[Prompt] Generate a stakeholder follow up draft about the incident

Use this prompt after you’ve filled in the incident post‑mortem template. Attach the completed post‑mortem to your agent of choice and input the prompt below.

The AI will craft a concise external announcement that:

Clearly states what happened and when, using lay terms.
Acknowledges which services or customers were affected and for how long.
Briefly explains how the issue was resolved.
Outlines the key next steps you’re taking to prevent a recurrence.
Includes an apology and contact details for those who need assistance.

Keep the tone transparent and empathetic, avoiding internal jargon. Once generated, review the draft for accuracy and clarity, then publish it via your status page, customer email list or other appropriate channels.

</aside>

Prompt

Context: You need to prepare a stakeholder‑facing incident update post based on the completed incident report.  The available data includes Summary & Impact, Key Details (type, severity, services affected, durations, participants), Contributing Factors & Mitigations, Resolution & Technical Analysis, Lessons & Risks, Follow‑up Actions and Communication.

Task: Generate a public‑facing announcement that:

1. Clearly states the issue and its timeline — explain, in lay terms, what went wrong and when, using information from the “Summary & Impact” and “Chronology & Response” sections.
2. Acknowledges the impact on customers — specify which products or services were affected and the duration of the disruption without revealing sensitive internal details.
3. Describes the resolution — briefly outline how the issue was fixed, drawing from the “Resolution & Technical Analysis” section.
4. Outlines next steps and reassurances — mention the most important follow‑up actions being taken to prevent recurrence and improve reliability, highlighting the organisation’s commitment to learning from the incident.
5. Includes an apology and contact information — convey empathy for any inconvenience caused and provide a way for customers to ask further questions.

Tone: Transparent, empathetic and professional.  Avoid internal jargon; use customer‑friendly language.  Be factual and concise, avoid assigning blame, and focus on restoration and improvement.

Example Output

Status Update: Elevated Response Times on User API – Resolved Incident window: 1 August 2025, 14:03 – 14:33 UTC

What happened: During the window above, some customers experienced slow responses when interacting with our User API. A recent application update introduced a database query that did not perform as expected, increasing average response times to a few seconds.

Impact: Approximately one quarter of User API requests were delayed during this time. There was no data loss, and all other services remained fully operational.

What we did: Our on‑call engineering team detected the issue within minutes and immediately worked to restore normal performance. We reverted the update that triggered the slowdown and optimised the underlying database query. Response times returned to normal by 14:33 UTC.

Next steps: We take service reliability very seriously and are implementing additional safeguards to prevent similar issues. These include: • Expanding our performance testing to better mimic production data volumes. • Automating checks for inefficient database queries before code is deployed. • Reviewing our deployment processes to ensure robust performance evaluations.

We’re sorry for any inconvenience this latency spike caused you. If you have questions or are still experiencing issues, please contact our support team at [email protected]. Thank you for your patience and for trusting us with your business.