Returns: domestic return vs international return vs abandoning

This is the cheapest branch and it has a closing date. A return is usually workable while the goods are still at the consolidation warehouse, because the domestic leg back to the seller is short, cheap relative to an international leg, and entirely inside one customs territory. The option set shrinks the moment the international freight is paid, and nothing reopens it, which is why the return decision belongs to the inspection window rather than to the delivery week. The window is not the same length everywhere, so read the period the desk publishes rather than assuming one, and note that it is counted from intake rather than from the day you placed the order.

The mechanics go in one order. Raise it with the desk while the item is still in the warehouse, with the inspection photographs attached and a plain description of the discrepancy between the listing and the item. Ask for the return address and the refund route before anything moves, because a parcel sent to an unconfirmed address is no longer your item and not yet the seller's. Then decide the form: full return for refund, exchange for another size, or a partial refund with you keeping the item. What goes into the first message decides how long the branch takes. Send the inspection photographs, a description of the mismatch that a stranger could verify, the listing text as it stood when you ordered, and the outcome you are asking for. A message that names the outcome gets an answer rather than a question, and on a branch with a closing date, one round trip saved is often the difference between a return and a write-off.

That third form is underrated and settles faster than either of the others. Where the discrepancy is cosmetic or small, a partial refund costs the seller less than a restocking cycle and costs you nothing in freight, and both sides end the exchange having given up less than a return would have consumed. Where the seller offers it and the item is usable, taking it is usually the better trade even when a full return is technically available.

The costs to compare are the domestic return freight, any restocking deduction, any handling line the agent raises for processing a return, and the time. The time is the item people forget. A return that takes three weeks to settle has also frozen the value of the item for three weeks, and on a purchase you intend to replace immediately, that freeze can cost more than the freight.

Refusals are not all bad faith. Sellers decline returns for reasons that are legitimate, including buyer-selected size errors, items marked as made to order, and discrepancies that fall inside the tolerance the listing stated. Before escalating, compare the listing text with what arrived, because a dispute built on a description that never promised the thing you wanted is a dispute that will fail slowly.

Where the listing did promise it, the dispute path runs through the platform rather than through the seller, and it rewards evidence over argument. What the platform will want is the listing as it stood at purchase, the inspection photographs, the dates, and a description of the mismatch that a stranger can check. What it will not weigh is the seller's tone or yours, which is worth remembering before writing a long message.

Outcomes in this branch cluster into three. A full refund with return, which resolves the same way as script one and costs you the domestic leg. A partial refund without return, which is the most common resolution and usually the fastest. Or no refund, in which case the item is yours and the useful question changes from recovery to salvage: resell locally, use it, gift it, or write it off.

Writing it off deserves a straight sentence rather than a euphemism. For a low-value bulky item, the domestic return freight plus the handling line plus three weeks of attention can exceed the amount in dispute, and the rational move is to stop. Abandoning is not the same as failing to pursue a claim you would win; it is declining to spend a large fixed cost to recover a small variable one.

Once the parcel has left, the return becomes an import in reverse, and the costs change category rather than size. International return freight is charged on the same chargeable-weight logic as the outbound leg. The destination may require an export declaration or a formal re-export entry for goods that already cleared. And the original seller frequently cannot receive an international parcel at all, since the address that accepted a domestic shipment is not set up to clear one.

The arithmetic rarely survives contact. When the round-trip freight, the re-export paperwork and the seller's restocking deduction are added together, they consume most or all of the item value on anything but a genuinely expensive single item, and they do it in weeks rather than days. Our own observation, from the four return cases we followed to settlement, is that the case which reached a foreign address recovered nothing net of freight.

What replaces the return in this branch is local disposition. Resell it where you are, which recovers a fraction of the cost without any freight; repair it if the defect is repairable and the repair is cheaper than the loss; keep it and use it; or pass it on. None of these are elegant, and all of them beat paying international freight twice so that a seller can decide whether to accept a parcel they may not be able to receive.

One exception is worth acknowledging. Where the item is high value, the defect is material, and the seller has agreed in writing to accept the international return and to refund in full, the arithmetic can work. The condition is the written agreement, obtained before the parcel moves, because the entire branch depends on a counterparty who has committed to receiving something they have no obligation to receive. What that agreement needs to state is specific: who pays the return freight, which address will receive the parcel, what condition triggers the refund, and when the refund is issued relative to receipt. An agreement missing any one of the four is not an agreement, and the gap is normally discovered while the parcel is already travelling in the wrong direction.

Recovery is the first axis and it runs in a clear order. Domestic returns resolved at the warehouse recovered the most in the cases we followed, with the seller-refusal branch recovering less and the abroad branch recovering nothing after freight. The orders of magnitude are what matters here rather than the specific figures, which came from four cases and are too few to present as a distribution.

Time is the second axis and it runs in the same direction for a different reason. A warehouse-stage return settles in a timeframe measured in days to a couple of weeks, a dispute in weeks, and an international return in weeks to months because it involves a second clearance. Since the money at stake is usually modest, the time cost frequently decides the comparison on its own. Time is not only waiting, either. A return in progress occupies attention, and attention spent on a small recovery is attention not spent on the next order, which is a real cost that never appears on any ledger. That is also the argument for batching: where one inspection turns up three defects, raise them as one case with three items rather than three cases, because the desk work, the messaging and the follow-up are then paid once instead of three times.

Failure modes are the third and the least discussed. The domestic branch fails when the return address is wrong or the seller disputes the condition. The refusal branch fails when the evidence is thin or the listing was ambiguous. The abroad branch fails at the point where the seller cannot receive the parcel, which strands the item in a third place with freight already paid. Measuring the branches by recovery alone hides the fact that the abroad branch can end with neither the item nor the refund nor the money.

Which reduces to one rule worth carrying: make the return decision at inspection, while the goods are in the warehouse and the options are all still open. If the decision cannot be made there, the default is to keep the item and settle later, because keeping preserves the option to resell while returning destroys it. Where the comparison is genuinely close, model it against the item value rather than reasoning from a feeling about fairness. The second half of the rule is about timing. Costs decided at inspection are cheap and costs decided after departure are not, and the gap between those two states is usually a single day of attention, which is why we treat the inspection window as a decision deadline rather than as a review period.