based on 84.9 million estimated diluted weighted-average shares outstanding. For the full-fiscal year 2025, we expect revenue to be in the range of $1.973 billion to $1.977 billion, non-GAAP income from operations to be in the range of $242 million to $245 million, and non-GAAP net income per share to be in the range of $3.01 to $3.03 based on 84 million estimated diluted weighted-average shares outstanding.

Note that the non-GAAP net income per share guidance for the fourth-quarter and full-fiscal year 2025 includes a non-GAAP tax provision of approximately 20%. I'll now provide some more context around our updated guidance. First, in terms of Atlas consumption, we expect to see a typical seasonal slowdown in Q4, driven by underlying application usage moderating during the holiday season. Second, since Atlas consumption remained lower on a year-over-year basis in Q3, we expect to see continued deceleration of Atlas year-over-year growth in Q4. Third, we expect to see a sequential decline in non-Atlas revenue in Q4, which is contrary to our normal pattern.

The reason for this is that we experienced a significant additional benefit from multi-year deals in Q3, which we do not expect to recur in Q4. In addition, I want to provide some incremental color on some of our recent product and -- how some of our recent product and go-to-market changes will impact the growth of our reported customer count going forward. First, as Dev explained, we are reallocating a portion of our go-to-market resources from the mid-market to the enterprise channel. As a result, we expect to see significantly fewer mid-market direct sales customer net additions and as a result, slower direct sales customer growth going forward.

We believe this reallocation of investment dollars will drive higher revenue growth over time. So, it's a trade-off that makes sense. Second, as we introduce Atlas Flex clusters in Q4 and automatically migrate customers in Q1, we expect to see a one-time negative impact to our customer count since we have approximately 4,000 serverless customers who are very low spending and we do not expect them to transition over to Flex. These customers have a negligible impact on our revenue but will impact our reported customer count.

To summarize, we're pleased with our third quarter results and especially our ability to win new business. We have a small share in one of the largest and fastest-growing markets in all of software with a number of secular tailwinds, including AI at our back. We'll continue investing judiciously and focusing on our execution to capture this long-term opportunity. With that, we'd like to open it up to questions.

Operator?

[Operator Instructions] Our first question for the day will be coming from Sanjit Singh of Morgan Stanley. Your line is open.

Thank you for taking the questions and congrats, Michael, on [an astounding] (ph) career. You had an absolutely fantastic run at MongoDB. I'm excited to see what you do next, or excited if you just take a breather. So congrats, Michael.

I guess to take the question -- to start off with the questions, when we look at what Atlas has been doing in the past two quarters, correct me if I'm wrong, but I think consumption is coming in at least modestly ahead of your expectations. Relative to what we've seen at the beginning of the year, what sort of -- is it a function of sales execution? Is it a function of the end-user activity sort of improving? What's driving at least the improvement in Atlas consumption in the past two quarters?

Yeah. So a few different things. So I think if you look at our outlook at the beginning of the year, we had indicated that we thought we would see stable Atlas growth from a consumption standpoint. What we've seen and what we've talked about is -- we've actually seen lower year-over-year growth based off of the underlying consumption.

And so -- and that's incorporated into our Q4 guide. We have seen the Q3 and Q4 be -- sorry, Q2 and Q3, excuse me, be better than our expectations, but it's still down on a year-over-year basis. And so I want to make sure that we're not sort of confusing the comparative set of year-over-year versus relative to our expectations.

Understood.

And the core of it, Sanjit, to your question is really the underlying usage of the applications.

Yeah, that makes total sense. And then, Dev, I have to ask you the AI agent question. In terms of an AI agent needing more context, it has going to have a set of tools to take its actions. What does it mean for MongoDB as an operational data store as customers start to roll out more agentic applications?

