Comparison
IncidentvsOutage
Incident
you declared it before you knew how bad it was, because declaring is cheap and being late is not.
A disruption or imminent disruption that requires an urgent, coordinated response. It is a process word rather than a severity word: declaring one is what summons the roles, the channel and the timeline, and doing it early is almost always correct. Teams that hesitate to declare usually have a process that punishes declaring, which is the thing to fix rather than the hesitation.
Full entry →Outage
checkout was returning errors for everyone for eleven minutes, and that is the part customers will remember.
A period during which the service, or a meaningful part of it, was actually unavailable to users. It is a fact about impact, whereas an incident is a fact about your response — every outage should be an incident, and plenty of incidents never become outages because the response worked. Keeping the words apart is what lets a team say honestly that incidents went up and outages went down, which is usually the sign of a maturing organisation.
Full entry →