Forge-Backup-and-Recovery

Define backup scope, frequency, and retention per data criticality classification. Start from a data inventory, classify each dataset by business impact, then set schedules and retention periods that match the documented recovery objectives.

Set recovery-point and recovery-time objectives together with the Product Owner based on business tolerance for data loss and downtime. Document the agreed values explicitly rather than leaving them implied, and review them when data or architecture changes materially.

Verify backups through completeness checks, integrity checks, and restorability sampling. Then rehearse full restores against the recovery objectives on a schedule and record the evidence, since a backup is only proven real once it has been restored.

Do not use it for planning application rollback paths, executing recovery during a live incident, or operating the underlying storage infrastructure. Those belong to rollback planning, incident management, and infrastructure management capabilities respectively.

Backup failures are treated as incidents rather than being silenced. Detection feeds into incident management so the gap in data protection is addressed before it affects recovery capability.