How MongoDB Makes its Money: Revenue Breakdown (FY2025)

How Does MongoDB Make its Money?

MongoDB (NASDAQ: MDB) is a database company that generated $1.92 billion in revenue for fiscal year 2025 (ending January 31, 2025). The company earns money through two channels: Atlas — a fully managed, consumption-billed cloud database service — and Enterprise Advanced — self-managed database software sold on annual subscription licenses with enterprise support.

MongoDB’s core product is the MongoDB database: a document-oriented NoSQL database that stores data as flexible, JSON-like documents rather than the rigid rows and columns of traditional relational databases (SQL). This flexibility makes MongoDB popular with software developers building modern applications — web apps, mobile backends, real-time analytics, and increasingly AI workloads — where data schemas evolve rapidly and horizontal scalability matters.

The company’s strategic arc over the last five years has been a deliberate migration of its business from on-premise software licenses (high margin, low growth) to cloud consumption (rapidly growing, improving margins at scale). Atlas now represents 66% of total revenue and is growing at 23.5% annually, while Enterprise Advanced (34% of revenue) grows at ~6%.

Key Takeaways

• MongoDB generated $1.92B in FY2025 revenue, up 14.3% year-over-year — a deliberate deceleration from 29% growth in FY2024 as the company anniversaried a period of large enterprise deal timing
• Atlas ($1.26B, +23.5%) is the growth engine: a consumption-billed cloud database running on AWS, Azure, and GCP that now accounts for 66% of total revenue
• Gross margin of 72.4% — healthy for a database company with managed cloud infrastructure costs; Atlas margins expand as workload sizes grow
• Operating margin of -4.2% (GAAP) — near breakeven; non-GAAP operating margin is approximately +16%, with the gap almost entirely due to stock-based compensation
• Free cash flow margin of ~15% — MongoDB is already FCF-positive despite GAAP losses, consistent with the SaaS pattern where non-cash SBC suppresses GAAP income
• Net Revenue Retention (NRR) above 120% for Atlas — existing customers grow their usage significantly year-over-year, compounding revenue without proportional new customer acquisition spend
• Atlas Vector Search positions MongoDB for the AI application wave — developers building RAG (retrieval-augmented generation) and AI-powered search applications need vector database capability, and MongoDB bundles it natively into the Atlas platform

MongoDB (MDB) Business Model

MongoDB operates two fundamentally different revenue models simultaneously. Understanding both is essential to reading the financials.

Atlas: Consumption-Based Cloud Database

Atlas is MongoDB’s Database-as-a-Service (DBaaS) offering, available on all three major cloud platforms (AWS, Azure, GCP). Customers pay based on what they consume — measured in compute hours (cluster size), storage (GB), and data transfer (bandwidth). There are no seat licenses; billing scales directly with application usage.

How consumption pricing works:

• A developer creates a MongoDB Atlas cluster — selecting compute (e.g., M10, M30, M50 tiers) and storage
• The cluster runs 24/7 and Mongo bills by compute hour and GB-month
• As the application serving that cluster grows (more users, more data, more queries), the developer scales up the cluster — automatically increasing MongoDB’s revenue
• Peak-hour scaling, sharded clusters, and multi-region deployments all increase consumption and therefore revenue

Why this model is powerful: Revenue from existing customers grows organically as their applications scale — without any sales effort from MongoDB. A startup that seeds with a $200/month Atlas cluster and grows to 10M users becomes a $50,000+/month customer over time. This organic expansion is measured by Net Revenue Retention (NRR) — MongoDB’s Atlas NRR above 120% means existing customers collectively spend at least 20% more each year.

Why this model has risk: Revenue can decelerate quickly if customers optimize workloads, downscale clusters during budget tightening, or reduce development activity. MongoDB experienced this directly in FY2023 when customers began aggressively right-sizing their Atlas clusters in response to macro pressure, causing revenue to miss expectations and the stock to drop significantly. This consumption volatility is the primary near-term revenue risk for MongoDB.

