Seeing a green backup notification can create false confidence. Recovery depends on whether the right information can be restored, by the right people, quickly enough to keep the organisation operating.
In this article
Decide what needs to be recoverable
List the information and systems the organisation cannot operate without: customer records, finance data, source files, configuration, email and calendars may all matter.
For each item, decide how much recent work the organisation could afford to lose and how long restoration could take before the impact becomes serious.
Keep recovery copies separate
Ransomware and account compromise can affect storage connected to the same network or administrator account. Use separation so one incident cannot easily damage both the working data and every copy.
Protect access without creating a single point of failure
Limit who can delete or change backups and use strong authentication. At the same time, make sure recovery does not depend on one unavailable person or one password stored inside the failed system.
Document suppliers, account ownership, support contacts, encryption keys and the approval needed to start restoration.
Test a real restore
The NCSC advises organisations to test that backups work. Choose a representative file, database or service and restore it into a safe environment.
Record how long it took, whether the copy was complete, which instructions were missing and whether the restored information was usable. A successful download is not the same as a successful recovery.
Turn the test into an operating rhythm
- Review backup failures and missed jobs
- Test after major system changes
- Rotate the people involved
- Check supplier assumptions
- Update the incident and continuity plan
Recovery confidence comes from evidence. A modest, tested process is more valuable than an elaborate backup design nobody has attempted to restore.
English (UK) 
Français 