RethinkDB
The real-time open-source database company that was loved but could not make money
- Founded
- 2009
- Closed
- 2016
- Lifespan
- 7 years
- Total funding
- $12.2 million
- Country
- United States
- Sector
- Enterprise software (SaaS)
- Founders
- Slava Akhmechet, Michael Glukhovsky
- Investors
- Y Combinator
- At its peak
- Around 100,000 users when the production-ready version 2.0 was released in April 2015.
- Cause of closure
- Business model failed
What did it do?
RethinkDB developed an open-source distributed database that stored JSON documents and pushed changes in query results to applications in real time. For revenue the company offered support, training and application development services, and it was also preparing a platform for real-time apps called Horizon.
Why did it shut down?
Having received seed investment from Y Combinator in June 2009, the company started out with a MySQL storage engine for SSDs, then developed its own document database and open-sourced it in 2012. The team, which prioritised correctness, a clean interface and consistency, did not release its production-ready version 2.0 until April 2015, by which time MongoDB was far ahead in the market. According to founder Slava Akhmechet, users saw the company as an open-source developer tool and were unwilling to pay. Turning to real-time features in mid-2014, the company now ran into Firebase and Meteor, and it ran out of cash before a small team could finish Horizon and the cloud service. On 5 October 2016 the company closed, saying it had been unable to build a sustainable business, and announced that its engineers would join Stripe.
The story
RethinkDB was a database start-up founded in 2009 by Slava Akhmechet and Michael Glukhovsky. Having received seed investment from Y Combinator in June 2009, the company started out with a MySQL storage engine designed from scratch for SSDs. Over time the team developed its own document database and open-sourced the product in 2012. RethinkDB's most distinctive feature was its ability to push changes in query results to applications in real time, which made it attractive for real-time apps.
When the production-ready version 2.0 came out in April 2015, it had around 100,000 users; over the same period MongoDB's downloads had passed 9 million. According to the assessment Akhmechet later wrote, two fundamental mistakes were made. The first was the choice of market: users saw RethinkDB as an open-source developer tool and were unwilling to pay even the price of a cup of coffee for it. The second was that while the team put correctness, a clean interface and consistency first, users wanted a product that arrived on time, ran fast on the jobs they tried it on and solved a specific problem. This perfectionism left the company three years behind the market.
Realising in mid-2014 that they could not compete with MongoDB, the team turned to real-time features, but this time ran into Firebase and Meteor. The money ran out while a small team was trying to build the database, an application platform called Horizon and a cloud service all at once. On 5 October 2016 the company closed, saying it had been unable to build a sustainable business; some of its engineers moved to Stripe.
In February 2017 the Cloud Native Computing Foundation bought the database's source code and copyrights for $25,000 and contributed them to the Linux Foundation under the Apache licence. The database, whose users included GM, NASA and the US Department of Defense, thus lived on as a community project, but the company that created it passed into history.
Lessons for founders
- The market your users see you in matters more than the market you are aiming for; being perceived as an open-source developer tool limited RethinkDB's ability to charge.
- Engineering excellence loses value if it does not arrive on time; users wanted a product that ran fast and solved a specific job rather than one that was correct.
- Building a database, an application platform and a cloud service at the same time with a small team scatters your focus; even under existential threat, commit to a single product.
Sources
- RethinkDB (Wikipedia)
- RethinkDB is shutting down (RethinkDB, 2016-10-05)
- RethinkDB: why we failed (Slava Akhmechet, 2017-01-18)
- CNCF Purchases RethinkDB Source Code and Contributes It to The Linux Foundation Under the Apache License (Cloud Native Computing Foundation, 2017-02-06)
- RethinkDB is dead? Rethink that thought: NoSQL database is back (The Register, 2017-02-07)
This section is compiled from public sources (Wikidata, news archives and company announcements). Figures are given as they appear in those sources; unknown fields are left blank. If you spot a mistake in an entry, you can send us a correction.
Is something in this entry wrong? Request a correction → Method