Returns disposition: grade, restock, refurbish or scrap

The return does not end at the warehouse door

Most returns reporting stops at receipt. The dashboard says 3,412 units came back last month, finance books the refunds, and the units go into a bay near the inward dock. What happens after that decides whether a return costs you 8 percent of the order value or all of it. Disposition is the decision, taken per unit, on what the item becomes next: sellable stock, repacked stock, discounted stock, or scrap.

Brands that treat disposition as a queue rather than a decision produce the same pattern every year. A returns bay that grows, a stock file that disagrees with the shelf, and a write-off in the final quarter that nobody forecast.

Grade at receipt, in under two minutes

Grading is a fixed protocol performed the moment the unit is scanned in, by a trained grader, against a printed standard taped to the bench. Four grades is enough, and the definitions must be physical rather than subjective.

• Grade A: seal intact, no wear, primary pack undamaged. Back to sellable after a 30-second check.
• Grade B: product fine, outer packaging opened or scuffed. Needs repack, typically Rs 12 to Rs 25 in materials and labour.
• Grade C: usable but shows wear, a missing accessory, or a minor defect. Refurbishment or bundle route.
• Grade D: unusable, contaminated, expired, or beyond economic repair. Scrap or recycler.

Two rules keep grading honest. The grader assigns a grade, never a disposition, and the grade drives disposition automatically through a mapping table. And every grade capture requires a photograph attached to the return ID, because photographs settle disputes with marketplaces, couriers and customers far better than notes do.

What the split usually looks like

Across Indian D2C catalogs, apparel and accessories typically land at 58 to 72 percent Grade A, because most returns are size or fit driven and the product is untouched. Beauty and food run far lower, often 20 to 35 percent, because seal-break rules force anything opened to Grade D regardless of condition. Electronics and appliances sit in between, with a wide Grade C band driven by missing cables, manuals and boxes.

If your Grade A share is materially above the band for your category, you are not grading, you are restocking. The tell is a rising rate of second returns on the same SKU, where a customer receives a unit that was returned once already and sends it back with a complaint. Track second-return rate on restocked units separately. Anything above 4 percent means your Grade A definition is too generous.

When refurbishment stops paying

Refurbishment is worth doing when expected net realisation, minus refurbishment cost, minus the cost of carrying the unit until it sells, exceeds what a liquidator would pay today. Write it out per SKU and the answer usually becomes obvious.

Take a Rs 1,999 MRP accessory. Grade C, needs a replacement retail box and a cleaning pass at Rs 55 in labour and materials. It sells through your outlet channel at Rs 900 with a 12 percent channel cost, and it sits for an estimated 70 days at a carrying cost of roughly 1.4 percent a month on value. Net realisation is around Rs 740. A liquidator offers Rs 430 today. Refurbishment wins comfortably. Now run the same maths on a Rs 599 item where refurbishment costs Rs 55 and the outlet price is Rs 250: the margin over liquidation collapses, and the labour is better spent elsewhere. As a rough boundary, refurbishment stops paying once its cost crosses about 20 percent of the realisable price.

Controls that stop the bay from becoming a graveyard

Three controls do most of the work. First, ageing buckets on the returns location: 0 to 7 days, 8 to 21 days, 22 to 45 days, and beyond. Anything reaching 22 days triggers a forced decision by a named owner, and anything beyond 45 days is provisioned at full value whether or not it has been scrapped, so the P and L stops flattering you.

Second, a blind re-grade audit. Every week, pull 30 units already graded, have a second person grade them without seeing the first result, and track disagreement rate. Above 10 percent disagreement, the standard is unclear and needs re-training rather than a policy memo.

Third, a reconciliation between returns credited to the customer and returns physically received. On marketplace channels this gap is real money: units refunded to a buyer but never received, or received empty. Run it weekly and file claims inside the platform window, which is usually 30 to 45 days from the return date and is enforced strictly.

Feed the loop back to buying and catalog

Disposition data is the only honest source of return reasons, because it records what the warehouse actually saw rather than what the customer selected from a dropdown. Map every graded unit to a reason taxonomy of no more than eight codes, and reconcile the warehouse code with the customer-stated reason monthly. A large gap between customer-stated size issues and warehouse-observed defects tells you the problem is in the size chart, not the factory.

Then rank SKUs by total return cost, not return rate. Return cost is reverse freight, plus grading and handling at roughly Rs 30 to Rs 60 per unit, plus the realisation gap between original price and disposition value. A SKU returning at 9 percent on a Rs 3,000 ticket usually costs far more than one returning at 22 percent on a Rs 400 ticket. Review the top 20 by cost every quarter with the buying team, and set a written sunset rule so the same items do not reappear in next season’s order.