Mean time to resolve (MTTR)

Mean time to resolve (MTTR) is the average time from when an incident is detected to when it is fully resolved and normal service is restored, calculated as total resolution time across incidents divided by the number of incidents. The same acronym is also used, inconsistently across teams and vendors, for mean time to repair or mean time to recovery, which can measure slightly different endpoints, so it is worth confirming which definition a given dashboard uses.

MTTR differs from mean time to detect, which covers only the time before an incident is noticed; a team can improve overall incident duration by shortening either metric, and the two are usually reported together. It also differs from mean time between failures, which measures reliability between incidents rather than response speed to any one of them.

MTTR matters as a core input to service level objective performance and to the error budget a team is spending, and a version of it is one of the four DORA metrics used to assess software delivery and operational performance. The common pitfall is chasing a lower average MTTR by fixing many minor incidents quickly while a small number of severe, slow incidents drive most actual customer impact.