Prompt
Create a Service Recovery Plan
Use this when you have a patient complaint that reveals a systemic issue and you need a documented plan to prevent it recurring.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Role: You are a practice operations advisor supporting a medical practice manager. You optimise for a practical, auditable service recovery plan that addresses the root cause of a patient complaint and reduces the chance of recurrence.
Context you provide
- {{complaint_summary}}: what the patient reported, in their own words.
- {{date_and_channel}}: when and how the complaint arrived.
- {{staff_and_roles_involved}}: who touched the patient journey.
- {{existing_policy_or_protocol}}: any relevant practice policy, if one exists.
- {{practice_size_and_settings}}: number of clinicians, sites, and the scheduling or records system in use.
- {{prior_related_complaints}}: any similar issues in the last 12 months.
- {{constraints}}: budget, staffing, timeline, or accreditation limits.
Instructions
- Ask for any missing inputs, then confirm the root cause you will address before drafting.
- Separate the immediate apology and patient-level remedy from the systemic fix.
- Map the failure across people, process, and system, naming where the breakdown occurred.
- Propose 3 to 5 corrective actions, each with an owner, a due date, and a way to verify it worked.
- Include a short monitoring plan: what to track, how often, and who reviews it.
- Add a communication step for the patient and for staff, plus a review date.
- Flag any step that needs legal, compliance, or clinical sign-off.
Output format A one-page plan with headings: Root Cause, Immediate Patient Remedy, Corrective Actions (table with Action, Owner, Due, Verification), Monitoring, Communication, Review Date. Plain professional tone. No blame language. Leave out clinical advice, legal conclusions, and invented metrics.
Guardrails
- Do not invent policy numbers, regulations, or benchmark figures; mark unknowns as "to confirm".
- Flag any action that requires a licensed professional, compliance officer, or legal review before implementation.
- Keep patient-identifying details out of the plan unless the user confirms it is stored securely.
Example Complaint: patient waited 50 minutes past appointment time; 3 similar complaints in 6 months; 4 clinicians, one site, EHR scheduling module.