MongoDB: The Company Behind the Document Database

Matt George

Fractional CTO · Founder of Injuria.ai

MongoDB Inc. is the technology company behind the popular source-available, cross-platform document database, MongoDB. The company provides a suite of commercial products and services built around its database, primarily targeting developers and organizations building modern applications. Its core offering, MongoDB Atlas, is a fully managed cloud database service that simplifies deployment, scaling, and operations. The company's central value proposition is to offer a more flexible, scalable, and developer-friendly alternative to traditional relational databases for a wide range of use cases, from web and mobile apps to real-time analytics and AI.

History

Founded in 2007 in New York City by Dwight Merriman, Eliot Horowitz, and Kevin Ryan, the company was originally named 10gen. The founders had previously built DoubleClick (acquired by Google) and experienced firsthand the scalability and agility challenges of using traditional relational databases for large scale, ad-serving applications. Their goal was to create a database that was built for the scale and speed of modern web applications, prioritizing developer productivity and horizontal scalability. The first public version of the MongoDB database was released in 2009[1].

The company officially rebranded from 10gen to MongoDB Inc. in August 2013 to align its corporate identity with its flagship product. This move signaled a focused strategy on building a business around the database. A significant milestone in the company's history was its initial public offering (IPO). MongoDB went public on the NASDAQ stock exchange under the ticker symbol MDB on October 19, 2017, raising $192 million.

Since its IPO, MongoDB has pursued a strategy of expanding its platform capabilities through strategic acquisitions. In October 2018, it acquired mLab, a popular MongoDB-as-a-service provider, to accelerate the growth of its own cloud offering, Atlas. A more transformative acquisition occurred in April 2019, when MongoDB acquired Realm, a mobile database and synchronization platform, for $39 million. This acquisition allowed MongoDB to extend its reach from the cloud backend to mobile and edge devices, offering a unified data platform for developers.

Core Products and Technology

MongoDB's product suite is centered on its document database, but has expanded into a comprehensive data platform.

The MongoDB Database

The core of the company's offering is the MongoDB database. It is a NoSQL database that uses a document-oriented data model. Instead of storing data in tables with rows and columns like a relational database (e.g., PostgreSQL or MySQL), MongoDB stores data in collections of JSON-like documents. These documents use a binary variant of JSON called BSON (Binary JSON)[2], which supports a richer set of data types.

This document model maps directly to objects in modern programming languages, which can make it more intuitive and faster for developers to work with. A key feature is its dynamic schema. Unlike relational databases that require a predefined schema, collections in MongoDB do not enforce a single document structure. This flexibility allows for easier iteration and evolution of the data model as application requirements change, a common pattern in agile development methodologies. The database is designed for horizontal scalability through a process called sharding[3], which distributes data across multiple servers. This architecture allows a MongoDB cluster to handle massive volumes of data and high throughput loads.

MongoDB Atlas

MongoDB Atlas[4] is the company's flagship product and primary revenue driver. Launched in 2016, Atlas is a fully managed, global cloud database-as-a-service (DBaaS). It automates the operational and administrative tasks of running a database, such as provisioning, configuration, patching, backups, and scaling. This allows development teams to focus on building applications rather than managing database infrastructure.

A key feature of Atlas is its multi-cloud architecture. Customers can deploy their databases on Amazon Web Services (AWS), Google Cloud Platform (GCP), and Microsoft Azure. This provides flexibility and helps organizations avoid vendor lock-in with a single cloud provider. Atlas offers a "serverless" instance type that automatically scales storage and compute resources up or down based on application demand, providing a consumption-based pricing model. The platform also includes a suite of integrated services, such as:

  • Atlas Search: A managed full-text search capability built on Apache Lucene, eliminating the need to run and synchronize a separate search engine like Elasticsearch.
  • Atlas Vector Search: A feature for storing and querying vector embeddings, designed for building applications with AI, such as semantic search, recommendation engines, and generative AI.
  • Atlas Data Federation: A service that allows users to query data across MongoDB clusters and cloud object storage (like Amazon S3 and Azure Blob Storage) using the MongoDB Query Language (MQL).
  • Atlas Stream Processing: A feature for processing streams of data in real-time before it lands in the database.