Yeah. So, just to talk about agents, I think when you think about agents, there's jobs, there's -- sorry, there's a job, there's projects and then there's task. Right now, the agents that are being rolled out are really focused on task like say something from Sierra or some other companies are rolling out agents. But you're right, what they deemed to do is to deal with being able to create a rich and complex data structures.

Now why is this important for AI is that AI models don't just look at isolated data points, but they need to understand relationships, hierarchies, and patterns within the data. They need to be able to essentially get real-time insights. For example, if you have a chatbot where someone is querying, a customer is kind of trying to get some update on the order they placed five minutes ago because they may have not gotten any confirmation, your chatbot needs to be able to deal with real-time information. You need to be able to deal with basically handling very advanced use cases, understanding like, to do things like fraud detection to understand behaviors of supply chains, you need to understand intricate data relationships.

All these things are consistent with MongoDB offers. And so we believe that at the end of the day, we are well-positioned to handle this. And the other thing that I would say is that we've embedded in a very natural way of search and vector search. So we're just not an OLTP database, we do tech search and vector search.

That's all one experience and no other platform offers that and we think we have a real advantage. And so we're integrated with the leading AI frameworks and platforms. We have enterprise-grade security and compliance and customers can run us anywhere, either on 118 cloud regions or on-prem, and that again is a huge differentiator for us.

Awesome. Appreciate the thoughts, Dev.

Thanks, Sanjit.

Thank you. One moment for the next question. And our next question will be coming from the line of Taylor -- Tyler Radke of Citi. Your line is open.

Hi, thank you very much for taking the question. And, Michael, all the best, and congratulations on 10 years. Going back to the sales execution, I mean, one of the things that you talked about earlier this year was some challenges just in terms of the recently-acquired workloads ramping. And I think a lot of those were from the past fiscal year.

So curious how the quality of workload acquisition has trended this year. And as you think about the ramp in consumption potential into next year, how does that sort of look versus this time a year ago?

Yeah. Maybe I'll just talk about what we're doing and the changes we made and then Michael can talk a little bit about the consumption trends. So, we did make some changes at the beginning of the year and we really wanted to focus on both the volume and the quality of the workloads and there were some slight adjustments that we made. We think those changes are having a reasonable positive impact.

Again, it's too early to declare victory because these workloads usually start small and grow over time, but we're really pleased with the results we're seeing so far. And -- but again, it's early days. And obviously, we'll know about the fiscal '25 workloads as we go into fiscal '26. But so far, so good. Michael, on consumption?

Yeah, just a couple of things, and thanks, Tyler. The fiscal '24 cohorts that we called out earlier, that slower growth does continue. They've been in line with our revised expectations. We made some changes that we talked about at the -- earlier in the year that should affect the fiscal '25 cohorts, but it's just too early to tell.

On those, we need a few more quarters of data before you can really see if we're -- how we're seeing those behave differently. I will say we've talked about the new business environment and our success in new business. We have been pleased with that, but that just shows kind of the initial piece and we need to see how they grow and how those cohorts evolve.

Great. Thanks. And follow-up on the EA side, you talked about the outsized strength in non-Atlas business this quarter. Maybe if you could unpack like the relative upside that was driven by simply duration versus new business.

I know you called out the duration impact year-over-year, but are -- do you feel like this was sort of a one-off or do you feel like maybe some of your bigger customers are indexing more towards EA and how does that impact the way you think about the product and introducing things like vector search and stream processing onto the on-prem product?

Yeah, thank you. So overall, we continue to find the EA product resonate with customers. It's an important part of the run anywhere strategy and we've continued to see success with that and people wanting to increase their investment in MongoDB. There's always been a multiyear component and we continue to see that.

