Cancellation Prevention Message
Draft a calm, tailored retention message for a customer requesting cancellation.
Customise & copy
Add the context you have. The prompt updates as you work.
Enter the customer's name.
Paste or summarize the customer's stated reason for cancelling.
Select the customer's current subscription tier.
Summarize recent product usage that may inform the alternative solution.
Live prompt
4 fieldsDraft a concise cancellation-prevention message from a Customer Success Manager to {{customer_name}}. Customer context: - Cancellation reason: {{cancellation_reason}} - Current plan tier: Professional - Recent usage activity: {{recent_usage_activity}} Acknowledge the customer's frustration without defensiveness. Ask exactly one clarifying question about the underlying cause. Suggest one specific alternative solution that fits the context, such as a configuration change, onboarding support, training, temporary plan adjustment, or workflow recommendation. End with a low-pressure invitation to stay. Do not invent facts, use hype, include fabricated proof, or ask the sender for more information. Return only the finished message, ready to send after one light edit. Include a subject line followed by the message body.
Proof before you use it
A real example
Tested on gpt-5.6-luna on 2026-09-10
{
"plan_tier": "Professional",
"customer_name": "Maya Chen",
"cancellation_reason": "The marketing team says the platform is too difficult to maintain and they are not getting enough value from the reporting features.",
"recent_usage_activity": "Four of ten seats logged in during the last 30 days. The team created two dashboards but has not automated reports or invited the sales team."
}A small ritual that works
How to use it
- 01Enter the customer's name.
- 02Add the stated cancellation reason.
- 03Select the current plan tier and summarize recent usage activity.
- 04Review the drafted message and make one light edit before sending.
Why it works
The prompt gives the message a clear communication path: acknowledge frustration, ask one focused question, offer one context-based alternative, and close with a low-pressure invitation. The customer name, cancellation reason, plan tier, and recent usage activity provide specific inputs for tailoring the message without requiring the writer to invent supporting details. Requiring a subject line and a ready-to-send body also defines the intended format and keeps the response focused.
Where it fails
The structure is less suitable when the cancellation reason or usage summary is too vague to support a relevant alternative. Its single-question limit also leaves little room to address several unrelated concerns in one message, and the prompt does not provide policies for handling refunds, contractual commitments, security concerns, or other specialized cancellation situations.
Customisation tips
Adjust the alternative-solution examples to match the product's actual support model, such as guided setup, training, a billing review, or a temporary downgrade. Add a tone preference if messages should be more formal, warmer, or more concise. If the sender needs a specific sign-off, escalation path, or approved retention offer, state that directly in the prompt. Keep the one-question and one-solution limits when a short, focused message is the priority.
Watch-outs
- The prompt asks for one light edit while also requiring the message to be ready to send; clarify whether the edit is limited to inserting the sender's name or signature.
- The alternative solution must fit the supplied context, so generic suggestions can weaken the intended personalization.
- The low-pressure closing should not introduce unsupported discounts, guarantees, product claims, or urgency.
- The instruction not to ask the sender for more information means the message must rely only on the four supplied variables.
Works in
Was this useful?
Submit your variation
If your version helps, it may be published with your name and a link back to you.
Related prompts
Want to make AI useful across your team? Explore Atul's training.