MongoDB Confirms Multiple Frontier AI Labs as Customers While Atlas Hits $2 Billion Run Rate
Q1 Fiscal 2027 Earnings Call, May 28, 2026 — Revenue beats high end of guidance, full-year outlook raised by 200 basis points
The most consequential disclosure on MongoDB's first quarter fiscal 2027 earnings call was not the revenue beat. It was CEO CJ Desai's deliberate and carefully chosen word: "labs," plural. When asked by Scotiabank's Patrick Colville whether MongoDB was working with multiple frontier AI laboratories, Desai confirmed it without hesitation. "Short answer to your first question, yes, it is plural, and it was chosen carefully." Multiple frontier labs have selected MongoDB for what Desai described as mission-critical workloads that are "among the most demanding data workloads in the industry," with use cases varying by lab. The depth of engagement is still early and deal sizes remain unspecified, but the directional signal — that organizations at the absolute frontier of AI infrastructure are choosing MongoDB over alternatives including Postgres — is significant for the long-term thesis.
Atlas Growth Accelerates for the Fourth Consecutive Quarter
On the financial headline, MongoDB delivered total revenue of $688 million in Q1, up 25% year-over-year, beating the high end of guidance and accelerating from the 22% growth posted in the same quarter of each of the prior two fiscal years. Atlas, the cloud database service, grew 29.4% year-over-year and added a record $117 million in year-over-year dollar growth, its fifth consecutive quarter of absolute dollar expansion. Atlas now sits at a $2 billion annualized run rate and accounts for approximately 75% of total revenue. CFO Mike Berry confirmed guidance for Q2 Atlas growth of approximately 26%, and raised the full-year Atlas growth expectation by 200 basis points to a range of 23% to 25%. Non-GAAP operating margin came in at 18%, above the high end of guidance, and the company achieved its second consecutive quarter of GAAP profitability.
The Agentic Memory Layer Opportunity is Real, Not Yet Material
The most strategically interesting product narrative on the call centered on MongoDB's emerging role as the memory and reasoning layer for AI agents themselves — a use case distinct from simply storing operational data. Desai cited Adobe's Journey Agent as a live production example: "A composite multimodal AI agent that unifies Adobe's marketing suite and orchestrates end-to-end customer journeys for their global B2C user base with MongoDB as the agent's long-term memory and reasoning layer." Adobe uses Atlas Search and Atlas Vector Search together to deliver sub-100 millisecond hybrid search for real-time agent decision-making.
Desai was direct about why MongoDB's architecture suits this use case in a way relational databases structurally cannot. "Agents don't behave like traditional applications. They read, write and act continuously across multiple simultaneous threads with a single agent spawning subagents that each make independent reads and writes in real time. Analytical systems built for off-line processing weren't designed for this." He added a pointed observation about the origins of MongoDB's architectural advantage: "We didn't have AI workloads in mind, but this architecture is perfectly suited for AI workloads." MongoDB 8.3, released during the quarter, delivers up to 45% more reads, 35% more writes and 15% more ACID transactions over version 8.0 without requiring any application code changes.
Investors should note Desai's honesty about the current financial contribution: "Our results today are driven primarily by core workloads, but we are seeing real and growing momentum from AI and agentic workloads." When asked by Goldman Sachs analyst Matt Martino whether agentic workloads are approaching the point where they move the needle on consumption, Desai said flatly, "It's still early." The production examples are real, but the revenue scale from AI is still emerging rather than transformative in the current period.
AI Natives Validating the Platform at Scale
MongoDB is accumulating a growing set of credible AI-native customers who chose the platform under stress. Desai recounted the ElevenLabs case — now at $500 million in ARR — whose engineering team had been running separate databases for operational data and search before consolidating onto Atlas. "They said, gee, we should have done this a lot sooner. Otherwise, we would have not had to deal with all these outages." Endor Labs, an AI-native application security platform protecting over 7 million applications, selected Atlas to support 225% year-over-year revenue growth. Zomato built Nugget, an AI-native customer support platform now orchestrating 15 million conversations per month on Atlas, after evaluating and rejecting DynamoDB and DocumentDB. Nugget has reduced support costs by 55% and improved human agent productivity by 40%, and Zomato is now selling the platform to other enterprises.
The go-to-market model for AI natives remains a work in progress. Desai acknowledged that many of these companies enter through self-serve and the company is still determining the right intervention point for field sales coverage. "With Ryan now in place, we are figuring out what is the right point to intervene, and that is a work in progress." New CRO Ryan Mac Ban joins from Confluent, where he led a cloud-native consumption-oriented business. The parallel is instructive given Confluent's experience scaling enterprise consumption models.
Enterprise Advanced Defies Obsolescence Narrative
EA and other revenue, which covers on-premise and hybrid deployments, grew 13% year-over-year in Q1 and Berry guided for approximately 20% growth in Q2, driven by the timing of several large multiyear deals with existing customers. The company is raising its full-year EA expectation to mid-single-digit growth, though Berry was explicit that EA revenue will be approximately flat in the second half due to tough comparables from the prior year's strong Q4. The durability of EA is notable given the widespread assumption that on-premise workloads were in secular decline. Desai attributed this to customers running both Atlas and EA simultaneously, particularly in financial services and technology, where regulatory, latency and cost-at-scale considerations are pushing back against the assumption that every workload migrates to public cloud.
Federal Vertical Strategy Takes Shape with Clarity Acquisition
MongoDB disclosed the acquisition of Clarity Business Solutions, a federal services firm that has been a partner since 2021 with high-level security clearances for classified government workloads. Financially, Clarity represents approximately $10 million in annual services revenue at roughly breakeven profitability — small enough to be immaterial today but strategically positioned ahead of MongoDB's expected FedRAMP High certification later this year. Berry noted that the federal business is "a pretty small piece of our business today" but that the company wants to address civilian, intelligence and defense segments comprehensively. Desai added that many federal customers are currently running the community version of MongoDB and would upgrade once proper enterprise coverage is in place. The long-term argument centers on unstructured document data at government agencies requiring high-performance retrieval — a description that fits MongoDB's native strengths.
Platform Depth Deepening Among Large Customers
One metric that deserves more attention from investors is platform adoption depth. Of Atlas customers generating at least $100,000 in ARR, 45% are now leveraging two or more platform features, up from 37% a year ago, driven largely by Vector Search and text search adoption. The net ARR expansion rate for the total company improved to 121% from 119% a year ago, with Atlas running above the company average. The cohort of customers above $100,000 in ARR grew 16% year-over-year to 2,895, and revenue growth from that cohort outpaced total company revenue growth. The company added 2,500 net new customers in Q1, bringing the total to 67,700.
Guidance and Capital Allocation
For Q2, MongoDB guided revenue of $729 million to $734 million, representing 23% to 24% growth, with non-GAAP operating margin of approximately 21% at the high end. For the full fiscal year, the company guided revenue of $2.92 billion to $2.96 billion, representing 19% to 20% growth, with non-GAAP operating margin of approximately 20% at the high end — putting MongoDB on track for a Rule of 40 performance at the top of its range. Full-year non-GAAP EPS is guided at $5.95 to $6.14. Operating cash flow in Q1 was $202 million versus $110 million a year ago, and free cash flow was $198 million versus $106 million. The company ended the quarter with $2.4 billion in cash and short-term investments after deploying $100 million in share repurchases during the quarter. An Investor Day is scheduled for September 29 in New York City.