Turn a Cancellation Request Into a Save Script
If you run on subscription or recurring revenue (a digital membership, a box subscription, a recurring order), the most expensive moment is cancellation: finding a new customer costs far more than keeping the one you already have. But most sellers do one of two things when a cancellation comes in: nothing, or panic and throw a discount at it. Both are wrong. The first one lets money walk out the door, the second one burns your margin and teaches the customer that canceling gets them a discount. The right move is to understand the REAL reason first, then make a save offer that fits that reason. Is it price, is it that they stopped using it, is it a real problem, are they switching to a competitor? Each reason needs a different answer. This prompt reads the incoming cancellation, extracts the actual reason underneath the stated one, and builds a save script suited to it, one that does not reach for the discount first. The goal is not to force everyone to stay; it is to keep the ones who can be kept with the right move, and send the rest off on a good note.
You are a retention strategist who specializes in subscription and recurring-revenue businesses. Your expertise: reading the REAL reason behind an incoming cancellation request and building a save script tailored to that reason, one that protects margin. Your job is NOT to force every customer to stay; it is to keep the ones who can be kept with the right move, and send off the ones who can't with a good impression. I'll give you the following: - Business model: [what you sell, e.g. monthly digital membership / box subscription / recurring order] - Customer's cancellation message: [paste it verbatim] - Customer history (if available): [how long they've been subscribed, how often they use it, last login, any past issues] - Save tools available to you: [what you can offer: pause, plan change, discount cap, extra support, etc.] RETURN THE OUTPUT IN THIS EXACT ORDER, WITH THESE EXACT HEADERS: === REAL REASON ANALYSIS === Compare the reason the customer WROTE against the reason the data SHOWS. These can differ (stated "too expensive," real "I stopped using it"). Which category does it fall into: [PRICE / FAILED TO SEE VALUE / A PROBLEM OR COMPLAINT / SWITCHING TO A COMPETITOR / TEMPORARY NEED ENDED / UNCLEAR] With a short justification. === SAVE MOVE (ordered from no-discount to discount) === A step-by-step save script matched to the real reason. Do NOT lead with a discount. The usual order: 1. Value reminder / usage support (if the reason is "not using it") 2. Flexibility (pause, downsize the plan, freeze it) 3. Problem-solving (if the reason is a complaint, fix that first) 4. Last resort: a targeted incentive/discount (only for the "values it but price-blocked" customer) Write the actual line to say to the customer at each step. === READY-TO-SEND MESSAGE === Fold the move above into a single, sendable message. Empathetic, not defensive, doesn't corner the customer. Copy-paste ready. === IF YOU CAN'T KEEP THEM: SEND-OFF MESSAGE === If the customer still leaves, a warm send-off that leaves the door open. A bad cancellation experience turns into a bad review; a good send-off is a door back in. === PRODUCT/PROCESS NOTE === What did this cancellation teach you? What should change in the product or process to prevent it from repeating (e.g. weak onboarding, value shown too late). 1-2 concrete notes. RULES: - Don't make the discount your reflex fix. Lead with a discount and you teach customers "cancel and you get a deal," burning your margin. A discount is a last step, only for a valuable customer whose real reason is price. - Don't corner, blame, or guilt-trip the customer. Stay empathetic and direct. - Don't invent features or offers that aren't real; use only what's listed under "save tools available to you." - Write in natural, direct English. Messages should be ready to send as-is.
How to use it
- Paste the customer's cancellation message and whatever account info you have: how long they have been a customer, how often they used the product, what they wrote as the reason. The more concrete the input, the sharper the save move.
- Look at the 'real reason' analysis first. What the customer wrote (too expensive) and the reason underneath it (I never got enough value, I barely used it) can be different things. Pick the save move based on the real reason, not the stated one.
- Keep the discount as a last resort. The prompt gives you non-discount options first (pause, plan change, usage support). Try those; a discount only makes sense for someone who values the product but is genuinely price-blocked. Discount everyone and you have made canceling profitable.
Example / tip
Input: 'Hi, I want to cancel my subscription, it got too expensive.' Customer: subscriber for 8 months, hasn't logged in for the last 2. AI output (shortened): REAL REASON: They wrote 'too expensive' but the data says 'stopped using it' (no login in 2 months). The actual problem isn't price, it's that they never saw the value. SAVE MOVE (in order): (1) Don't lead with a discount. Say 'we noticed you haven't logged in for a while, want me to show you the 3 features that would help you most in two minutes?' (2) Offer a free 1-month pause instead of a cancellation. (3) If neither lands, one question before they go: 'what was missing that would have brought you back?' (product feedback). SEND-OFF: If you can't keep them, a warm send-off that leaves the door open: 'you can pick up right where you left off whenever you're ready.' A bad cancellation experience becomes a bad review.
Usage example
Copy the prompt as is, fill in the bracketed fields for your own business, and paste it into ChatGPT or Claude. The more concrete your input, the more useful the output. Take the first draft, ask for a one-sentence fix on whatever you don't like, then give it a final pass in your own voice.
When not to use it
Don't force it if you have no real input to work with. A prompt doesn't fill an empty field with data, it organizes and speeds up what you already know. Never publish the output unchecked; AI sometimes produces a claim that sounds right but is wrong or made up. Always verify anything that needs proof yourself: numbers, prices, guarantees.
Output quality checklist
- Is the output specific to your input, or generic filler that fits any business? If it's generic, run it again with more concrete input.
- Does every number, ratio, or claim have something backing it? Delete the unbacked ones or soften them honestly.
- Does it sound like you? Simplify the cliches and inflated lines in your own voice.
- Does it give one clear next step? If the reader knows what to do, the output did its job.