We talked about that at the beginning of the year as to how fiscal '24 had an abnormally high amount of multi-year benefit and therefore, we were anticipating that being a headwind and we quantified that in roughly the $40 million range. What we talked about on this call earlier is we saw from a few large accounts, a surprising amount of multiyear that positively benefited Q3 at a little more than $15 million in revenue compared to what we saw Q3 a year ago. So, not as much of a headwind as we had been expecting. Obviously, with the 606 dynamics, some of these things, especially for a large deal can be kind of meaty and chunky and lumpy, which is why we try and call it out and sort of help people understand, but there's a pretty healthy kind of baseline flow, not just of EA, but also of multiyear.

And when we see spikes, we just try and call it out for you.

Yeah. I just want to add, Tyler, that we are investing in our what we call our EA business. First, we're starting by investing with search and vector search and a community product. That does a couple of things for us.

One, whenever anyone starts with MongoDB with the open-source product, they immediately get all the benefits of that complete highly integrated platform. Two, those capabilities will then migrate to EA. So, EA for us is an investment strategy. We definitely see lots of large customers who are very, very committed to running workloads on-prem.

We even see some customers want to run to run AI workloads on-prem. So, the optionality they get by using MongoDB to not just be on-prem and the cloud, but also cross-cloud is a very compelling one.

Thank you.

Thank you. One moment for the next question. And our next question will be coming from the line of Brad Reback of Stifel. Your line is open.

Great. Thanks very much. And, Michael, best of luck. It's been a great run. Dev, you started the call talking about a bunch of investments, which are great given the growth of the business. And obviously, you talked about reallocating some expenses. But net-net, should we think about this incremental investment phase next year as gating margin upside?

I think that's something that we obviously are not ready to talk about next year just now, but I would say that we -- the reason we're looking to invest and just to summarize again, going up-market on legacy app modernization where we see very large workloads potentially at play and being the ideal database for GenAI apps, which is the future as important investments to drive long-term growth. And we're quite energized by those investments and that's something that we have high conviction on.

That's great. And then on the MAAP program, are most of those workloads going to wind up in Atlas or will that be a healthy combination of EA and Atlas?

I think it's again early days. I would say -- I would probably say more on the side of Atlas and EA in the early days. I think once we introduce a search and vector search into the EA product, you'll see more of that on-prem. Obviously, people can use MongoDB for AI workloads using other technologies as well in conjunction with MongoDB for on-prem AI use cases.

But I would say you're probably going to see that happen first in Atlas.

Great. Thanks very much.

Thank you.

Thank you. And one moment for the next question. Our next question will be coming from the line of Jason Ader of William Blair. Your line is open.

Yeah, thank you. I'm not going to belabor congratulating Michael, but it has been it has been fun working with you and best of luck. The question I had is on the strength in EA. Do you think, Dev, it represents a comment on how enterprises might be rethinking or reassessing the kind of on-prem versus cloud workload placement decision?

Well, when I think about large enterprises, I think large enterprises have meaningful workloads that are still running on-prem. I think the belief that everything would go to the cloud was probably something that was really popular in the good old days of Zurp. But I think now as customers assess their investments that they already have in place, they're being much more judicious about where they run those workloads and if they think they can leverage their existing investments in their own infrastructure, then they're going to do so. Also for a bunch of other reasons like regulatory reasons, some customers are quite not moving as aggressively to the cloud.

We see that in particularly in Europe, where we see a lot of the European banks still running majority of the workloads on-prem. So it also varies by region where conversely in Asia, we're seeing people much move much more aggressively to the cloud. So I think it really depends on industry, on geography, and on the personal dynamics of what's happening in that particular account. I mean, we see some large US banks are also very committed to running things on-prem.

So it really varies. And that's why we feel really good about our run anywhere strategy because it gives customer optionality. They can build something and run on-prem. And if and when they choose to move to the cloud, it's very easy to do so with MongoDB.

All right. And then just as a follow-up also on the investments you're making in strategic sales and enterprise. Could you just get a little more specific on what those investments might be? Is it hiring a lot of new salespeople? Is it working more with SIs, investing more in SIs? Any additional detail would be helpful. Thanks.

