Glossary · Database

Read Replica for Analytics

A read replica for analytics is a continuously synced, read-only copy of a production database used to run reports and heavy analytical queries without slowing down the live application.

Analytical queries are greedy: a single revenue-by-cohort scan can chew through CPU and I/O that your checkout flow needs. A read replica isolates that risk–the database continuously streams changes to a copy, analytics runs against the copy, and production stays fast. Replicas are typically seconds behind the primary, which is fine for almost all reporting. For teams adopting AI query tools, pointing the tool at a replica is the standard pattern: everyone can ask anything, as often as they like, with zero chance of impacting customers.

Fig — every answer ships with the tables, rows and SQL behind it.