RethinkDB: why we failed
In this postmortem from 2017, @kiyanwang candidly breaks down the dissolution of their startup, examining the fatal assumptions, misallocated resources, and market dynamics that led to its shutdown.
Core Principles for Independent Builders:
Early signals were often rationalized away instead of prompting an immediate pivot in distribution or pricing.
The psychological toll of running out of capital compounds operational friction in a solo or small team setup.
Key Antidote: Validate paying willingness before committing months to building sophisticated product features.
Community Debate Themes:
Community discussions on HN centered on identifying product-market fit early, avoiding vanity metrics, and the emotional recovery process after shutting down.
CATEGORY Postmortems & Failures →
HISTORICAL ERA Cloud & Mobile Era (2014–2019)
COMMUNITY SCORE 58 points
TOTAL DISCUSSIONS 6 comments
I really wonder how much of it was timing. Startups with very little to lose aren't going to pay for support contracts on open source software. That takes established companies with SLAs, or quantified dollars lost per minute of downtime, or companies that just can't hire fast enough and too much work to do to bother with investigating DB issues.
But these companies that would pay aren't going to roll out new products on a database only a few folks have heard of.
Let's suppose in an alternate universe Slava had made Rethink his hobby project and it had taken ten years to get to where it is today. Even then, companies that had taken a gamble on it may still be a few years out from where the survivors can (and it would make economic sense to) buy a support contract, but their runway would have potentially carried them into the sweet spot of having early adopters become paying customers.
I think the mentioned examples are less developer tools companies than developer services companies.
I get that they are saying the cloud offering is hard, but if they didn't have that, what was their business model? Were they relying on support and consulting fees? Consulting is hard to scale.
Cloud offering with build-in redundancy (herokudb, RDS, Redis Cloud) is what most app stakeholders are looking for. We want to try RethinkDb but we don't want to deal with servers.
If rethinkdb is good, i hope someone can resurrect this as a cloud service.
Back in 1996, having just joined a developer tools company, I recall my new CEO at our first meeting saying "The developer tools market is such a shitty, shitty market"
Previous discussion: https://news.ycombinator.com/item?id=13421608 ↗