Yeah. So just for everyone's benefit, we've identified a number of accounts, which we call strategic accounts, which we think that have high upside for us. We've seen a number of accounts that grow very quickly when we deploy the right mix of resources. Now they're all not necessarily quota-carrying resources.

They could be additional technical sales resources, additional PS resources, additional customer service resources to better service and support those accounts. We even do things like run education sessions for developers of the accounts, they’re called either hackathons or like what we call developer days or even design reviews where we'll meet with our development teams who are looking to build an application, help them think about how they would potentially use MongoDB to build that particular app. And what we find is that because many of these developers, the experience with MongoDB is quite limited, the more we can engage with them, the more we can educate them and the more we can show them how simple and easy it is.

Like for example, most customers today think like they have to use an OLTP database, then a search database, maybe a vector database, and then like a caching database. And all that is integrated in MongoDB. So all of a sudden, customers can say, wow, I can simplify my life, simplify my back-end infrastructure, build this app far more quickly and it will be much more easier to manage long-term if I do everything on MongoDB. And it's really a function of just educating them on the power of MongoDB that really opens up a lot of opportunities for us.

So that's why we're doubling down and the mix of resources is really predicated on the accounts, but it's not just quota counting resources, it's the whole suite of resources that we're bringing to the table.

Thank you.

Thank you. One moment for the next question. Our next question will be coming from the line of Andrew Nowinski of Wells Fargo. Your line is open.

Okay. Good afternoon. Thank you very much for taking the question and congrats on a nice quarter. You gave an example of a customer that migrated off Postgres and I think you said they had issues with their PG vector function.

I was wondering how long was that customer using Postgres before they decided to make a change to Mongo, meaning was this some sort of like a rebound type customer where they chose PG -- or excuse me, Postgres and it didn't work? And then how frequently are you seeing this type of transition?

Yeah. So I can't give you the specifics on how long they were using Postgres, but this is not -- this is a trend that we're seeing in our business. You have to remember Postgres is a 40-year-old technology. It's -- and they have been the beneficiary of people lifting and shifting from other types of relational databases, Oracle, SQL Server, MySQL, et cetera.

And they're an open-source database. And but as part -- because they're an open-source and relational database, they have the same inherent challenges all relational databases do. They're quite inflexible. So once you build a schema, it's very hard to change the schema.

It's hard to scale and hard to distribute data. And if you have large data volumes, you have to do weird things like, for example, resort to off-road storage for large data objects, which creates performance bottlenecks. And so again, people default to Postgres if they don't know anything better because all they know is relational and everyone is kind of moving off those other relational platforms. And that's the whole point I was saying earlier.

Once we educate developers on the flexibility of schema, how easily or horizontally scale the rich query language where you can do aggregations and do sophisticated geospatial indexes, the productivity gains by using the document model and how easy is to organize data, it's -- people are just like, wow, life is so much easier. Now I want to be clear, this is not a zero-sum game. Postgres does not have to fail for us to be successful. It's a big market and we're quite excited about the opportunity, but we do see customers moving off Postgres and coming to MongoDB.

Thank you. That was very helpful. And maybe just a quick follow-up. If we normalize the $15 million multiyear deal impact you had in Q3, would EA still be down sequentially in Q4? Thank you.

We haven't given that level of guidance, but just trying to help you understand in the context of the full-year numbers and the headwind that we talked about at the beginning of the year, just given the strength that we saw in Q3.

Got it. Thanks.

Thank you. And one moment for the next question. The next question will be coming from the line of Raimo Lenschow of Barclays. Your line is open.

Hey, thank you. And if you think about the EA strength this quarter, Michael, you'd kind of give us a little bit there. Like how should we think about renewal -- the renewal situation renewal pool coming up like or that you had in Q3, coming up in Q4, et cetera, as well and then like what does it mean in terms of upsell, cross-sell opportunity as people think about starting AI projects by self-serve, as you kind of mentioned earlier on the call?