For how usage-based pricing works at the business model level, see the Usage-Based Business Model.

Enterprise Advanced: Subscription Software

Enterprise Advanced is the legacy business: annual (or multi-year) subscription licenses for the MongoDB database software deployed on customers’ own infrastructure — on-premise servers or private cloud environments. Enterprise features include:

• Advanced security (LDAP/Kerberos authentication, field-level encryption)
• Audit logging and compliance tools
• Encrypted storage engine
• Ops Manager (on-premise MongoDB management platform)
• Commercial support with dedicated SLAs

Enterprise Advanced customers are typically large enterprises with strict data sovereignty requirements, regulated industry constraints (financial services, healthcare, government), or existing investments in on-premise infrastructure that make a full cloud migration costly. Growth is slower (+6% in FY2025) because new application development increasingly goes to Atlas, and the on-premise market is structurally shrinking over time.

MongoDB Competitors

MongoDB’s competitive landscape varies by customer type and use case:

• Oracle — the dominant relational database vendor; enterprises that standardize on Oracle Database compete with MongoDB for new application workloads, though the two databases serve somewhat different use cases
• Snowflake — data warehouse and analytics platform; competes with MongoDB’s Atlas Data Federation and for enterprise data platform budget
• Datadog — competes for developer infrastructure and observability spending; not a database competitor but a comparable in enterprise developer tooling

For direct database comparisons:

• MongoDB vs Oracle — document NoSQL vs. relational database: use cases, pricing, and enterprise adoption compared
• Palantir vs Snowflake — data platform competitors in the broader enterprise data infrastructure market

Other database competitors not yet covered on Visuwire:

• AWS DocumentDB — Amazon’s MongoDB-compatible managed database; directly substitutes Atlas for AWS-centric customers at a lower price point. DocumentDB is MongoDB’s most significant competitive threat because it runs in AWS (where most MongoDB Atlas workloads run) and is API-compatible, reducing switching friction
• PostgreSQL — open-source relational database that has added JSON document support; increasingly adopted by startups that would previously have chosen MongoDB, particularly as AI-native applications often use Postgres for structured metadata
• Redis — in-memory database used for caching and real-time use cases that sometimes overlap with MongoDB’s real-time data capabilities
• Elasticsearch / OpenSearch — full-text search engines that Atlas Search competes with for developer workloads requiring text search

Revenue Breakdown

Financial data sourced from MongoDB FY2025 Annual Report (10-K). Fiscal year ends January 31.

Atlas — 66% of Revenue, +23.5% Growth

The cloud database-as-a-service product and MongoDB’s primary growth engine. Beyond the core database, Atlas includes a growing suite of integrated capabilities:

• Atlas Search — full-text search engine built directly into Atlas, eliminating the need for a separate Elasticsearch deployment. Competes with Elastic’s hosted Elasticsearch service
• Atlas Vector Search — vector database capability for AI applications; developers building semantic search, recommendation engines, and RAG (retrieval-augmented generation) workflows can store and query vector embeddings natively in MongoDB without a separate vector database (Pinecone, Weaviate, Qdrant)
• Atlas Stream Processing — real-time data pipeline processing within the Atlas platform; competes with Apache Kafka-based architectures for event-driven applications
• Atlas Data Federation — query data across Atlas, S3, and other sources using a single MongoDB query interface
• Atlas App Services — serverless backend services (authentication, functions, triggers) that make Atlas more than a database — positioning it as a development platform

The Atlas suite strategy is important: by bundling adjacent capabilities (search, vector, streaming, serverless) into the database platform, MongoDB increases switching costs and ARPU simultaneously. A team that uses Atlas Search, Atlas Vector Search, and Atlas Triggers alongside the core database is deeply embedded in the MongoDB platform.

