Case Study: Spotify achieves sub-second, petabyte-scale A/B experimentation and 10× reduction in low-variance work with ClickHouse

Spotify - Customer Case Study

Spotify needed to run A/B experimentation at scale—serving thousands of sub-second queries per second against petabyte-scale datasets—while cutting down manual, low-variance work and making results accessible to product teams. To meet those demands they adopted ClickHouse as the backend for their experimentation platform.

ClickHouse delivered a unified SQL interface for the Data Platform plus tools for automatic decision making, enabling feature teams to self-serve insights. The implementation supports thousands of sub-second queries per second on petabyte-scale data and reduced low-variance work by an order of magnitude, accelerating experiment analysis and automating decisions.

Spotify