Production Workflow
Enterprise-ready. Security hardened, scalable, monitored, and documented. The product is production-grade.
What production means
Production is the steady state. The system is fully operational, monitored, and supported. You have runbooks for common issues, incident response processes, and continuous improvement cycles. This isn't a destination — it's an ongoing commitment.
Overview
Who it's for
Enterprise customers, high-value users, and anyone who expects 24/7 reliability. No tolerance for downtime or data loss.
Duration
Ongoing. Production is not a phase — it's the steady state for live products. Continuous improvement, not project completion.
Operational cadence
Monitoring review
Check dashboards for anomalies, errors, and performance trends.
Incident review
Review any incidents. Update runbooks. Improve response processes.
Maintenance windows
Regular updates, patches, and infrastructure maintenance.
Security review
Review access controls, audit logs, and security posture.
Key activities
Monitoring and alerting setup
Comprehensive observability: application logs, infrastructure metrics, and business KPIs. Alert on what matters.
Backup and disaster recovery
Automated backups, tested recovery procedures, and documented RPO/RTO targets.
Runbooks and documentation
Step-by-step guides for common issues, deployment procedures, and incident response.
On-call and support processes
Clear escalation paths, response time SLAs, and 24/7 coverage if required.
Continuous improvement
Regular retrospectives, performance optimization, and incremental feature additions.
Compliance and audit trails
Maintain audit logs, access controls, and documentation for regulatory requirements.
AI-assisted operations patterns
AI helps with incident triage: Analyze logs, correlate events, and suggest root causes. Human makes the final call.
AI monitors for anomalies: Pattern recognition across metrics to catch issues before they become outages.
AI assists routine maintenance: Generate release notes, update documentation, and manage repetitive operational tasks.
Quality bar
What "production-ready" looks like:
Full operational readiness
Monitoring and alerting active
Backup and recovery tested
Support runbook complete
SLA defined and achievable
Incident response process documented
Boundaries
Allowed
- • Continuous improvement
- • Feature additions via normal process
- • Scheduled maintenance windows
- • Gradual rollout of updates
- • A/B testing and experiments
Not Allowed
- • Untested deployments
- • Missing monitoring
- • No disaster recovery plan
- • Deploying without rollback ability
- • Missing incident response process
Anti-patterns to avoid
Deploying without a rollback plan
Every deployment should be reversible. If you can't roll back in 5 minutes, you're not ready.
Missing incident response process
When things break (and they will), you need a documented process. Chaos during outages costs trust.
No on-call or support coverage
Production systems need someone responsible. Define coverage hours and escalation paths.
Undocumented operational procedures
If only one person knows how to fix it, you have a single point of failure.
Stage skipping risks
Each stage exists for a reason. Skipping stages leads to:
POC → PROD skip
Security vulnerabilities, data loss risks, and operational fragility. You haven't validated with real users or hardened the system.
MVP → PROD skip
Poor user experience, missing features, and support chaos. Early feedback wasn't incorporated; polish didn't happen.
MMP → PROD skip
Operational failures, missing runbooks, and incident response gaps. You're running production without operational readiness.
Operational readiness checklist
Before going live, verify all items are complete:
- All audits pass at complete level
- Monitoring dashboards configured and alerting active
- Backup procedures tested with successful restore
- Support runbook written and reviewed
- On-call rotation or support hours defined
- Refine checkpoint passed with stakeholder sign-off
- Rollback procedure tested and documented
Design in Production
Design system updates in production go through proper review. Use feature flags for visual changes and A/B test significant updates.
Explore the Design System