Churn Diagnosis
Part of Retention after a bad customer experience
Measuring whether a resolved complaint restored trust
Separate remedy completion, customer feedback and later purchasing to assess a resolved complaint without overstating the result.
A closed complaint does not measure restored trust. Assess three separate outcomes: whether the agreed remedy was completed, whether the customer says it addressed the problem, and what happens when another genuine buying opportunity arises. Report each separately; none proves the others.
Define the resolved-case group
Choose the complaint types and period to review. Start follow-up when the agreed remedy was completed, not when an agent changed the ticket status. Check a sample for evidence: delivered replacement, processed refund or completed repair. Keep pending, reopened and unclear cases separate.
Record context to interpret later behaviour: product, issue severity, one-off or repeat purchase, earlier buying pattern and any plausible later need. A customer who recently bought a durable item may have had no opportunity to buy it again.
Key Metrics to Track After Complaint Resolution
- Remedy Completion Rate
- Percentage of cases with verified delivery, refund or repair
- Follow-Up Response Rate
- Proportion of customers who answered the post-resolution question
- Later Purchase Rate
- Customers who made a relevant purchase within the follow-up window
- Reopened Complaints
- Number of complaints reopened within the observation period
Ask a narrow follow-up question
After confirming the action, ask whether the agreed outcome addressed the issue and whether anything remains outstanding. Keep this question separate from a rating, review request or sale. Route an unresolved answer back to service. Record how many customers were invited and how many answered; respondents may differ from non-respondents.
“The replacement arrived and works” supports a conclusion about that respondent's case at that time. It does not establish preference for the business over alternatives. Silence does not establish dissatisfaction. Neither should become a definitive trust label on an individual profile.
Measure later behaviour at a fair opportunity
For customers with a plausible reason to buy again, define a relevant completed purchase and follow-up window before reviewing results. Count only customers who have had the full window.
Show eligible customers, later purchasers, reopenings and further complaints. A replacement supplied as a remedy is not fresh demand. Decide in advance whether a suitable substitute counts as a later purchase.
Where useful, compare with earlier resolved cases that had similar products, purchase periods and buying opportunities. Those cases may still differ in severity or in the service process. Customers who never complained are a less direct comparison because their experiences and willingness to contact support can differ. State these limits with the result.
| Signal | What it supports | What it cannot prove alone |
|---|---|---|
| Remedy completion record | A recorded action occurred | The customer received or accepted it |
| Customer confirmation | That respondent's view at the time | The view of every affected customer |
| Later relevant purchase | The customer bought again | That the complaint response caused the purchase |
| No later purchase | No qualifying order was recorded in the window | That trust was lost, particularly if no need arose |
Report a bounded conclusion
State the case group, completion rule, response count, buying opportunity, follow-up length and data gaps. Report actual counts only after measuring them. Describe later orders and reopened complaints as observed outcomes, rather than proof that recovery restored trust.
To assess whether a changed recovery process improves retention, plan a fair comparison before introducing an optional service step. Provide every customer with any remedy they are entitled to receive. Interpret later purchasing alongside feedback and recurring problems.
