Recovery concepts, defined
What is Mean Time to Clean Recovery (MTCR)?
The measure that backup success and RTO miss: how long until the business is trusted and clean again.
Why the usual metrics mislead after an attack
Most recovery dashboards report two things: did the backup job succeed, and what is the recovery time objective. Both quietly assume the recovery point is good. After ransomware that assumption breaks, because the most recent backups often already contain the attacker. A green backup and a fast restore can still deliver you a compromised environment, which is not recovery at all.
What MTCR actually measures
Mean Time to Clean Recovery measures the full path to a trusted state, including the steps backup metrics ignore:
- Time to select a known-clean point, using anomaly and threat signals to find a backup that predates the attacker.
- Time to validate it in isolation, restoring into a clean room and confirming it is healthy and malware free before you trust it.
- Time to restore identity and critical systems in order, so the business can actually authenticate and operate.
- Time to confirm the business is running, not just that servers booted.
MTCR versus RTO and RPO
RTO asks how fast you are back and RPO asks how much data you can lose. Both are useful, and both assume the point is clean and restorable. MTCR does not. It can be longer than your RTO precisely because it includes finding and proving a clean point, and it sometimes accepts a worse RPO to get a point that predates the attacker. If you only report RTO, you can hit your target and still be compromised. MTCR is the number that tells the truth after an attack.
How to reduce your MTCR
- Pre-select clean pointsRun anomaly and threat scanning continuously so a known-clean point is already identified, not hunted for during the incident.
- Keep a clean room readyStand up and rehearse an isolated recovery environment so validation does not start from zero.
- Automate identity-first recoveryScript and test the sequence that restores Active Directory, Entra and DNS before dependent applications.
- Rehearse and time itRun full clean recoveries on a schedule and measure MTCR, so the number shrinks and stops surprising you.
Why measure it at all
What gets measured gets improved. Time-to-restore rewards speed to any state, including a compromised one. Time to clean recovery rewards the thing that actually matters: speed to a trusted state. Tracking it turns recovery readiness from a belief into a number you can lower, report to leadership, and stand behind on the worst day.
How KELYN makes this operational
KELYN designs recovery on Commvault to shorten the real path to clean: pre-identified clean points, an isolated recovery environment, identity-first sequencing, and tested runbooks. Then it rehearses and times the full clean recovery, so the number you report is the number you can deliver. See where you stand with the two-minute Recovery Readiness Scorecard.
Sources
Your next backup will run
Will your business come back?
You can prove it in two minutes. Test your recovery readiness and see where the gaps are before an attacker does.