MongoDB (MDB) Blitzscaling Score: 65/100 (B) — Moat, Speed & Risk

Growth re-accelerated to 30%, the company turned GAAP-profitable, and Atlas now carries almost three-quarters of revenue. The score still lands exactly on the line — because the evidence pack was thin, and the model scores what it can verify.

9 min read · Blitz Index v1.0 · 2026Q4 · T = August 31, 2026 · Not investment advice

The business in one paragraph

MongoDB sells a document database for the cloud era. Its managed cloud service, Atlas, is now the company: in the second quarter of fiscal 2027 Atlas revenue was $565.9 million, up 29% year over year — 73% of total quarterly revenue of $771.8 million. Total revenue grew 30% year over year, the fastest quarterly growth since fiscal 2024, and the older Enterprise Advanced product grew 36%, its best in three years. Remaining performance obligation reached $1.52 billion, up 91% year over year, and full-year fiscal 2027 guidance was raised to $2.99–3.03 billion. Trailing-twelve-month revenue is roughly $2.8 billion. The quarter also marked a structural milestone: GAAP operating income of $28.4 million, against a $65.3 million loss a year earlier.

On those numbers alone, a reader might expect a score well clear of the threshold. It is not. MongoDB lands at 65 — exactly on the line, alongside Palantir and Salesforce in the same quarter. The reason is not the business. It is the evidence.

Why it scored 65: three drivers

1. Exactly on the threshold — a mature SaaS company on the borderline. The 2026Q4 theme is that good companies are not necessarily current blitzscalers, and MongoDB joins it. The estimated history makes the point: 66 in 2023, exactly 65 in 2024, 63 in 2025, and now a measured 65. The business did not collapse in 2025 — growth simply slowed to 19% that year as the $2 billion revenue base got heavier. The score tracks speed, and speed is what moves.

2. Re-acceleration plus a return to profit — but a conservative speed score. The second quarter of fiscal 2027 was genuinely strong: 30% growth, the best since fiscal 2024, and a GAAP profit turn. Growth (A1) earns 8 of 10 — the score file judges 30% at a $2.8 billion scale as the equivalent of Snowflake's 35% at $5.4 billion. But the Speed axis totals only 15 of 25, because the two items the pack could not verify were scored conservatively. Capital velocity (A2) earns 2: no market-capitalisation trend or fundraising-speed figures were available, so the conservative floor applies. Hiring and productivity (A3) earns 2 for the same reason — headcount and revenue-per-employee figures were unconfirmed. The score file is explicit: had the pack been thicker, Speed could have been 3–4 points higher. Share velocity (A4) earns 3: MongoDB leads the document-database category, but with AWS DocumentDB, Azure Cosmos DB and Oracle competing, number-one status cannot be asserted outright — though the customer evidence is strong, with 2,999 customers at $100,000-plus in annual recurring revenue, up 17% year over year, and a record 2,900 net new customers in the quarter.

3. The consumption-based transition is complete. Atlas at 73% of revenue, growing about 29% for five consecutive quarters, with net revenue expansion of 122% and RPO up 91%, is the portrait of a cloud database-as-a-service franchise that has finished its transition. The Moat axis (38 of 60) reflects that maturity: network effects (B1) earn 8, built on a developer community of millions, MongoDB University, and data gravity — once an application is embedded on MongoDB, switching is expensive — with 48% of Atlas customers above $100,000 in ARR now using two or more platform features, up from 42% a year earlier. Distribution (B2) earns 9 on the dual structure of Atlas self-serve product-led growth plus enterprise direct sales; the record quarterly customer additions are the product-led evidence, though enterprise sales still dominate enough to keep the score at the bottom of the higher band. Market size (B3) earns 10: a database software market above $100 billion is more than 30 times current revenue, and AI workloads — Vector Search and Voyage AI — are expanding it. Gross margin (B4) earns 11 on a 74% GAAP gross margin, improved from 71%, and 76% non-GAAP. Creative destruction (C) is the strongest axis at 12 of 15 — 4 on each item: relational databases are being displaced by document-model cloud databases, the schema-less developer workflow has become standard and hard to reverse, and Atlas's pay-for-what-you-use pricing has replaced perpetual licences and self-hosted infrastructure spending.

The scorecard

Total: 15 + 38 + 12 − 0 = 65, Grade B. No FLAG — demerits are zero, far below the 15-point FLAG line.

The gates

Both gates pass comfortably, which is part of why the threshold score reads as a measurement artefact rather than a verdict on the business. G1 — growth continues or a second growth curve exists — passes on the re-acceleration itself: 30% growth, Atlas at 29% for five straight quarters, Enterprise Advanced at 36%, RPO up 91%, and guidance raised. G2 — the new market's ten-year ceiling must be at least ten times current revenue, a bar of roughly $28 billion — passes with room to spare: the global database software market is roughly $100 billion and expanding with AI workloads, more than 30 times trailing revenue.

The other side: what to worry about

No demerits fired, so the concerns are again observational — but they are real. The first is the gap between GAAP and non-GAAP profit: net income of $40.9 million against non-GAAP income of $162.6 million, a difference driven by stock-based compensation, which the market has repeatedly flagged. Non-GAAP operating margin is 24%, up from 15%, and free cash flow doubled to $137.6 million — the underlying economics are improving, but the compensation load is the price of the talent that builds the product, and it is large.

The second is the competitor that is also the channel: the hyperscalers. AWS DocumentDB and Azure Cosmos DB compete directly with Atlas while also hosting it, and the score file notes that MongoDB pays infrastructure costs to the hyperscalers — one reason the gross-margin ceiling is capped at 11 rather than 12. The third is the consumption model's sensitivity to the cycle: when customers optimise cloud spending, a usage-based revenue line bends immediately. And the fourth is the AI question itself: Vector Search adoption is expanding and Voyage AI's customer count has doubled for two consecutive quarters, but management itself concedes that AI revenue contribution is still "small". The second curve is being verified, not yet proven.

A note on confidence

This score carries medium confidence, and the reason deserves to be stated plainly rather than buried. The evidence pack was thin: hard numbers existed essentially for growth alone, and the rest was supplemented from secondary web sources. The model answers that with a rule — unconfirmed means scored low — so A2 and A3 both sit at the conservative floor of 2. That is the honest way to handle a thin pack, and it is also why this 65 should be read differently from a 65 built on a thick pack. The next quarter's task is verification: capital speed and productivity, confirmed or not, will move this score by more than any change in the business itself. The score file lists exactly that as the open item, and proposes a v1.1 rule to fix the default score when A2 cannot be verified, plus a three-step confidence scale — high, medium, low — because two steps cannot express "a listed company with a thin pack".

What would change the score

Up: the cheapest upside is evidence, not performance — confirming capital velocity and productivity figures could add 3–4 points to Speed on its own. Beyond that, sustained 30%-plus growth would firm up A1, and AI workloads turning from "small" into a measured revenue line would strengthen the second-curve case behind G1 and the C axis. If multi-product lock-in keeps deepening — the share of large Atlas customers using two or more features is already rising — B1 has room to rise.

Down: growth falling back toward the high teens, as in fiscal 2025, would push A1 down and the total below the line. A cloud-spending optimisation cycle would hit the consumption model first. And if the hyperscalers' competing document databases keep pace while Atlas's infrastructure costs stay high, the margin ceiling that already caps B4 at 11 could start to bite. MongoDB's score is a borderline score with a strong quarter behind it; the next quarter decides whether the re-acceleration was a turn or a blip.