Backup jobs report success with impressive consistency. Restores are where the truth lives, and most organisations do not find out which of the two they have until the worst possible week.
The failures a green tick does not catch
- The job runs but excludes a folder somebody added last year
- The data is present but the application cannot be stood back up around it
- The restore works but takes four days, and the business needed four hours
- The backup is intact but reachable with the same credentials as the thing it protects
Two numbers worth agreeing on
How much data you can afford to lose, and how long you can afford to be down. Until those are written down and agreed by someone who owns the consequences, a backup regime cannot be judged adequate or inadequate — only present.
The only backup you can rely on is one you have already put back.
What we do
For managed clients we test restores on a schedule rather than on request, we document what was restored and how long it took, and we report the times back. When a restore takes longer than the agreed target, that is a finding to fix — not a number to quietly file.
If you do one thing
Pick a file from a month ago, restore it to a scratch location, and open it. If nobody in the organisation can complete that exercise this week, the problem is not your backup software.