Loading
An incident is a service interruption: restore it, quickly. A problem is the underlying cause: find it and remove it, thoroughly.
They have different goals, different urgency and usually different people. Restoring service fast often means a workaround, which is correct for the incident and does nothing for the problem.
If every occurrence is only an incident, you restore service repeatedly and never remove the cause. The same fault recurs monthly, each time handled efficiently, and the total cost is enormous and invisible.
Nobody decides to work this way. It happens because there is no record type for the cause, so there is nowhere to put it.
When it recurs, affects many people, or was resolved with a workaround.
Usually not the person restoring service. Different goal, different urgency.
Link the incidents to them. Aggregate cost persuades where a description does not.
Half an hour with your own data usually saves reading three of these. The guides will still be here afterwards.