Why Returns Disposition Decides Inventory Trust in Apparel

A returns trailer arrives at a 3PL on a Tuesday afternoon during a drop week. Two hundred forty units come off, most from a DTC promotion, about forty from a wholesale account that shipped short and got the balance back as a refusal. The receiving team scans the outer cartons, marks the ASN closed, and stages the totes near the QC bench. By Friday, half the units are still in totes. Some have been inspected but not dispositioned. A handful were quietly returned to pick locations without QC. The ops manager pulls an inventory report on Monday morning and cannot tell you, within a thousand units, what is actually available to sell.

That is the problem. Not the receiving speed. The disposition workflow.

What is a returns disposition workflow in apparel, and why does it decide inventory trust?

A returns disposition workflow apparel operations can trust is the sequence a returned unit moves through from the moment a carrier scan hits your system to the moment that unit has a final status recorded against a specific inventory location. The statuses that matter are resellable to A-grade stock, hold for QC, repair or steam and re-tag, B-grade or outlet channel, donation, and destroy. Each status has to post to inventory in the correct bin, in the correct location, at the correct valuation, and it has to do that in days rather than weeks.

Inventory trust is the confidence that the number your system shows as available to sell matches the number you can actually pick, pack, and ship today. Returns disposition is where that trust breaks first in apparel, because apparel returns are high volume, size and condition variable, and often seasonal. A jacket returned in March is not the same asset as a jacket returned in September, and a top returned with deodorant marks is not the same asset as a top returned with tags still on.

This is Breakpoint 5 territory in the 6 Breakpoints framework. Warehouse execution gets less predictable, and the 3PL blind spot is where finance stops believing the inventory number. Disposition is the specific mechanic inside BP5 that most brands underinvest in.

Why does receiving speed get the attention while disposition gets ignored?

Because receiving is measurable and disposition is judgmental. A 3PL scorecard usually tracks receiving dock-to-stock time on inbound POs, and returns get treated as an inbound flow with the same metric. Trailer arrived, units scanned, dock cleared. Green.

But a returned unit is not a PO receipt. A PO receipt is one SKU, one condition, one destination bin. A returned unit is one SKU, unknown condition, and five possible destinations. The scan tells you the SKU. It does not tell you whether the garment can be sold again, and it does not tell you where the unit should physically go. That decision requires eyes on the garment and a rule that says what to do next.

When I sit in on a customer kickoff, one of the first questions I ask the warehouse lead is how many days between a return arriving on dock and that unit being either available to sell or written off. The answers I hear most often are between seven and twenty-one days at a 3PL, and between three and ten days in a brand-run warehouse. In both cases, the units in flight during that window are invisible to the sell side of the business. Merchandising cannot allocate them. DTC cannot show them as in stock. Wholesale cannot promise them to an account. They are inventory, they belong to you, and they do not exist in your ATS.

What actually happens inside a broken disposition workflow?

The pattern across the implementations my team has shipped is remarkably consistent. There are three failure modes, and most brands have all three running at once.

The first is the limbo bin. Returned units sit in totes near QC, not received into any sellable location. They exist physically. They do not exist in the system. A $15M brand running wholesale, DTC, and 3PL will typically have six to nine hours a week of someone reconciling inventory across Shopify, the 3PL WMS, and wholesale commitments. A meaningful share of that reconciliation time is chasing units that are physically in the building but not in the ATS.

The second is the premature receive. To close the dock ticket faster, receiving posts units back to available stock before QC has looked at them. The DTC site immediately sells them. The pick ticket comes down. The picker finds a damaged garment in the bin, or no garment at all because it was still in the return tote. Now the customer gets a cancellation, a refund, and a reason to lose trust. Oversell rate at peak for the profile I described tends to sit at two to three percent, and premature receives on returns are a real contributor to that number.

The third is the silent write-off. Damaged units get pushed to a destroy pile that nobody counts, nobody values, and nobody reports on. Finance sees a shrink number at year-end and cannot tell you what it is composed of. Merchandising cannot tell you which styles returned damaged at a rate that suggests a fit or fabrication issue. The data is gone before it was captured.

What does a working disposition workflow look like?

A working workflow has a defined ladder with an owner, an SLA, and an inventory posting for each rung.

Rung one is dock scan. Owner is receiving. SLA is same day. Inventory posting is to a returns hold location, not to available stock. The unit now exists in the system as on hand, not available.

