MongoDB Says Enterprise Advanced Has Quietly Become a Double-Digit Growth Engine, Debunking Atlas Cannibalization Fears
Piper Sandler 5th Annual Growth Frontiers Conference, September 15, 2026
MongoDB's CFO Michael Berry and Chief Product Officer Benjamin Cefalo used a fireside chat at Piper Sandler's Growth Frontiers Conference to reframe how investors should think about the company's business mix, just two weeks ahead of its Analyst Day. The headline message: Enterprise Advanced, the self-managed license business long treated as a legacy afterthought next to Atlas, has become a genuine second growth engine, and it is not cannibalizing the cloud database business the way the market has assumed.
EA Emerges as a Real Growth Driver, Not a Legacy Drag
Berry told the audience that EA annual recurring revenue has accelerated for three consecutive quarters and is "now turning into a double-digit growth driver." He attributed the reacceleration to a renewed enterprise appetite for self-managed deployments over the past 12 to 18 months, driven by regulated industries and governments that need to run workloads on-premise or across specific clouds due to data sovereignty rules, cloud capacity constraints, and the sheer economics of absorbing incremental AI spend. "Now you bring all the additional regulations, you bring AI into it. Can they even deploy in a cloud given the capacity constraints? Economics matter," Berry said. Cefalo pushed back explicitly on the idea that customers view EA as a retreat to legacy technology, noting that MongoDB brought search and vector capabilities—originally built for Atlas—into the self-managed product only about four to five months ago, driven entirely by customer demand rather than a defensive bet. "Customers like the flexibility and portability that we're able to give them that, quite frankly, not a lot of other technologies can give them," he said, pointing to use cases where customers operate in regions without a cloud provider at all.
No Evidence of Atlas Cannibalization
The more consequential data point for investors: among the cohort of large customers running both self-managed and Atlas, Atlas growth is actually higher than the total company average. Berry called out this statistic as a direct rebuttal to a persistent market narrative that EA growth comes at Atlas's expense. "There's this perception, I think, Rob, which is, hey, it cannibalizes it. Our view is that simply hasn't been the case," he said, adding that EA and Atlas share the same underlying codebase, so incremental R&D spend to serve both is minimal. Berry also pushed back on the idea that MongoDB's AI monetization story hinges solely on Atlas acceleration, arguing that customers are agnostic about where they run AI-driven workloads as long as they get the functionality. "They're saying we want to deploy run anywhere. We might want to self-manage that," he said, suggesting self-managed could become a larger share of AI-driven consumption than investors currently model.
cRPO Spike Explained, and a Caution Against Overreading It
Berry addressed the sharp jump in current remaining performance obligations investors flagged last quarter, cautioning against using cRPO as a primary revenue leading indicator for a hybrid business like MongoDB's. He explained that the spike reflects a shift in go-to-market motion as the company moves upmarket into larger enterprise accounts that prefer longer-term commitments, across both self-managed and Atlas. With total RPO around $1.4 billion and current RPO around $700 million against a roughly $3 billion revenue guide, Berry noted the metric captures only commitments above one year and includes renewals of existing customers, not just incremental new business. "I would also ask investors, don't assume that's all incremental because it's not," he said, characterizing cRPO as "a part of" the growth picture rather than the biggest driver.
Margin Expansion Mechanics and the AI Reinvestment Flywheel
Berry walked through the operating discipline behind recent margin gains—operating margin up 5 points and free cash flow margin up 15 points—attributing it primarily to scale, alongside a fiscal 2026 sales restructuring that trimmed capacity that wasn't driving productivity. Going forward, he guided to R&D dollar growth of roughly 30%, sales and marketing growth in the mid-teens, and low-single-digit G&A growth, with R&D growing at a lower percentage rate than prior years but still expanding in absolute dollars. Notably, Berry disclosed that MongoDB has not yet meaningfully deployed AI internally to drive its own efficiency gains, framing this as an underappreciated lever. "We've not really benefited from AI internally, and we've only started to roll it out. Our goal is, hey, we want to reallocate those dollars to drive growth," he said, specifically flagging customer support and business development functions as areas ripe for automation. He argued AI-driven internal automation is structurally different from prior efficiency waves because it reduces dependence on people-centric processes, which should let software companies sustain higher margins alongside growth rather than trading one for the other.
Voyage Acquisition Is Driving the Customer Acquisition Inflection
Berry attributed the recent inflection in net new customer additions largely to the Voyage embeddings acquisition, closed a little over a year ago, rather than core Atlas or EA sales motions. He was direct that Voyage's customer contribution has not yet meaningfully shown up in revenue and that the next phase of value creation depends on converting those Voyage customers into full Atlas customers. This is a useful data point for investors trying to disaggregate organic core growth from acquisition-driven customer count inflation.
Product Roadmap: MCP, Agent Memory, and the Data Platform Pitch
Cefalo detailed where incremental R&D dollars are being directed, starting with MongoDB's MCP server, launched roughly four weeks before the conference, which lets AI agents interface with MongoDB services in a standardized way. The company is embedding this into connector libraries across Frontier Labs and Vercel's v0 platform to drive what Cefalo called "unaided AI awareness"—getting coding agents to default to MongoDB without a human developer explicitly choosing it. He also described an organic use case emerging among customers: using MongoDB as a memory layer for AI agents, leveraging its native JSON storage since agents communicate in JSON natively. "This happened organically with some customers where they were like, hey, this is the best place to stick this," he said, noting a partnership with LangChain to formalize this within agentic frameworks. On the competitive top-of-funnel problem versus Postgres, Cefalo acknowledged the battle is partly about influencing how large language models get trained, which he called "a slow game" given the lag between documentation updates and new model training cycles.
Leadership Transition Under New CEO CJ
Asked about the shift since CJ took over as CEO from Dev, Berry credited the new leadership with three changes: a sharper customer-facing focus, with CJ spending the majority of his time directly with customers; a heightened push into enterprise and C-suite selling, an area CJ has publicly said the company should have been further along on; and greater operating rigor, including monthly and quarterly operating reviews focused on return on investment rather than backward-looking reporting. "Our focus is what are we seeing as leading indicators to allow us to change," Berry said, contrasting this with a prior tendency to manage by rearview mirror.