QC scorecard
Photos arrive, a decision is due, and the return window is not generous. Score what you can see, mark what you cannot, and get a reading that names the weak item rather than the whole order.
Reading
12 / 24
Treat as a loss
At this total the item is unlikely to be what the listing described. Weigh the return path against writing it off, and keep every photo either way.
| Stitching | 2 | |
|---|---|---|
| Label or tag | 2 | |
| Material or leather | 2 | |
| Colour | 2 | |
| Hardware | 2 | |
| Size tolerance | 2 |
The copied text is written to be pasted into a message to whoever is holding the items. It names the weak item rather than the whole order, which is the version that tends to get an answer.
Why these six items
The six items are the ones that decide most photo-based disputes: stitching, labelling, material, colour, hardware and size tolerance. Each has a failure mode that can be shown in a photograph, which matters because a claim you cannot show is a claim that goes nowhere.
They are also the six where a warehouse photo actually beats the listing photo. A listing photo is chosen by the seller and lit to flatter. A warehouse photo is taken quickly, from fixed angles, with no interest in how the item looks. That is exactly why it is useful: it is the least flattering image set you will ever get, and it is the one to judge from.
Where the thresholds come from
The bands used here assume a short return window and a platform that is no longer operating. In that situation the cost of a wrong "accept" is not just a bad item; it is a bad item plus the knowledge that you could have said something while the window was open. The bands are set accordingly, nudged towards raising a question rather than letting one pass.
A total of 19 or more with nothing scoring at or below 1 reads as accept. A total between 13 and 18, or any single item at 1 or below, reads as a request about that item. A total of 12 or under reads as a loss to weigh rather than a shipment to send. The bands are crude on purpose: a finer scale would imply a precision that six photographs cannot support.
The option that matters most
Every item has an "I cannot tell from the photos" option. Setting three or more of them stops the tool from producing a verdict at all, and it tells you to ask for the missing angles instead. That behaviour is the point of the tool rather than a limitation of it.
The failure this guards against is a confident decision built on a blurry photo. Scoring an unclear item at the middle of the scale feels neutral and is not: it pushes the total towards a verdict while contributing no evidence. Marking it as unjudged keeps the total honest and makes the missing information visible.
What this reading is not
It is not an authentication check, and it cannot become one. Nothing in a photo set establishes whether an item is authorised, and this ledger does not claim otherwise anywhere. Batch quality, seller reliability and price fairness are also outside what six photos can settle.
It is also not a substitute for the platform’s own return process. The reading is written to be pasted into a message because messages get answers; it does not itself open a case or change a deadline. Check what your account actually says about the window you are working inside.
Do this before you score
Run the photo set through the checklist first. Scoring items you have no photo for produces a low total that reflects your image set rather than the item, which is a misleading number to act on.
- The QC checklist — confirm the photo set covers the angles this scorecard assumes.
- The shipping weight estimate — if the item is going to ship, work out what the parcel costs before the window closes.
- What a QC photo can and cannot settle — the longer version of the limits section above.