UNDERSTANDING WHY COUNT(*) CAN BE SLOW IN POSTGRESQL

Summary from the TLDR Web Dev newsletter

COUNT(*) can be slow in PostgreSQL due to MVCC and the need to verify row visibility, but optimizations like filtering on indexed columns or using approximate counts from pg_class can speed the query up.