Enterprise Advanced — 34% of Revenue, +6% Growth

Self-managed MongoDB for on-premise and private cloud deployments. Growth is stable but structurally limited by the long-term migration of enterprise workloads to public cloud. Enterprise Advanced has high retention (enterprises rarely rip out a core database system) but limited expansion opportunity compared to Atlas consumption growth.

Revenue Trend (3-Year)

The deceleration from 47% to 14% over three years reflects: (1) the normalization from COVID-era developer spending acceleration, (2) customer workload optimization that compressed Atlas consumption growth in FY2023-FY2024, and (3) growing comparables making percentage growth harder to sustain at scale. Atlas growing at 23.5% on a $1.26B base is more absolute-dollar growth than 57% on a $500M base.

Customer and Retention Metrics

Net Revenue Retention (NRR) above 120% for Atlas is exceptional. It means that even if MongoDB acquired zero new Atlas customers, the existing Atlas customer base would grow revenue by 20%+ annually through organic consumption expansion. For context: Snowflake targets 130%+ NRR; Datadog runs around 130% NRR; MongoDB’s 120%+ is strong but below the very top tier. See annual recurring revenue for how NRR compounds.

The growth in customers spending >$100K ARR (from 2,300 to 2,700, +17%) is the most important enterprise health metric — these are the accounts with the deepest product integration and the highest consumption expansion potential.

MongoDB (MDB) Income Statement

Financial data sourced from MongoDB SEC filings.

Key Financial Metrics

• Gross Margin: 72.4% — Solid for a managed database service. Atlas carries lower gross margins than Enterprise Advanced because of cloud infrastructure costs (compute, storage, networking paid to AWS/Azure/GCP). As Atlas clusters grow larger and MongoDB’s negotiated cloud infrastructure rates improve with scale, gross margins should expand gradually. Enterprise Advanced’s software-and-support model carries ~80%+ gross margins

• GAAP vs. Non-GAAP Operating Margin gap (~20 points) — The -4.2% GAAP operating margin vs. ~16% non-GAAP reflects approximately $370M in annual stock-based compensation (nearly 20% of revenue). See GAAP vs. Non-GAAP for why this distinction matters. GAAP operating margin has improved dramatically — from -37% in FY2022 to -4.2% in FY2025 — as MongoDB has scaled revenue on a flatter cost base

• Free Cash Flow: ~$290M — FCF margin of ~15%. MongoDB is already FCF-positive despite GAAP losses, consistent with the SaaS accounting pattern where non-cash SBC suppresses GAAP income without consuming cash. FCF generation enables investment in growth without dilutive equity issuance

• Operating Leverage — MongoDB’s path to consistent profitability is visible: R&D and S&M spend are growing slower than revenue, driving incremental margin expansion each year. Non-GAAP operating margin improved from ~13% to ~16% in one year and is trending toward 20%+

• Billings — For Atlas (consumption-based), traditional SaaS billings metrics are less relevant than for seat-based businesses. Monthly consumption invoices mean there is limited deferred revenue build from Atlas. Enterprise Advanced does generate deferred revenue from annual prepayments, which investors track via billings

Is MongoDB Profitable?

On a GAAP basis, MongoDB barely turned profitable in FY2025 — reporting $30 million in net income on $1.92B in revenue. This was the company’s first GAAP-profitable fiscal year.

However, the more meaningful profitability metrics are:

• Non-GAAP operating income: ~$310M (~16% non-GAAP operating margin) — what the business generates before non-cash SBC charges
• Free cash flow: ~$290M — cash the business actually produces after capital expenditures

MongoDB’s path to consistently strong GAAP profitability runs through two levers: revenue scaling faster than headcount (operating leverage) and stock-based compensation declining as a percentage of revenue over time. Both trends are in motion.

Atlas Vector Search: The AI Opportunity

The most significant new growth catalyst for MongoDB is Atlas Vector Search — the vector database capability built natively into the Atlas platform.