Rung two is QC and grading. Owner is QC. SLA is 48 hours from dock scan during normal periods, 72 hours during peak. Grading uses a fixed rubric, not judgment. A-grade means tags on or off but garment is in resellable condition with no marks, no damage, no odor. B-grade means minor cosmetic issue that would not clear DTC standards but is acceptable in an outlet or sample channel. Repair means the unit needs steam, re-tag, or minor mend before it can move to A-grade. Destroy means the unit cannot be sold anywhere and needs to be written off.

Rung three is the inventory move. Owner is the WMS. This has to be automatic based on the QC grade. A-grade posts to available stock in the DTC location. B-grade posts to a separate location tied to the outlet channel. Repair posts to a work-in-process location that is not visible to sell channels. Destroy posts to a write-off account with a reason code and a cost value.

Rung four is the exception review. Owner is the ops manager. Cadence is weekly. Anything sitting in returns hold longer than the SLA gets a name against it and a decision by Friday. Nothing sits for a month.

The test of whether this workflow is working is a single query. Show me every returned unit received in the last thirty days, its current disposition status, its current location, and its current inventory posting. If that query takes more than five minutes to run and returns clean data, you have a workflow. If it takes a day and comes back with gaps, you do not.

How does disposition connect to what wholesale and DTC can actually promise?

This is where the inventory trust question gets sharp. A brand running wholesale and DTC on the same pool of stock has to decide what percentage of on-hand inventory is committed to open wholesale orders, and what percentage is genuinely available to DTC. If returns hold is not visible in that calculation, the ATS number is wrong in two directions at once. Units that are physically in the building but stuck in hold are missing from ATS, so DTC underpromises. Units that got prematurely received back to available stock inflate ATS, so DTC overpromises. Both errors happen in the same warehouse, on the same day, on the same SKU.

Magnolia Pearl is a useful reference here. Their operation runs drops, same-day fulfillment, and a global returns flow that includes international duties reconciliation. Before they moved to a connected system, reconciliation across Shopify, the warehouse, and wholesale was consuming a full FTE’s worth of time each week. After the rebuild, reconciliation time dropped by roughly two thirds and oversell rate stayed under half a percent through peak. The mechanic that made that possible was not faster receiving. It was a disposition ladder that posted each returned unit to a specific location with a specific status, so the ATS calculation had no phantom units and no missing units.

When should returns post to available stock, and when should they never?

Here is the point of view. Returns should post to inventory in days, not weeks, and they should never post to available stock before QC has graded them. Those two rules sound obvious and they are violated constantly, because the receiving team is measured on dock clearance and the QC team is measured on nothing in particular. The fix is not a new system. It is moving the measurement. QC owns the SLA on grading, and receiving cannot close the returns line until QC has signed off.

The second point of view is about B-grade and outlet. Most brands treat B-grade as a garbage bin, which means B-grade inventory sits invisible until someone remembers it exists, usually in January when finance asks about shrink. B-grade should be its own channel with its own ATS and its own allocation rules. If you cannot see B-grade inventory in your reporting the same way you see A-grade, you are not managing it. You are hoping it goes away.

What does a healthy returns disposition metric set look like?

There are four numbers worth tracking weekly, not monthly.

First, average days from dock scan to final disposition. Target under five during normal periods, under seven at peak. If this number is drifting, disposition is falling behind receiving.

Second, percentage of returned units still in returns hold longer than SLA. Target under five percent. This is the phantom inventory metric. Everything on this list is a unit that exists physically and does not exist in your ATS.

Third, disposition grade mix. What percentage of returns are grading A, B, repair, and destroy. Watch this by style and by season. A style that returns at a fifteen percent B-grade rate has a fabric or a fit problem, and merchandising needs to know before the next buy.

Fourth, oversell rate on styles that had a return in the last thirty days. If this is materially higher than oversell on styles with no recent returns, you have premature receives.

None of these numbers require a new tool. They require a system that treats returns as its own workflow with its own inventory postings, not as a variant of PO receiving.

What this means for an apparel operations team

Returns disposition is not a warehouse problem. It is an inventory truth problem that shows up in the warehouse. The reason it lives at Breakpoint 5 in the framework is that the symptoms surface in warehouse execution, but the cost is felt in merchandising, in wholesale allocation, in DTC customer experience, and in finance shrink reporting.

The first ninety days after a system go-live is the window where disposition either becomes a real workflow with owners and SLAs, or gets deferred as a phase two problem. My experience is that phase two never comes. If disposition is not designed into the initial rollout, brands live with the limbo bin, the premature receive, and the silent write-off for another year, and they explain the resulting inventory noise as 3PL flakiness.

The practical next step for an ops team is small. Pull a list of every returned unit received in the last thirty days. Ask the warehouse where each unit is and what its status is. If more than five percent of the units cannot be answered cleanly, disposition is your work for the quarter, and receiving speed is not.