Yeah. So I think about -- if you think about EA for Q4, it tends to be a large renewal quarter, but what we're talking about in terms of the guide is because we had such strength in multiyear, that's where we would expect to see our EA down sequentially, which is not typically our pattern, which is why we called it out. In terms of AI workloads and some of those other things, I think it's early to tell and obviously, we'll continue to evolve and assess our view when we get to the full-year guide in March. And then we'll also have an updated view on how the cohorts are behaving and sort of how multiyear played out.

But I think in terms of Q4, the comments that I made earlier hopefully will help.

Yeah. Okay, perfect. And then the -- and can you talk a little bit about like obviously, there's a debate of like which database will be the persistent layer if you do AI projects, et cetera. What do you see from the big hyperscalers in terms of working with you guys and partnerships?

We obviously just have AWS kind of summit, et-cetera. Can you speak a little bit like how your relationship with those big guys is evolving around this? And Michael, All the best in case I don’t talk to you.

Yeah. So, I'll start with the partnerships first, like our -- with AWS, as you said, they just had their re:Invent show last week. Our -- it remains very, very strong. We've closed a ton of deals this past quarter.

Some of them very, very large deals. We're doing integrations to some of the new products like Q and Bedrock and the engagement in the field has been really strong. On Azure, I think we -- as I've shared in the past, we start-off with a little bit of a slow start, but in the words of the person who runs their partner leadership, the Azure MongoDB relationship has never been stronger. We've closed a large number of deals.

We're part of what's called the Azure native IC service program and have a bunch of deep integrations with Azure, including Fabric, Power BI, Visual Studio, Semantic Kernel, and Azure OpenAI Studio. And we're also one of Azure's largest marketplace partners. And GCP does -- we have actually seen some uptick in terms of co-sales that we've done this past quarter and GCP made some comp changes where they -- that were favorable to working with MongoDB that we saw some results in the field and we're focused on closing on -- closing a handful of large deals with GCP in Q4. So in general, I would say, things are going quite well.

And then in terms of, I guess, implying your question was like the hyperscalers and are they potentially bundling things along with their AI offerings? I mean, candidly, since day one, the hyperscalers have been bundling their database offerings with every offering that they have and that's been their predominant strategy. And we've -- I think we've executed well against strategy because databases are not like a by-the-way decision. It's an important decision.

And I think the hyperscalers are seeing our performance and realize it's better to partner with us. And as I said, customers understand the importance of the data layer, especially for AI applications. And so the partnership across all three hyperscalers is strong.

Okay, perfect. Thank you.

Thanks, Raimo.

Thank you. And one moment for the next question. The next question will be coming from the line of Brad Sills of Bank of America. Your line is open.

Great. Thank you so much, and congratulations, Michael, on your next move. I wanted to ask about new workloads here on vector search, stream processing, relational migrator. Is there any one of those three that's ramping faster than maybe you expected? Just a little bit of color on how those new workload types are ramping. Thank you.

Yeah. I'll kind of give you just a rundown of some of the -- I mean, essentially you're asking about the new products, like our -- on search, we introduced a new capability called Atlas search nodes, which where you can asymmetrically scale your search nodes because if you have a search-intensive use case, you don't have to scale all your nodes because they have become quite expensive. And we've seen that this kind of groundbreaking capability really well received. The demand is quite high and because customers like they can tune the configuration to the unique needs of their search requirements.

One of the world's largest banks is using Atlas Search to provide like a Google-like search experience on payments data for massive corporate customers. So this is a customer-facing application and so performance and scalability are critical. A leading provider of AI-powered accounting software uses Atlas Search to power its invoice analytics future, which allows end users and finance teams to perform ad-hoc analysis and easily find past-due invoices and voices that contain errors. So that search -- on vector search, it -- again, and it's been our kind of our first full-year since going generally available.

