shipping-and-launch

A production launch checklist covers six areas: code quality (tests, lint, reviews), security (no secrets, dependency audits, rate limiting), performance (Core Web Vitals, bundle size), accessibility (WCAG 2.1 AA), infrastructure (env vars, migrations, health checks), and documentation. All sections must pass before deploying.

Deploy with the flag off, enable for internal users first, then canary to 5% of users while monitoring error rates and latency for 24-48 hours. Advance through 25%, 50%, and 100% only if metrics stay within thresholds, then remove the flag after full rollout.

Roll back immediately when error rate exceeds 2x baseline, P95 latency rises more than 50%, user-reported issues spike, or data integrity or security problems appear. Feature flag rollbacks take under a minute; redeploying a previous version takes under five minutes.

An error budget release gate uses your SLO's allowed failure fraction to decide if it is safe to ship. Above 20% budget remaining you ship normally, between 0-20% you slow rollouts, and when exhausted you freeze feature work to focus on reliability.

Check that the health endpoint returns 200, error monitoring shows no new error types, latency has not regressed, and the critical user flow works manually. Also verify logs are flowing and confirm the rollback mechanism is ready.