What happens after the system goes live

Urgent gets argued during an outage, which is the worst moment to define it. Included gets argued when a request arrives that one side calls a fix and the other calls new work. Both are settled in writing before launch: a severity table with an example against each level, and a line between configuration, a fix, a small change and new work.

The response window for each severity is written into your agreement and depends on the tier you choose. A window stated without the tier behind it is a promise nobody can check, so this page explains how the levels work and the agreement carries the figures.

Early in the month a supervisor reports that photographs will not upload from one phone model, although the form still saves. Under the agreement's table that is severity 2: one function broken, a workaround in place. The fix changes how the app compresses images on that model. It is reviewed, tested on the same model and entered in the change log with the approver's name.

The operations manager asks for a column on the weekly client report showing who closed each job. That is a small change, inside the tier's allowance. The same week the manager asks for a second approval stage on jobs above a set value. That adds a state, a permission and a notification, so it is new work: quoted in writing, started once the quote is approved.

Meanwhile the contractor's administrator adds two technicians and rewords a checklist question from the operations manual. That is configuration, and it stays with them.

The monthly note lists the fix and the report change with who approved each, the approval stage quoted and waiting for a decision, two phones that have not synced for four days, and the TLS certificate due for renewal in six weeks. The contractor holds the change log, the note and an open issues list that matches the system as it runs.