When a return comes back marked “not as described,” the usual response is to check the size chart. Sometimes that's the right call. Often, the real problem is sitting in the product photo, not the sizing guide.
A product photo does more than show what something looks like. It sets an expectation. If a sweater's colour reads slightly warmer online than it does in person, a customer isn't just disappointed. They feel misled, even when nobody intended that. The return that follows gets logged as a sizing or quality issue, because that's the box on the return form. The actual cause never gets recorded anywhere.
Colour is the most common culprit, and the easiest to miss. A photo shot under one type of studio light, viewed later on a phone screen with different colour calibration, can shift a shade of navy toward black, or a cream toward yellow. Multiply that across a catalogue of thousands of images, and a brand can be quietly generating returns it has no way of tracing back to their source.
Texture is the second, quieter cause. Photos that flatten fabric detail, like a matte jumper that reads as flat and slightly plasticky, leave the customer with the wrong idea of what's arriving. So does over-smoothing that erases the texture entirely while trying to make an image look cleaner.
Return data usually gets sorted into a handful of standard categories: wrong size, changed mind, damaged in transit, item not as expected. Photo accuracy isn't one of them, so it never surfaces as a pattern, even when it's the actual driver behind a chunk of the “not as expected” pile.
The teams who do catch it usually find it by accident: someone finally compares the product photo to the item in hand, side by side, and sees the gap immediately.
Before assuming a return pattern is about fit, pull a handful of the returned items and compare them directly against their product photos. Do this on more than one screen: a laptop, a phone, and a colour-calibrated monitor if one is available. If the colour or texture reads noticeably differently across those three, that's worth taking seriously.
It's also worth checking whether the returned items cluster around a handful of specific colours or fabric types, rather than spreading evenly across the catalogue. A pattern that concentrates on, say, every product in one shade of blue points much more clearly at a photography or colour-calibration issue than sizing ever could.
This is also where a consistent retouching process pays off in a way that's easy to overlook. Our full breakdown of how an e-commerce retouching workflow actually runs covers exactly where colour accuracy gets checked, and why a consistent small team handling an account matters for catching this kind of drift before it reaches a customer.
How can I tell if returns are being caused by photo accuracy rather than sizing?
Compare a sample of returned items directly against their product photos on more than one screen. If the colour or texture reads differently, or returns cluster around specific colours rather than sizes, that's a signal worth investigating.
Does this only affect colour, or other things too?
Colour is the most common cause, but texture matters too. Over-smoothed or flattened images can misrepresent how a fabric actually looks and feels, which creates the same kind of mismatch.
Is this something retouching can actually fix, or is it a photography problem?
Both play a part. Accurate retouching, matching colour to a physical reference rather than just making an image look appealing, is what keeps a photo honest to the product, regardless of how it was originally shot.