Most conversations about returns start in the wrong place — with the item, and what is wrong with it. The more useful question is about sequence: given a unit has come back, what is the order of the exits available to it, and are you taking them in the order that recovers most?
The chain as it already exists
Every returns operation has a version of the same ladder, whether or not it is written down anywhere. A unit is checked in, graded, and then routed to the best exit it qualifies for. Sold as new if it never really left. Repackaged and discounted on your own site if it did. Sold as a mixed lot if it will not move individually. Liquidated if the lot will not move. Recycled if nothing else will take it.
That ladder is well understood and reasonably efficient, and each rung recovers less than the one above. What it lacks is a rung between "discounted on your own site" and "sold as a lot", which is precisely where most of the volume that frustrates people sits: units that are genuinely fine but that will not clear individually at a price worth having.
What each exit really costs
The recovery figure is only part of it. Every rung also has handling attached — shelf time, pick and pack, listing effort, photography, customer service on a discounted sale, and the carbon of a unit that gets moved several times before it settles. The lower rungs recover less and often cost more per unit of recovery, which is why they are exits of last resort rather than genuine channels.
A trial is an insertion, not a replacement
This is the part worth being precise about, because it is the objection every operations team raises first and it is a reasonable one. Listing a unit for trial does not commit it to anything. It is not consignment, it is not a sale to us, and it does not remove the unit from any route it was already on.
If the item does not sell it comes back to the same place it left, at the same grade, still eligible for every exit below.
So the trial rung is free to attempt. That is unusual in a returns chain, where most decisions are one-way: once a unit is in a clearance lot it is gone, and once it is discounted on your own site the price is public. Trying the highest-recovery option first has no downside beyond the time the unit spends listed, which it would otherwise spend on a shelf.
Who holds the stock
You do, or the reverse logistics provider you already use does. We do not run a warehouse, we do not take title, and we never take delivery of your stock. Title passes from you direct to the customer at the point they buy, and we collect as agent. Nothing sits on our balance sheet and nothing moves off yours.
This is not a positioning choice, it is the reason the model works operationally. The moment a marketplace starts holding inventory it needs to forecast, finance and store it, and those costs come back out of the recovery. Leaving the stock where it already is, with the people who already handle it, is what keeps the rate low enough to be worth taking.
Which means the integration is the product
What we build is the layer between your systems and a trial: stock and condition in, listings out, an order released when a customer passes their checks, a return booked when the trial ends, and settlement on what sold. We integrate with the APIs your operation already runs, or with your provider’s, and that work is ours to build and ours to maintain. Nobody is handed a specification and left to it.
For reverse logistics providers specifically, the point is that nothing about the existing operation changes. Stock is graded the way it is graded now, on the same benches, by the same people, against the same internal standard. We map that standard onto the three published consumer grades at the point of listing.
What happens to a unit that does not sell
It rejoins your flow at the grade it was listed at. The trial fee the customer paid has covered the round trip and the regrading, and 75% of that fee is yours. In practice the unit has spent a fortnight or a month out of your building, generating a fee, and has come back no worse than it went out — at which point every rung below the trial is exactly where it was.
It has also produced something the lower rungs never do: a reason. A liquidation lot tells you a pallet cleared. A trial that ends in a return tells you which item a real customer used for weeks and decided against, and what they said about it. Over a category that is genuinely useful demand data, and it arrives as a by-product rather than as a research cost.
Where this argument does not apply
It does not apply to units that should not be sold at all. Anything unsafe, incomplete in a way that cannot be put right, or genuinely faulty belongs on the disposal route, and inserting a rung above it does not change that.
It does not apply where your returns are overwhelmingly low value. The trial mechanism has a cost floor — a minimum trial fee of £5 — and below a certain retail price the arithmetic stops working for everybody, including the customer.
And it does not help with speed. If your constraint is clearing a warehouse by a date, a lot sale does that and a trial does not. A trial optimises for recovery per unit, not for how quickly a bay empties, and those are different problems that deserve different answers.