And the product uptake has been actually very, very high. In Q3, we released quantization for Atlas Vector Search, which reduces the memory requirements by up to 96%, allowing us to support larger vector workloads with vastly improved price performance. For example, a multinational news organization created a GenAI powered tool designed to help producers and journalists efficiently search, summarize, and verify information from vast and varied data sources. A leading security firm is using Atlas Vector Search to fight AI fraud and a leading global media company replaced elastic search with hybrid search and vector search use case for a user recommendation engine that's built to suggest -- that's built in to suggest articles to end users.

And so that's super exciting to see as well. We're also seeing a lot of interest in our streaming product, demand is very high. We just rolled it out to another hyperscaler and customers are commenting on that the use cases of being able to embed stream processing with MongoDB makes our life so much easier. So overall, we're quite pleased with the progress we're making on the new products.

And as I said before, natively bundling all these capabilities really reduces or eliminates the need for customers to have to bolt on a bunch of different technologies to solve the same problem, saving them a lot of time, money, cost, and risk.

That's really exciting. Thanks, Dev. And then I wanted to ask a question around Cedric's appointment. Any focus that may be different here under his leadership that we should be thinking about going forward? Thank you.

No, Cedric has been our CRO for, gosh, now like I think like five, six years, and he -- I was the Interim CRO for about three quarters until he took over when we last made a change and this is really an expansion of his responsibilities. I've known Cedric for a long time. He and I have worked with at multiple different companies. I think I have a good barometer for understanding sales leadership.

There's a number of sales leaders who worked at other top-tier software companies who used to work for me or with me. And so I'm super-excited by the role Cedric is going to take and then we're also making some changes under Cedric to better align the different organizations so that we can more tightly work together on going up-market, on app monetization and positioning ourselves well to be the ideal database for GenAI apps.

Super exciting. Thanks, Dev.

Thank you.

Thanks, Brad.

Thank you. And one moment for the next question, please. Our next question will be coming from the line of Mike Cikos of Needham and Company. Your line is open.

Hey guys, thanks for taking the question here. I just wanted to come back to the consumption growth being slightly better than expectations again for the second quarter in a row now. And apologies if I missed it, but this improvement that we're seeing, is this across all vintages and geographies or is it potentially more concentrated in scope? Just trying to get a better understanding of what's seeking place out there and what's embedded in the guide.

Yeah. No, I would describe it as broad-based, Mike. And obviously, we're pleased to see it and we're continuing monitoring and slicing and dicing it in different ways. And as we have information or insights to you, we'll share it. And without trying to throw a whole bunch of cold water on our mind, it was slightly better, or a step-function changed better, but good to see.

Terrific. And maybe for a quick follow-up for Dev. I think it builds off maybe Tyler's question at the top of the Q&A, but you had cited that some customers are thinking about their workloads more holistically and even looking to run AI workloads on-prem. How much of that do you think is just a function of customers are still trying to figure out how to optimize for latency and cost or is this more a demonstration of we really are in the early phases of the exploratory phase versus going into production?

Is there any way to coarse that out or is the two not necessarily connected? Thank you.

No, I think it's kind of a little bit of both. I think you have some customers who are very committed to running a big part of the estate on-prem. So by definition, then if they're going to build an AI workload, it has to be run on-prem, which means that they also need access to GPUs and they're doing that. And then other customers are leveraging basically renting GPUs from the cloud providers and building their own AI workloads.

I do think we're in the very, very early days. They're still learning, experimenting. More-and-more apps are entering production. And as I said on the prepared remarks, we have thousands of workload -- AI workloads running on MongoDB, but a very small percentage of them have demonstrated meaningful product market fit.

And so the initial traction is kind of still small. But I think as people get more sophisticated with AI, as the AI technology matures and becomes more and more useful, I think applications will -- you'll start seeing these applications take-off. I kind of chuckled that today I see more senior leaders bragging about the chips they're using versus the apps they're building. So it just tells you that we're still in the very, very early days of this big platform shift.

