The Pre-Database Dream
In 2007, the tech world was buzzing with the promise of cloud computing. This is when Dwight Merriman, Eliot Horowitz, and Kevin Ryan, fresh off their success building the internet advertising powerhouse DoubleClick, founded a new company named 10gen.
Their experiences at DoubleClick, scaling to handle immense web traffic, had shown them the painful limitations of traditional relational databases. But their first idea wasn't to build a better database. Instead, they aimed to create an all-in-one 'Platform as a Service' (PaaS). This product, called Babble, was envisioned as a complete open-source toolkit that would let developers build and launch scalable web applications without worrying about the underlying infrastructure. It was a concept similar to what Google App Engine and Heroku would later popularize. The goal was ambitious: to offer a full application stack, from the front end to the back end, as a seamless, integrated service.
A Solution Without a Problem
Despite the team's pedigree and the forward-thinking concept, Babble failed to find its audience. The market, it turned out, wasn't quite ready for a PaaS from a small startup. More importantly, developers—the very people 10gen wanted to attract—pushed back against the all-or-nothing approach. They preferred the freedom to assemble their own 'stack' of technologies, picking and choosing the best tools for each part of their application. Babble was seen as too restrictive. The company struggled to gain traction as potential customers showed little interest in buying the complete platform. The grand vision of a unified app server was proving to be a solution in search of a problem the market didn't feel it had.
Discovering the Diamond in the Rough
As part of building the Babble platform, the 10gen team had to solve a core problem: they needed a database that was as flexible and scalable as their vision. Frustrated with existing options, they decided to build their own. This internal database was designed from the ground up to handle the messy, evolving data of modern web applications. It didn't force data into rigid rows and columns but instead used a flexible, document-based model (now known as BSON or Binary JSON). This component, initially just an internal tool for the 'real' product, was a radical innovation. As the 10gen team pitched Babble to potential users, they noticed a surprising pattern: the customers weren't excited about the platform, but they were intensely curious about the novel database powering it. They had inadvertently built their most valuable asset as a side-effect.
The Pivot That Launched a Unicorn
Faced with a failing primary product and surging interest in a secondary component, the founders made a courageous choice. In 2009, they decided to scrap the Babble platform entirely and pivot the entire company to focus on their database. They pulled the database out of the PaaS, gave it a name that evoked its ability to handle massive datasets—MongoDB, for 'humongous'—and released it to the world as a standalone, open-source project. The strategy shifted from selling a platform to building a community around a powerful, free tool. 10gen would then provide commercial support, enterprise features, and eventually cloud hosting to monetize their popular database. The pivot worked. Developers flocked to MongoDB, loving its flexibility and ease of use. The company's focus was now clear, and in 2013, 10gen officially renamed itself MongoDB Inc., aligning its identity with the product that had saved it.