MongoDB Enterprise Advanced

For organizations that prefer or require self-hosting their databases, MongoDB offers the Enterprise Advanced subscription. This is a commercial package that includes the core MongoDB database along with advanced security features (like LDAP and Kerberos integration), management and monitoring tools (Ops Manager), enterprise-grade support, and a commercial license. This option is typically chosen by companies in highly regulated industries or those with existing data center investments who need more control over their database environment.

Realm Mobile Database

Following its acquisition, Realm[5] was integrated into the MongoDB ecosystem. It is a mobile-first database designed to run directly on iOS and Android devices. As of September 2024, MongoDB deprecated Atlas Device Sync and the Realm SDKs, with end-of-life scheduled for September 2025. Previously, its key feature was Atlas Device Sync, which automatically synchronized data between mobile clients and a MongoDB Atlas backend. This simplified the development of applications that needed to work offline or have real-time collaborative features, as it handled the complexity of data conflict resolution and network management.

Business Model

MongoDB's business model has evolved from traditional software licensing to a cloud-first, consumption-based strategy. The company generates revenue primarily through subscriptions.

As of its first quarter fiscal 2025 financial results, MongoDB reported total revenue of $450.6 million. Of that, $436.9 million (approximately 97%) was derived from subscriptions. The remaining revenue comes from services, which includes consulting and training.

The main subscription offerings are:

  1. MongoDB Atlas: This is the company's largest and fastest-growing revenue source. Atlas operates on a consumption-based model where customers pay for the resources they use, including compute, storage, data transfer, and usage of additional features like Search or Vector Search. This model provides a low barrier to entry, allowing developers to start with a free tier and scale their spending as their application grows. As of the end of Q1 FY25 (April 30, 2024), MongoDB reported having over 49,200 total customers; while a specific number of Atlas customers was not provided for the quarter (in Q4 FY24, the total number of customers was 47,800, and the number of Atlas customers was over 46,300), the company reported that Atlas revenue accounted for 70% of total revenue.
  2. MongoDB Enterprise Advanced: This subscription is sold on a term-based license, typically on a per-server or per-core basis, for self-managed deployments. It provides a predictable revenue stream from large enterprises that are not yet on the cloud or have specific security and compliance needs for on-premises hosting.

The company also maintains the MongoDB Community Server, a free-to-use, source-available version of the database. While it does not generate direct revenue, it is a critical part of MongoDB's go-to-market strategy. It fosters widespread developer adoption and a large user community, creating a funnel that drives users towards the paid Atlas and Enterprise Advanced products as their needs for scale, support, and advanced features grow.

Market Position and Competition

MongoDB is a prominent player in the global database market and is often considered the leading NoSQL database. According to the DB-Engines ranking, which measures database popularity based on web mentions, job postings, and search frequency, MongoDB consistently ranks as the most popular NoSQL database and among the top five databases overall, alongside established relational systems like Oracle, MySQL, and PostgreSQL.

MongoDB's competition can be categorized into several groups:

  • Other NoSQL Databases: Its most direct competitors are other document or wide-column NoSQL databases. Amazon DynamoDB[6] is a major competitor, deeply integrated with the AWS ecosystem and offering a serverless model. Other competitors include Couchbase[7], which also offers a flexible JSON model and a mobile-to-cloud sync solution, and Google's Cloud Firestore[8].
  • Relational Databases with JSON Support: Traditional relational databases have responded to the popularity of the document model by adding support for JSON data types. PostgreSQL[9], with its powerful JSONB data type and rich indexing capabilities, is a formidable competitor. Many developers choose to use PostgreSQL for both relational and document-style data to avoid managing two different database systems. Similarly, MySQL[10] and SQL Server have also incorporated native JSON support.
  • Cloud Service Providers: The largest cloud providers, AWS, Microsoft, and Google, represent a complex dynamic of "co-opetition". While MongoDB partners with them to run Atlas on their infrastructure, these providers also offer their own competing database services. For example, a developer on AWS might choose DynamoDB over MongoDB Atlas due to tighter integration with other AWS services and potentially simpler billing.

