performance-optimization

Fix N+1 queries by replacing per-record lookups with a single query using a join or include clause, such as passing include: { owner: true } to an ORM findMany call. Confirm the fix by checking the database query log before and after the change.

Improve Core Web Vitals by first measuring with Lighthouse and the web-vitals library, then targeting the failing metric: optimize hero images with responsive srcset and fetchpriority for LCP, break up long main-thread tasks for INP, and set explicit image dimensions for CLS.

Add an index only after running EXPLAIN ANALYZE and seeing a sequential scan where an index should apply. Index for the query shape with equality columns first, and skip indexing for low-selectivity filters, leading wildcards, or write-heavy tables.

Pool exhaustion shows as all endpoints slowing at once while the database sits idle. Raising the pool size just moves the queue; instead find what holds connections, and for serverless or autoscaling workloads use a multiplexing proxy like pgbouncer or RDS Proxy.

Do not optimize without measurement evidence of a problem, since premature optimization adds complexity without improving what matters. Also revert any change whose re-measured result falls within run-to-run variance, because neutral changes cost maintenance forever.

Prevent regressions by setting explicit budgets such as JavaScript under 200KB gzipped and API p95 under 200ms, then enforcing them with bundlesize and Lighthouse CI in your pipeline. Complement synthetic gates with RUM alerting on p75 movement in production.