Great point. Thank you again, guys.

Thanks, Mike.

Thank you. And one moment for the next question. Our next question will be coming from the line of Eric Heath of KeyBanc. Your line is open.

Hi, thanks for taking the question. Dev, Michael, it sounds like the takeaway from the call is a greater focus on EA and on enterprise. So should we structurally rethink the EA business differently and think of this more as a healthy double-digit growth business going forward for the foreseeable future? And then if I could just ask a follow-up question separate to that.

But Michael, I understand that it's still early to identify the fiscal '25 cohort of workloads, but just curious at a high-level if they look and feel like of higher quality than the fiscal '24 cohort of workflow.

Do you want to take on the first one in terms of EA...

Yeah, I mean, I would say, I mean, we are very committed to our run anywhere strategy. And as I said, we are first investing in community where for many customers is the first way they experience MongoDB. And we want them to have the full experience of integrating search and Vector search into our core product. And so they can out of the gate really start building applications.

That will then transition to building those capabilities into EA. So, we are clearly investing in the EA product. But Atlas is still a big, big part of our business and a big, big part of our growth engine and we typically launch new features on Atlas and because of the capabilities we already have, the fact it's multi-cloud makes it a very, very compelling offering for many customers.

Yeah. And I think in terms of the workloads, I do think it's early. Just as a reminder for folks, they tend to start small, although grow quickly. I think the only other thing that I can add is, we've been pretty consistent and that we've been pleased with the new business that we've done.

But we need some time to let the cohorts play out as we track them. But I think like I said, we've been happy with the new business that we were winning.

Thank you. One moment for the next question. And our next question will be coming from the line of William Power of Baird. Your line is open.

You there, William?

Sorry, yeah. Thank you. Dev, you had some encouraging comments on relational migrator. I wonder if you could just touch on what you think is driving the higher interest here.

I mean, it sounds like AI is contributing and helping, but it'd be great to get some more color there because that still feels like obviously a meaningful long-term opportunity. And then maybe the second part of the question for Dev or Michael, just be great to get any other framework around the professional services investments. Any way to kind of think about quantification and timing of that?

Yeah. So the reason we're so excited about the opportunity to go after legacy applications is that, one, it seems like there's a confluence of events happening. One is that the increasing cost and tax of supporting and managing these legacy apps are just going up enough. Second, for many customers who are in regulated industries, the regulators are calling their -- the fact that they're running on these legacy apps with systemic risk, so they can no longer kick the can down the road.

Third, also because they no longer can kick the can around, some vendors are going end-of-life, so they have to make a decision to migrate those applications to a more modern tech stack. Fourth, because GenAI is so predicated on data and to build a competitive advantage, you need to leverage your proprietary data, people want to access that data and be able to do so easily. And so that's another reason for them to want to modernize. And then you also have people who built those applications who are retiring or just no longer in the firm, so it just creates more and more risk for the companies.

Given all that, customers are incredibly interested in figuring out a way to easily and safely and securely migrate those -- off those applications. And we always could help them very easily move the data and map the schema from a relational schema to a document schema. The hardest part was essentially rewriting the application. Now with the advent of GenAI, you can now significantly reduce the time.

One, you can use GenAI to analyze existing code. Two, you can use GenAI to reverse engineer tests to test what the code does. And then three, you can use GenAI to build new code and then users test it to ensure that the new code produces the same results as the old code. And so all that time and effort is suddenly cut in a meaningful way and that's suddenly creating a lot of interest from customers saying, my goodness.

And if you're already on a relational app, moving to another relational app doesn't feel like modernization. So, the other advantage is that moving to MongoDB gives them a much more modern platform, a much more agile, flexible, performant, and scalable platform for their future needs. And that's why we're so excited. Again, it's early days.