MongoDB's strategy to compete is to position Atlas as a superior, multi-cloud data platform that offers more integrated features (Search, Vector Search, mobile sync) and freedom from single-cloud lock-in.

Strengths, Criticisms, and Controversies

Like any major technology, MongoDB has genuine strengths as well as a history of limitations and controversy.

Genuine Strengths

  • Developer Productivity: The document model and MongoDB Query Language (MQL) are often cited as highly intuitive for developers working with object-oriented programming, as data in the database closely mirrors objects in their code. This can lead to faster development cycles.
  • Flexible Data Model: The dynamic schema is a significant advantage for applications with evolving requirements, common in agile and fast-moving projects. It allows developers to change the application's data structure without performing complex and slow database migrations.
  • Scalability Architecture: The database was designed from the ground up for horizontal scaling via sharding. This makes it a strong choice for applications that are expected to grow to handle very large datasets or high traffic loads.
  • Unified Data Platform: With Atlas, MongoDB has successfully bundled database hosting with full-text search, vector search, data federation, and mobile data sync. This integrated platform can simplify the tech stack and reduce the operational burden of managing multiple disparate systems.

Limitations and Criticisms

  • Historical Data Consistency Concerns: In its early years, MongoDB faced significant criticism regarding its default write concern settings, which prioritized performance and availability over data durability. A series of analyses by Jepsen, a distributed systems testing firm, highlighted scenarios where data could be lost under certain network partition conditions. In response, MongoDB has since changed its default write concerns to be safer and has improved its replication protocols.
  • Transactions: For many years, MongoDB lacked support for ACID (Atomicity, Consistency, Isolation, Durability) transactions across multiple documents. This was a major limitation for applications requiring strong consistency, such as financial or e-commerce systems. The company addressed this by introducing multi-document ACID transactions in version 4.0 in 2018, though some in the community still view relational databases as more mature for transaction-heavy workloads.
  • Storage Overhead: The BSON format, with its self-describing nature (storing field names in every document), can lead to higher storage consumption compared to the more compact row-based format of relational databases, especially for datasets with many small documents and long field names.
  • Complexity of Self-Hosting: While Atlas simplifies operations, setting up, managing, and correctly configuring a sharded, self-hosted MongoDB cluster is a complex task that requires significant expertise. Misconfiguration can lead to performance problems or data imbalances.

Licensing Controversy

One of the most significant controversies surrounding MongoDB was its change of license. In October 2018, MongoDB announced it was moving its open-source database from the GNU AGPL v3 license to a new, custom license called the Server Side Public License (SSPL)[11]. The company stated this was done to protect its business from large cloud providers who were taking the open-source code, offering it as a commercial service, and capturing all the revenue without contributing back to the community or the project.

The SSPL requires that any organization offering the software as a service must also open-source all of the management, automation, and other supporting code that makes the service possible. This move was highly controversial. The Open Source Initiative (OSI)[12], the stewards of the official definition of "open source," reviewed the SSPL and ultimately stated that it does not conform to their Open Source Definition. As a result, MongoDB is now more accurately described as "source-available" rather than "open source," a distinction that matters for organizations with strict policies on using only OSI-approved licenses.

Decision Lens for Technical Leaders

Choosing a database is a critical architectural decision with long-term consequences. Technical leaders should evaluate MongoDB based on their specific use case, team skills, and business goals.

Good Fit Scenarios

  • Agile and Evolutionary Projects: MongoDB's flexible schema is ideal for early-stage startups or projects where the data model is expected to change frequently. It allows for rapid iteration without being bogged down by schema migrations.
  • Content Management and Catalogs: Use cases involving semi-structured or unstructured data, such as product catalogs, articles, or user-generated content, are a natural fit for the document model.
  • High-Scale, Read/Write Intensive Applications: Applications like IoT data ingestion, logging, real-time analytics, and social media feeds benefit from MongoDB's horizontal scaling and high throughput capabilities.
  • Teams Seeking Operational Simplicity: For teams that want to minimize database administration overhead, MongoDB Atlas provides a compelling, managed DBaaS solution with a rich feature set, allowing them to focus on application development.
  • Modern Application Stacks: Developers using JavaScript-heavy stacks (e.g., Node.js, React) often find MongoDB's JSON-like document model to be a very natural and productive fit.

