Case Study: Supabase cuts weekly crashes by 60x with Sentry

Supabase cuts weekly crashes 60x with Sentry

Supabase, an open-source platform providing managed Postgres and backend services, faced a significant challenge with the error monitoring for its web dashboard, Studio. Its frontend team's Sentry project was overwhelmed with approximately 475,000 weekly noise events from non-issues like user cancellations, which drowned out real bugs and led to a complete loss of trust in the alerts.

The solution from Sentry involved a complete rebuild of the error monitoring setup. The team implemented client-side filtering to eliminate noise, fixed the root cause of the largest errors, and then utilized Sentry’s features to prioritize and fix high-impact bugs. This resulted in a 60x reduction in weekly crashes, from over 2,000 to just 32, and allowed the team to find and fix bugs affecting ~6,300 users before any customers reported them.

Supabase