We've run a number of pilots that have gone well. We're in the process of working with some customers now in the migration process. This will take time because these are very, very complex applications. And actually, one thing I also mentioned was that they're not just going after -- saying go after some tertiary Tier 2 or Tier 3 application, they're saying, hey, we want you to look at some of our crown jewels because these are the apps that are most painful for us.

So that's also very exciting. But again, this will take time, but we're very committed to this and we think this is going to drive -- help us drive long-term -- meaningful long-term growth.

Yeah, Will. And to the last part of your question on the professional services investment, we're really building out that capacity in order to meet the demand that we're seeing relative to the opportunity. We're calling it in particular because it has a gross margin impact because that's where that will typically show up. And then maybe the last thing and it's probably obvious, but just to sort of underscore it is the reason we're doing this though is for the ARR, right, to drive the new workloads, the additional workloads over to MongoDB as part of that migration.

And over time, as we've talked about before, we hope and expect to be able to leverage technology more and more, but at least initially and into the medium-term, there's going to be a healthy human/services component to that. Just wanted to sort of effectively telegraph that out to folks.

That's helpful. Thank you.

Thanks, Will.

Thank you. One moment for the next question. And our next question will be coming from the line of Rudy Kessinger of DA Davidson. Please go ahead.

Hey guys, thanks for squeezing me in here. I believe last quarter you said consumption growth slightly ahead of expectations. And while down slower year-over-year growth versus Q2 last year, the year-over-year growth did improve from Q1. And so, I guess I'm curious for Q3, could you make a comment in that same regard? Obviously, slower on a year-over-year basis than Q3 last year, but was it stable with year-over-year consumption growth in Q2 or better or worse?

Yeah. Rudy, thanks for the question. We haven't specifically called that out relative to Q2. We did see a lower year-over-year growth as we called out. We did see a seasonal rebound. Usually Q3 is stronger than Q2 and we talked about how that was smaller than in the prior year. So hopefully, that will help you all triangulate.

Okay. And then just a quick follow-up. I believe it was on your Q4 call back in March. At that point, Dev, you said it would be at least another year until AI applications are being deployed at scale.

It sounds like the commentary that some early large workloads, but out of the thousands, just not many that are at large scale. I guess, is your expectation now that maybe it's still at least another year until we're seeing broad AI application rollouts at scale?

Yeah. I think a lot of it's a function of the what's happening in the R&D side of AI, right? So for example, today, we don't have a very compelling model designed for our phones, right, because today the phones don't have the computing horsepower to run complex models. So, you don't see a ton of very, very successful consumer apps besides, say, ChatGPT or Claude.

So we don't -- we also don't see like hundreds of apps taking off like you saw kind of the first generation of like the Internet or the cloud era, right, or the mobile era. So like I think we're still in the early days of AI. And so while we see a lot of people building AI apps, a lot of them have kind of fairly rudimentary functionality. But I think that over time that's going to change.

In fact, I know it will change. I just can't predict when that will happen. But where we do see apps having production, we are having traction, we're seeing them grow very, very quickly and we have a lot of them on our platform. It's just very few of them are really have meaningful.

Thank you. And that concludes today's Q&A session. I would like to go ahead and turn the call back over to Dev for closing remarks. Please go ahead.

Thank you, everyone. I just wanted to say we're really pleased with our Q3 results with strong new business performance and revenue exceeding expectations both -- across both Atlas and EA. We're making the necessary investments to expand our enterprise channel where we see the largest opportunity to establish MongoDB as a standard and the strongest returns on our go-to-market investments. Looking ahead, we are encouraged by the progress we're making on both accelerating legacy app modernization with AI as well as establishing ourselves as a standard of the emerging AI tech stack for greenfield AI applications.

And last but not least, I would like to thank Michael again for his contributions over the past 10 years and wish him well. Thank you, everyone, and we'll talk to you soon.

Thank you for participating in today's conference call. You may all disconnect now.