Poor Fit Scenarios

  • Highly Relational and Interconnected Data: Systems that are fundamentally about relationships between stable entities, such as a traditional financial ledger or a human resources system, are often better served by a relational database. While MongoDB has a $lookup operator for joins, complex, multi-table joins are the native strength of SQL databases.
  • Strict ACID Transactional Systems: For applications where complex, multi-step transactions with absolute guarantees are the primary workload, a mature relational database like PostgreSQL or Oracle might be a more conservative and trusted choice, despite MongoDB's addition of ACID support.
  • Organizations with Strict Open Source Policies: Companies that can only use software with an OSI-approved open-source license will not be able to use recent versions of MongoDB Community Server due to the SSPL.
  • Analytics Requiring Ad-Hoc Joins: While MongoDB has analytics capabilities, data warehousing and business intelligence systems that rely heavily on ad-hoc queries across many different, normalized tables are typically built on columnar or relational data warehouses.

Key Tradeoffs to Weigh

A technical leader must consider several tradeoffs:

  • Flexibility vs. Discipline: The freedom of a dynamic schema is powerful, but it shifts the responsibility for data consistency and validation from the database to the application layer. This requires strong team discipline and clear coding standards to prevent inconsistent or "dirty" data.
  • Developer Velocity vs. Querying Complexity: While simple queries are easy, complex aggregations and joins can be more difficult to write and optimize in MQL compared to SQL for those with a relational background.
  • Managed Service vs. Cost at Scale: MongoDB Atlas offers immense convenience, but its consumption-based pricing can become a significant expense for very large, high-throughput applications. Leaders must model the total cost of ownership (TCO) of Atlas against the cost and complexity of self-hosting or using an alternative DBaaS.
  • Single Platform vs. Best-of-Breed: Using Atlas's integrated Search and Vector Search simplifies the architecture. However, it may mean forgoing the more advanced features of a dedicated system like Elasticsearch or a specialized vector database like Pinecone. The tradeoff is between platform simplicity and the power of specialized tools.
Sources (12)
  1. MongoDB Our Story
  2. MongoDB Docs: BSON Types
  3. MongoDB Docs: Sharding
  4. MongoDB Atlas Product Page
  5. MongoDB Realm Product Page
  6. Amazon DynamoDB
  7. Couchbase
  8. Google Cloud Firestore
  9. PostgreSQL JSON Types Documentation
  10. MySQL JSON Data Type Documentation
  11. MongoDB Server Side Public License (SSPL) FAQ
  12. Open Source Initiative (OSI) on SSPL

Recommended reading

Oracle: Databases, Cloud, and a Long History13 min readOracle is a multinational technology corporation known for its relational database software, enterprise applications, and a growing portfolio of cloud infrastructure services.Datadog: The Observability Company, Explained12 min readA comprehensive overview of Datadog's history, products, business model, and market position as a leading observability and security platform for cloud applications.Microsoft Azure: Products, Business Model, and Tradeoffs14 min readMicrosoft Azure is a leading cloud computing platform offering a wide range of services, from virtual machines to artificial intelligence, primarily targeting enterprise customers and organizations with existing Microsoft investments.Cockroach Labs: Distributed SQL as a Company13 min readA technical overview of Cockroach Labs, the company behind the distributed SQL database CockroachDB, covering its history, technology, business model, and strategic tradeoffs for engineering leaders.Netlify: The Jamstack Platform Company14 min readA comprehensive overview of Netlify, the web development platform known for pioneering the Jamstack architecture, detailing its history, products, business model, and competitive landscape for a technical audience.VMware: Virtualization, the Cloud, and the Broadcom Era11 min readA comprehensive overview of VMware's history, core technologies, business model, and strategic shifts following its acquisition by Broadcom, written for a technical leadership audience.