Returns intelligence
Turn return reasons into a focused product review.
Read the cohort, compare reported reasons and carry the unresolved evidence into the next product discussion. A reason label starts an investigation; it does not establish a cause.
Illustrative local sample. No store is connected.
- 01Jacket
- 02Reported reason
- 03Local draft for discussion
Read the cohort before the pattern.
- What is counted
- Sold items and items reported returned. This sample counts one item per return record, not orders or return authorizations.
- When it is counted
- Order-created cohort: 1–14 September 2026. Reported returns through 30 September 2026.
- What a reason means
- A prepared normalized reason label from an illustrative record. It is not a verified product defect.
Later returns, unrecorded reasons and different reporting cutoffs can change a real cohort. These fixed samples are not mature observed outcomes.
Illustrative local sample. No store is connected.
Same return rate. Different questions.
Switch the product and select a reported reason. Counts, source references and the review question change together; unresolved evidence remains visible.
Order-created cohort: 1–14 September 2026
Reported returns through 30 September 2026
- Sold items
- 120
- Reported returned items
- 12
- Overall return rate
- 10%
- Physical inspection not supplied
- Later cohort outcomes not observed
These evidence gaps stay open for every product, reason and review focus.
From a reported reason to a review question.
Prepared normalized reason records. These references do not link to customer accounts or verified inspection records.
EX-JKT-01 / Fit
- Reported
- The sample record uses the Fit reason label. It contains no verified measurement or physical defect finding.
- Reason count
- 6 of 12 reported returned items (50%)
- Needs verification
- Check the published measurement method, size-guide version and the relevant physical sample before proposing a correction.
- Illustrative source record references
- J-F01, J-F02, J-F03, J-F04, J-F05, J-F06
Make the next discussion specific.
The focus is a human review preference. It can differ from the selected reason and does not change the records or diagnose a cause.
Local draft for discussion
- Product
- Jacket / EX-JKT-01
- Cohort and reporting cutoff
- Order-created cohort: 1–14 September 2026. Reported returns through 30 September 2026.
- Return-rate numerator / denominator
- 12 / 120 = 10%
- Selected reason
- Fit: 6 / 12 = 50% (Share of reported returns)
- Source references
- J-F01, J-F02, J-F03, J-F04, J-F05, J-F06
- Proposed investigation
- Size guide: Review the applicable guide, measurement definitions and version with the product owner.
- Unresolved evidence
- Physical inspection not supplied · Later cohort outcomes not observed
This local draft does not create a ticket, send a task, approve a fix or change a return decision.
Evaluate a change with a comparable later cohort.
- 01
Check the source
Confirm product keys, counted units, reason mapping and dates.
- 02
Inspect product evidence
Review the actual specification, description, guide and sample.
- 03
Document a proposal
Name the change, responsible reviewer and question to evaluate.
- 04
Observe a later cohort
Use a comparable order cohort and reporting window, retaining known differences.
Reason concentration does not identify causality. A before-and-after difference can reflect cohort composition, reporting delays or other changes; this specimen measures no effect or savings.
Read the sample arithmetic
Overall return rate = reported returned items ÷ sold items. Reason share = records for that reason ÷ reported returned items. Jacket: 12/120 = 10%; Fit: 6/12 = 50%, not 50% of sales. Overshirt: 8/80 = 10%; Fit: 2/8 = 25%. The four reason counts sum to 12 and 8 respectively. Shares are displayed to at most one decimal place; the underlying counts stay exact.
Bring the missing evidence to the review.
Agree the source scope, definitions and responsible owner before using actual return records. No store data is transferred through this specimen.
Discuss the data for a review