Pricing & Product · ANALYTICS
Shrink the returns hole
−25%
Returns are almost never spread evenly across a catalogue — they concentrate in a handful of SKUs, sizes and channels. This engagement finds the concentration first, so the fix is aimed at a cluster rather than at everything.
The process
Join returns to orders
Return records matched back to the original order line in the warehouse, carrying SKU, variant, channel, courier and the stated reason code.
Normalise the reason codes
Free-text reasons and platform codes mapped to one taxonomy — fit, damage, description mismatch, delivery — because the raw codes rarely agree across channels.
Find where returns concentrate
Return rate decomposed by SKU, size, channel and campaign, with each cluster costed at its real margin impact including inbound and refurbishment.
Ship the targeted fixes
Size charts corrected, images and copy rewritten against the mismatch reasons, packaging changed on the damage cluster — one intervention per cluster.
Watch the cluster, not the average
Return rate is tracked for the treated cluster against an untreated control, so an improvement can be attributed rather than hoped for.
What we need from you
- Twelve months of order and return records with reason codes
- Product attributes including size and variant structure
- Landed cost of a return, or the components to estimate it
What you get
- Unified returns taxonomy across channels
- Concentration analysis with margin impact per cluster
- A ranked fix list, with the tracking to prove the treated clusters moved
Timeline
Three to five weeks for the diagnosis and first fixes.