Why vector databases matter for AI: Large language models (LLMs) like GPT-4 and Claude process and generate text, but they need a way to efficiently search through large datasets of domain-specific information. The dominant architecture is RAG (Retrieval-Augmented Generation): convert documents, customer records, product catalogs, or any data into mathematical vectors (embeddings), store them in a vector database, and at query time retrieve the most semantically similar vectors to provide relevant context to the LLM.

MongoDB’s angle: Instead of forcing developers to use a separate specialized vector database (Pinecone, Weaviate, Qdrant, Chroma), MongoDB allows storing vector embeddings alongside the regular application data in Atlas — in the same cluster, queried with the same MongoDB driver. For developers already using MongoDB as their primary application database, this is a compelling reason not to add a second database vendor.

Early traction: MongoDB reported thousands of customers using Atlas Vector Search as of FY2025. The product generates additional Atlas consumption (storing and querying vectors is compute-intensive), meaning AI adoption directly increases Atlas revenue from existing customers.

The risk: Cloud providers (AWS, Azure, GCP) and specialized vector database companies are competing aggressively in this space. AWS’s own vector search (Amazon OpenSearch, Amazon Bedrock Knowledge Bases) and Microsoft Azure AI Search could absorb much of the enterprise AI search market before MongoDB achieves significant Vector Search revenue scale.

What to Watch

1. Atlas consumption growth stabilization — After years of deceleration, whether Atlas growth has found a floor around 20–25% is the most important question. A re-acceleration toward 30%+ would meaningfully re-rate the stock. A further deceleration toward 15% or below would signal that workload optimization headwinds are deeper than expected

2. AI/Vector Search revenue contribution — MongoDB has not broken out Vector Search revenue specifically. Investors should watch for any management commentary on AI workload growth as a share of Atlas consumption — this is the clearest signal of whether the AI opportunity is translating into incremental revenue

3. AWS DocumentDB competitive threat — Amazon’s MongoDB-compatible managed database is priced below Atlas and runs natively in AWS. As more enterprises move workloads to AWS, the decision of whether to use Atlas or DocumentDB is made millions of times per year. MongoDB has competed effectively so far by offering superior multi-cloud support, a better developer experience, and more advanced features (Vector Search, Stream Processing, App Services) — but the threat requires constant monitoring

4. Gross margin trajectory — As Atlas grows as a share of revenue (from 66% now toward 75%+), cloud infrastructure costs could pressure blended gross margins. If Atlas gross margins don’t improve with scale (via better AWS/Azure pricing), the margin mix shift from higher-margin Enterprise Advanced to lower-margin Atlas would be a headwind

5. Enterprise Advanced migration to Atlas — The strategic goal is to convert on-premise Enterprise Advanced customers to Atlas over time. Each converted customer increases lifetime revenue (Atlas consumption typically exceeds on-premise license value at steady state) and margin visibility. The migration pace is a key indicator of how quickly the revenue mix improves

6. Macro sensitivity of consumption — MongoDB is one of the most macro-sensitive enterprise software companies because consumption billing responds immediately to customer behavior. In a recessionary environment, developers right-size clusters fast — this happened in FY2023 and caused significant revenue disappointment. Monitoring enterprise customer count and average spend per customer quarter-over-quarter gives early warning signals

MongoDB (MDB) Financial Summary

MongoDB (NASDAQ: MDB) generated $1.92 billion in total revenue in fiscal year 2025 (ending January 31, 2025), up 14.3% year-over-year. The company achieved its first GAAP-profitable year with $30M in net income. Atlas — the cloud database-as-a-service product — grew 23.5% to $1.26B and now represents 66% of revenue, with net revenue retention above 120% and meaningful traction in AI workloads through Atlas Vector Search. Non-GAAP operating margin reached ~16% as MongoDB scales revenue faster than operating costs.

For the broader competitive landscape, see the Cloud Computing Sector analysis.