Product photography is budgeted as a marketing expense and behaves as a logistics variable. What the image promises determines how often the delivered item disappoints.
The photograph is the only inspection the buyer gets
In a shop a customer handles the item, judges its weight and sees its true colour. Online, every one of those judgements is made from an image.
Whatever the image implies becomes the expectation, and the return decision is made by comparing the arriving object against that expectation rather than against the description.
This makes photographic accuracy a commercial variable with a direct cost attached, not a question of aesthetic preference.
Colour is the most common and least controllable mismatch
Colour depends on lighting during the shoot, on processing afterwards and on the screen the buyer is using, and only the first two are within the retailer's control.
Shots optimised to look appealing tend to be lit in ways that shift colour, so the most attractive image is frequently the least faithful one.
Retailers that standardise lighting and processing accept slightly duller images and see fewer returns in the categories where colour matters most.
Scale is misjudged whenever there is no reference
An item photographed alone against plain background carries no size information. Buyers infer scale from category expectations, and those expectations are frequently wrong.
Including a familiar object, a hand or a room context lets the buyer calibrate, which resolves a whole class of returns that dimensions in a table do not prevent.
Dimensions are supplied in most listings and consulted by few buyers, because reading a measurement and picturing a size are different cognitive tasks.
Styled images sell the setting rather than the item
Photography that places a product in an aspirational scene raises engagement, and it also attaches expectations to the item that the item itself cannot meet.
A buyer who liked the scene receives only the object, and the gap between the two is experienced as the product being disappointing.
The effect is strongest in home and furniture categories, where the return is expensive and the styling temptation is greatest.
Return reasons are the cheapest available feedback
Reason codes collected at return, particularly free-text ones, point directly at which listings are creating false expectations and in what way.
Most retailers collect this data and route it to customer service, where it is used to process the individual case and never reaches the people who produce images.
Connecting those two functions turns an existing data stream into a specific list of listings to reshoot, which is among the cheapest improvements available.