Refund Request Handler
The Refund Request Handler generates professional, policy-compliant responses to refund requests across the full range of scenarios: approved refunds, denied refunds, partial refunds, and edge cases where exceptions might be warranted. It produces responses that communicate the decision clearly while preserving the customer relationship, which is the hardest part of refund handling.
Support agents, customer success managers, and small business owners who handle their own support use this template whenever a refund request arrives. It is especially valuable for newer agents who are unsure how to phrase a denial without sounding robotic, or how to explain a partial refund without confusing the customer.
The prompt works because it separates the decision logic (does this qualify?) from the communication (how do I say this?). You provide your refund policy, the customer's situation, and the facts of the transaction. The AI applies the policy to determine the appropriate response type and then drafts a message that is clear, empathetic, and specific to the situation. This eliminates both the "let me check and get back to you" delays and the inconsistency of different agents interpreting the same policy differently.
This prompt is just the starting point
Score it with AI, optimize it with one click, track versions, and build your prompt library.
The Prompt
Draft a response to a customer refund request based on the following: **Your Refund Policy**: ``` [PASTE YOUR REFUND POLICY OR SUMMARIZE THE KEY RULES, e.g., - Full refund within 30 days of purchase, no questions asked - 31-60 days: store credit only - After 60 days: no refund - Digital products: refund only if the product has a verified technical issue - Subscription: pro-rated refund for unused months on annual plans] ``` **Customer's Request**: ``` [PASTE OR SUMMARIZE THE CUSTOMER'S MESSAGE, e.g., "I purchased the annual plan 3 months ago but I've barely used it. My company switched to a different tool and I'd like a full refund for the remaining 9 months."] ``` **Transaction Details**: - Product/Plan: [WHAT THEY PURCHASED, e.g., "Annual Pro Plan"] - Purchase date: [DATE, e.g., "January 15, 2026"] - Amount paid: [AMOUNT, e.g., "$240 ($20/month billed annually)"] - Usage: [HOW MUCH THEY USED, e.g., "Logged in 8 times total, last login was 6 weeks ago"] - Previous refund history: [ANY PAST REFUNDS, e.g., "None" or "Received a courtesy refund 6 months ago"] **Your Preferred Outcome**: [WHAT YOU WANT TO DO, e.g., "Pro-rated refund for unused months" or "Follow strict policy" or "I'm open to making an exception to retain them"] Generate: ### 1. Policy Assessment A brief analysis of how the refund policy applies to this specific case. Include: - Which policy clause applies - Whether the request qualifies for a full refund, partial refund, or no refund - Any grey areas or exception considerations ### 2. Recommended Response (ready to send) A complete email/message response that: - Opens with empathy (acknowledge the situation without over-apologizing) - States the decision clearly in the first two sentences - Explains the reasoning briefly (reference the policy without quoting it like a legal document) - If approved: details the refund amount, method, and timeline - If denied: offers the nearest alternative (credit, downgrade, pause) - Closes with a specific next step and an open door ### 3. Alternative Response (if applicable) If the decision is borderline, provide a second version that takes the opposite approach (e.g., a version that approves the exception if the recommended version denies it). This gives you options. ### 4. Internal Notes A brief note for your records: the decision made, the reasoning, and any precedent this sets for future cases. Keep responses under 150 words. Customers requesting refunds want a clear answer, not a lecture about company policy.
Usage Tips
- Paste your actual refund policy: The AI can only apply rules it knows. Vague input like "standard policy" produces vague output. The more specific your policy text, the more accurate the recommendation.
- Include the customer's exact words: Tone matters. A customer who writes "I want my money back NOW" needs a different opening than one who writes "I understand it might be past the window, but I wanted to ask."
- Mention previous refund history: If the customer received a courtesy exception before, the AI will factor that into its recommendation. Repeat exceptions set expectations you may not want to set.
- Use the Internal Notes section: Documenting why you approved or denied a refund prevents inconsistency when the same customer or a similar case comes up again.
Get more from this prompt
Save it, score it with AI, optimize it, and track every version. Free to start.