What it is
MTTR (mean time to recovery) measures how fast a team resolves incidents, while MTBF (mean time between failures) measures how often they happen, and together they roll up into availability.
When to use it
SRE and infrastructure teams use these metrics in postmortems and reliability reviews to see whether incidents are getting rarer, faster to resolve, or both.
Gotchas & limits
Mean-based metrics are sensitive to outliers; one very long incident can dominate an MTTR average, so consider looking at the median or distribution alongside the mean.