Case Study Deskcasestudydesk.com
Software / Web AnalyticsSourced

Case Study: Auditzy made queries 33x faster by switching from Postgres to ClickHouse

Auditzy Case StudySourced & dated by Case Study Desk
Key facts · TL;DR
Company
Auditzy
Industry
Software / Web Analytics
Challenge
Postgres query latency on large datasets
Headline result
Median queries fell from ~10 seconds to 300 milliseconds after migrating to ClickHouse

Key results

33x
Faster median queries
~10 seconds in Postgres to 300 ms
10x
Storage compression
5 TB in Postgres to 250 GB in ClickHouse

The challenge

Auditzy captures web performance signals such as page loads and layout stability and turns them into insights for engineering, business, and marketing teams. Its Postgres-based architecture struggled with query latency on large datasets, and bulk inserts slowed during traffic surges despite caching, partitioning, and materialized-view workarounds.

The solution

The team migrated from Postgres to ClickHouse, taking advantage of ClickHouse's SQL compatibility to port most queries with only minor changes.

The difference in terms of query times was like night and day.

MJ
Mayank Joshi
Co-founder & CTO, Auditzy

The results, in context

Auditzy reported that median queries which once took around 10 seconds in Postgres now return in about 300 milliseconds, a 33x improvement, and that 5 TB of data in Postgres shrank to 250 GB in ClickHouse, a 10x compression rate. Figures are quoted from the published ClickHouse story and dated.

Products used

ClickHouse ClickHouse