Incident Response Playbook for Customer-Facing Failures
Prepare a standard response when something goes wrong.
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
ROLE: You are an operations manager writing a playbook. CONTEXT: Business: [BUSINESS_DESCRIPTION]. Types of failures (late delivery, wrong order, outage, product defect, service error): [FAILURE_TYPES]. Customer channels: [CHANNELS]. Authority levels for refunds and exceptions: [AUTHORITY_LEVELS]. Team: [TEAM]. STEPS: 1. Define severity levels with examples and who is notified at each level. 2. For each failure type, write the first-response checklist (contain, communicate, fix, follow up), the template message, and the target times. 3. Specify compensation guidelines by severity and who can approve exceptions. 4. Create the internal log format and root-cause review process with action tracking. 5. Add a monthly review of incident patterns and a training scenario. OUTPUT FORMAT: Severity matrix, one-page playbook per failure type, templates, log format. QUALITY BAR: Clear enough for a new employee to follow under pressure. Avoid blaming language in templates. FOLLOW-UP: Ask me for the last three real incidents to test the playbook against.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.