A cancellation interview should clarify what changed in the customer’s situation and experience. Keep leaving independent of research participation, ask about specific events and distinguish a disappearing need from an unmet promise. Record uncertainty rather than treating one account as the cause of every departure.
Separate the exit process from the research request
Complete or clearly explain the cancellation process before inviting an interview. State that participation is optional and will not affect access to the agreed exit arrangements. Offer a short format and accept a refusal without repeated persuasion. A customer should not need to defend leaving in order to stop a service or resolve an outstanding operational question.
Prepare by reviewing the original goal, relevant usage and support history that you are authorised to access. Avoid collecting a dossier of unrelated personal details. Use the record to ask informed questions, not to challenge the customer’s account. If a refund, complaint or contractual interpretation is involved, route it to the appropriate process; an interviewer should not improvise promises outside their authority. Keep research notes distinct from the operational actions needed to close the account.
Reconstruct the decision without arguing against it
Ask about the last occasion when the service helped, the first sign that it was no longer suitable and the alternatives considered. Follow the sequence of events. “What happened when the report was due?” is more revealing than “Was the dashboard disappointing?” Leave room for changing priorities, project completion or internal staffing shifts. These may explain departure without implying a product defect.
When the customer describes a problem, ask what consequence it had and what they did next. Separate their words from your interpretation in the notes. Check whether the customer would like any promised follow-up, then specify who will provide it. Do not use the interview to pressure them into another contract. If an available remedy is relevant, explain it accurately and allow the customer to choose whether to discuss it.
Classify reasons without flattening the story
Use a primary reason plus supporting context rather than forcing every account into a single permanent code. “Project ended” and “could not trust the reporting period” call for different responses. Mark whether a reason was explicitly stated, inferred from records or remains unknown. A lack of response is not evidence of price sensitivity. Review contradictions, such as high usage alongside dissatisfaction, rather than deleting them to tidy the story.
Compare several cases before choosing a broader change. Look for repeated conditions and examine the relevant delivery records. Assign an investigation owner when a specific failure is credible, while keeping the immediate customer action visible. SBA marketing guidance treats post-sale experience as part of the business relationship; a respectful exit is part of that experience. Your notes should help the organisation learn without pretending that the conversation establishes a universal explanation.
Set retention and access rules for research notes consistent with your organisation’s responsibilities. A short summary of the decision and relevant events is usually more useful to the product team than a widely forwarded transcript. Remove unnecessary identifying details when discussing patterns, and preserve a route to correct a mistaken interpretation.
Fictional worked example
A fictional scenario to explain the method, not a customer result or a market benchmark.
At the fictional service Harbor Ledger, two customers cancel during the same week. An optional conversation with the first reveals that its seasonal project has finished and the team no longer needs weekly reports. The second describes repeated delays obtaining an approved report before a board meeting. The interviewer records “need ended” and “delivery reliability concern” separately, with the customers’ own explanations.
Harbor Ledger closes both accounts through the normal process. It asks the delivery owner to examine the late-report sequence for the second account, but does not launch a discount campaign to address the first customer’s absent need. A third customer declines an interview and remains classified as unknown. These invented cases illustrate respectful classification, not actual cancellation findings or evidence that a remedy would restore either relationship.
TAKE THIS INTO YOUR NEXT CONVERSATION
A practical next-step checklist.
- Make the research invitation optional after clarifying exit.
- Ask about recent events and the decision sequence.
- Label stated, inferred and unknown reasons.
- Assign any operational follow-up separately from research.
Common questions.
Should we offer a discount during the interview?
Only discuss a relevant available option with the customer’s agreement. A discount does not explain or repair every reason for leaving, and research must not obstruct cancellation.
How many interviews establish the main reason for churn?
There is no universal number. Look for repeated evidence across comparable cases and records, disclose non-response and avoid treating a small voluntary sample as representative.
Further reading & source context.
These primary references provide background. They are not evidence for the fictional example above and do not endorse AccelerateLink.
Found an error or a missing qualification? Send an editorial correction.
Find my next move.
Use a practical tool to turn this idea into a clearer starting point.
Find my next move