Document an outage for your technical team.

What you need

A past incident affecting one service, a defined time window, approved alert and log excerpts, and a dated list of known changes. Remove secrets, access tokens and unnecessary personal data before sharing files with an approved assistant. Keep original timezones and source references. Incomplete logs and events close together in time cannot establish the cause of an outage on their own; the technical owner validates conclusions and next steps.

Getting started

1. Choose a closed incident and give every excerpt a reference, timestamp, timezone and affected service.
2. Request a timeline that separates observations, missing information and hypotheses, with supporting passages for each item.
3. Have the technical owner check the brief, then compare it with the known incident record before deciding which checks to make next.

What you can measure during a pilot

Information quality, missed actions and preparation time.