MongoDB

MongoDB announced on Sept. 28 that its current president and CEO, CJ Desai, is leaving the company for a senior role at Meta. Former CEO Dev Ittycheria is returning to assume the interim role while the board searches for a permanent replacement. MongoDB stock sold off 18% following the news.

Why it matters: While we acknowledge MongoDB Enterprise Advanced's growth acceleration under Desai's leadership, we believe its recent outperformance mostly comes from secular demand for high-performance artificial intelligence databases rather than any competitive edge that Desai formed. Ittycheria held the role of MongoDB's CEO for 11 years and led the company through its IPO. We think his deep knowledge of its product and clients will support a smooth transition as MongoDB searches for a new leader for its next stage of growth. We think Desai will bring rich enterprise go-to-market experience to Meta, but his track record at MongoDB alone does not guarantee Meta success in the enterprise market. We like that Meta keeps its revenue optionality open for different channels to monetize AI.

The bottom line: We maintain our $335 fair value estimate for no-moat MongoDB. Shares look fairly valued following the selloff. Desai's departure should not affect MongoDB's position as a leading database supplier that stores and feeds unstructured data for large language models. MongoDB Enterprise Advanced currently accounts for 21% of the firm's subscription revenue. While the number ticked up during Desai's tenure, we expect it to decrease over the long term. The cloud-native MongoDB Atlas remains the firm's main growth driver, as nearly all new customers come to MongoDB for Atlas. The number of Atlas new customers has been overshooting the number of all new customers over the past three quarters.

Coming up: MongoDB is holding its annual investor day on Sept. 29. We expect to hear more about AI-oriented product enhancements that deepen the synergy between Atlas and Voyage AI.

Our fair value estimate for MongoDB is $335 per share, which implies a fiscal 2027 enterprise value/sales multiple of 8 times. We expect MongoDB to achieve a five-year compound annual growth rate of 18%. MongoDB has established itself as the mainstream solution for document-oriented databases. We anticipate that MongoDB Atlas should benefit from incremental data workloads from the existing customer base and become the primary growth driver for the company, as more clients move their databases to public clouds. Between fiscal 2022 and the first half of fiscal 2026, the average annual recurring revenue for major customers who spend over $100,000 a year doubled. Expanding workloads from existing customers is key to supporting MongoDB’s double-digit top-line growth through fiscal 2035. The boom in new AI-powered applications should provide a continued tailwind for MongoDB’s expansion.

We expect to see moderate GAAP gross margin expansion of around 400 basis points for MongoDB over the next 5 years. In our view, the pace of MongoDB’s margin expansion should be slower than that of other high-growth database companies due to unfavorable unit economics when Enterprise Advanced customers transition to the cloud-based Atlas product. We believe MongoDB’s GAAP operating margin should turn positive for the first time in the next year or two and continue to expand, reaching 31% by fiscal 2036. The Non-GAAP operating margin should experience a similar upward trend, increasing from 19% to 43% over the next decade.

We assign MongoDB a no-moat rating because we currently see limited adoption of document-oriented databases for mission-critical workloads that embody strong switching costs. MongoDB’s structural design makes it an optimized tool for agile software development; however, too much flexibility can also lead to maintenance headaches in the long run. Therefore, customers tend to stick with relational databases like PostgreSQL for easier data governance. While MongoDB’s annual recurring revenue expansion rate is high at around 120%, and it is a widely-regarded leader among document-oriented databases, we need more evidence that MongoDB can remain an integral part of the enterprise tech stack in the long term before awarding the company a narrow moat. Also, the ramping trajectory of MongoDB’s new AI functionalities can bring higher uncertainty to the company’s future expansion of return on invested capital.

MongoDB is a NoSQL database that provides an alternative for software developers building new applications. Traditional SQL databases store data in tables, and developers need to follow rigid database structure guidelines, or database schema, to ensure data integrity and compatibility. Document-oriented databases such as MongoDB are schema-less, offering developers more flexibility by reducing the burden of upfront database structure design. In addition, MongoDB stores data in Binary JSON format. For developers who know the JavaScript language, adopting MongoDB as the back-end database significantly accelerates the application development timeline because the MongoDB Query Language should appear more familiar than the SQL language, thus flattening their learning curves. In our view, MongoDB’s popularity comes down to its ease of use, which increases developer productivity.

MongoDB’s revenue mainly comes from its commercial products—the fully managed MongoDB Atlas and the self-managed MongoDB Enterprise Advanced. We believe MongoDB offers enterprise customers an attractive value proposition, as its developer-centric features save labor costs by making software engineers more efficient. Meanwhile, MongoDB’s scalability and stability are outstanding for heavy-duty day-to-day usage under proper management. Deploying MongoDB helps a new application hit the market faster for revenue generation, which aligns well with the agile software development methodology. With over 65,000 customers, we think MongoDB has already established itself as the mainstream document-oriented database solution. Competing products like Amazon DocumentDB, Azure Cosmos DB, and Google Firestore all provide MongoDB compatibility.

However, we do not believe MongoDB’s performance as a general-purpose database has created high switching costs for the company. The agile software development approach has shortened the time frame between major application releases, and applications that run on MongoDB often have a shelf life of less than 10 years. This characteristic sets MongoDB apart from its wide-moat database peers, as those systems store key financial and operational data that will persist as long as the company exists, creating high switching costs that support an economic moat.

Document-oriented databases like MongoDB also have technological shortcomings for long-term use because they prioritize speed and efficiency to align with the agile mindset. For example, without schema enforcement, one document may store the user’s age as “age” in numeric format, while another may store it as “user age” in text format. Such inconsistencies can accumulate and dramatically slow down the database’s performance over time. When running applications powered by large language models, these data inconsistencies can also have a negative impact on the quality of the model's responses.

In our view, the migration process of MongoDB should be less challenging than that of other enterprise database systems. Data stored in MongoDB is usually tied to specific applications, which limits the impact of migration to certain business units. This contrasts with the migration process of Oracle Database, where the entire company’s back-office operations face the risk of disruption. Additionally, many specialized, lightweight NoSQL databases provide better performance in certain aspects compared with MongoDB. Key-value databases like Redis offer lower query latency, wide-column databases like Apache Cassandra enable higher data throughput, and services like Pinecone specialize in vector search, a key component for building AI applications. Some SQL databases, like PostgreSQL, also support data storage in the JSON document format and directly compete with MongoDB for workloads.

Although MongoDB, as a general-purpose database, stands out due to its versatility in handling different kinds of data workflows, we don’t identify a workflow type that leaves MongoDB users without alternatives. In other words, customers can always introduce a new system alongside MongoDB and shift some workloads for better performance; different specialized databases combined may provide a more desirable overall performance that fits the customers’ needs better than using MongoDB alone.

Demand for new applications powered by artificial intelligence should continue to increase and fuel growth for MongoDB.

MongoDB Atlas should enjoy an extended growth runway as the cloud migration of OLTP databases is still at an early stage.

MongoDB’s friendly user experience and free Community Server should continue to increase its penetration among the developer community and incentivize the adoption of commercial offerings.

The rise of new database technologies can reduce the number of optimized use cases for document-based databases, limiting MongoDB’s long-term growth potential.

MongoDB faces heavy competition to earn AI-driven demand for new applications from SQL-based data pipelines.

Competition could further intensify for MongoDB as hyperscalers and other NoSQL database vendors invest more resources into product development.