100% Free Forever
AI-Powered Learning
Industry Expert Content
Certificates & Badges
Learn At Your Own Pace
DevOps

Mean Time to Recovery (MTTR)

DORA reliability metric

IntermediateConcept6.4K learners

Mean Time to Recovery (MTTR) is a reliability metric that measures the average time it takes to restore service after a production incident or failed deployment.

Definition

Mean Time to Recovery (MTTR) is a reliability metric that measures the average time it takes to restore service after a production incident or failed deployment.

Overview

MTTR measures how quickly a team can detect, diagnose, and resolve an incident once something goes wrong in production. As one of the four DORA metrics, it is treated as a stability counterpart to the delivery-speed metrics of deployment frequency and lead time for changes — together they capture the idea that fast software delivery and reliable software are not in tension when engineering practices are mature. The clock for MTTR typically starts when an incident is detected — via alerting, monitoring, or user reports — and stops when service is restored to an acceptable state, which is not always the same as the underlying bug being fully fixed. A quick rollback or feature-flag kill switch that restores service counts toward MTTR even if a permanent code fix ships later. This distinction matters: teams that can revert or disable a bad change quickly will have a far lower MTTR than teams whose only recovery path is a fresh, fully tested deployment. DORA research shows elite teams recover from incidents in under an hour, while low performers can take a week or more. Low MTTR is usually the product of good observability (so problems are detected fast), clear incident response processes (so the right people engage quickly), and deployment mechanisms that support fast, safe rollbacks — feature flags, canary releases, and blue-green deployments all shorten recovery time by design. MTTR is sometimes confused with related terms like mean time between failures (MTBF) or mean time to detect (MTTD); MTTR specifically covers the recovery phase after a failure is known, not the interval between failures or the time to notice one.

Key Concepts

  • One of the four core DORA metrics for delivery and reliability
  • Measures time from incident detection to service restoration
  • Rollbacks and feature-flag kill switches count as valid recovery paths
  • Elite teams recover in under an hour
  • Improved by strong observability and incident response processes
  • Distinct from mean time between failures (MTBF) and mean time to detect (MTTD)
  • Often paired with postmortems to prevent repeat incidents

Use Cases

Setting incident response SLAs and on-call expectations
Evaluating whether rollback tooling is fast enough for production needs
Justifying investment in observability and alerting infrastructure
Benchmarking reliability performance against DORA cohorts
Informing postmortem discussions about recovery speed, not just root cause

Frequently Asked Questions

From the Blog