Change Plan

A short description. Use fictional names or generic service labels.

What problem does this solve? What should be different afterward?

Included systems, excluded systems and the smallest useful rollout group.

Affected roles or services, expected interruption and an approved workaround if one exists.

The responsible role. Avoid personal or client identifiers.

Proposed window, time zone and latest safe point to stop.

Upstream and downstream dependencies. Record “None identified” only after checking.

One intended step per line. Include decision points; never paste credentials or confidential commands.

Access, baseline evidence, prerequisites and checks required before starting.

Observable conditions that stop the work and trigger recovery.

How to return to a known working state, how long it takes and how that path has been checked.

The role authorized and available to make the recovery decision.

Required for data or irreversible changes: backup/snapshot scope, recoverability evidence and restore limitations.

Required for irreversible changes: an alternative recovery path, residual loss and the decision owner.

Expected technical results and evidence to capture. Include comparisons to the baseline.

How the original task will be tested from the affected user or service perspective.

Required for production, access or network changes: limited test scope, observation period and stop criteria.

The approval required, responsible role and any outstanding decision. This tool cannot grant approval.

Who needs notice, what they need to know and when progress or rollback will be communicated.