Content creation: photography is not a talent, it is a process

The article on social media management names the photograph as the bottleneck of that whole service, and stops there. This one takes over, and it begins with the single idea that makes the subject manageable.

Product photography is not a talent you either have or do not. It is a process: a fixed place, a fixed light source, a fixed background, a set of angles decided once. What decides the result is the repetition, not the eye.

That has a direct consequence for who should do it. A process is installed once and then carried out by anybody conscientious, including whoever tidies the shop in the evening. A talent, on the other hand, is rented by the month.

This article sets out that process, the part that costs money when it goes wrong — the misjudged size and the returned parcel — and the part nobody ever mentions: where the files are kept, and who owns them.

Daylight from a window is better than anything you could buy, and it is free. The only constraint is not to mix it with anything else, and that is the constraint people break without thinking because the strip lights are already on.

Switch off the shop’s strip lights while photographing. Mixed lighting — daylight and fluorescent — produces colours that cannot be corrected and that misrepresent the product’s real shade, which is precisely the information the customer is after.

Place the object beside the window rather than in front of it, so the light arrives from the side. An object lit from the front is flat; one lit from the side shows its relief, its stitching, its grain — which remains the only way of saying it is well made.

If the shadow side is too dark, hold a sheet of white paper opposite the window to bounce a little light back. It is the only accessory this process requires, and it is already in your printer.

Finally, always photograph at the same hour. Ten o’clock light and four o’clock light do not give the same colours, and a grid where the same shirt is beige and then grey is a grid that worries instead of reassuring.

Here is the part that costs money, and it is nearly always absent: nothing in a photograph states the size of the object, and the buyer guesses wrong. Not through carelessness — an image contains no dimensional information at all, and they had nothing else to judge by.

The mechanism is easy to describe. A sharp, well-framed image contains no indication of scale. The customer imagines a size, orders, opens the parcel, and finds something else. They refuse it, and the cost of the refusal is entirely yours: two journeys and an item that comes back handled.

Three corrections, chosen by product. A familiar object placed beside it — a coin, a pen, a hand. The object worn or in use, which settles the question without anyone thinking about it. And the dimensions written into the image itself rather than in the caption, because the caption is not always read.

For clothing, add the one piece of information that genuinely matters: the garment’s real measurements laid flat, not the size on the label. The article on the clothing sector explains why a size is a bet and a measurement is data.

This is not an aesthetic refinement. On the catalogues we have seen correct it, the first consequence was not more orders: it was fewer orders refused on delivery, which is better.

Nobody writes about this and it nonetheless decides whether your image library is an asset or a heap. A photograph you cannot find is a photograph you take again, and you pay twice for the same object.

The naming that works is dull and fits in one rule: item reference, then view number. No date, no platform name, no accents, no spaces. Somebody looking for an item in two years has to be able to type its reference and find its four views.

Filing follows the same logic: one folder per product family, and inside it one folder per reference. Filing by month or by campaign feels natural at the moment you do it and becomes unusable within six months, because nobody remembers the date of a photograph.

Always keep the original files, uncropped and uncompressed, beside the published versions. On the day a banner, a poster or a print is needed, an image resized for a social network will not do, and the object may no longer be in stock.

Finally, back up somewhere other than the phone that took the photographs. It is the most banal advice in this article and the one we have seen missing at the greatest cost: a lost phone sometimes takes two years of catalogue with it.

What we will refuse: selling monthly photo sessions to a business that can install the process itself. It is our easiest line to bill and it makes the client dependent on an appointment, when their products arrive every week.

We will refuse to publish manufacturer images in place of yours for a product you hold in stock, and to deliver a catalogue with no indication of scale. The first misrepresents the exact object, the second is paid for in refused parcels.

We will also refuse to keep the original files. They go to you, named and filed, with the uncompressed versions — and if that makes us easier to replace, that is the right consequence.

What we do: installing the process in your premises in one morning — the corner, the surface, the hour, the card of angles; the first session run with your team rather than instead of it; the naming and folder structure decided at the outset; and scale present on every product whose size could be misjudged.

And what you should do without us this week: choose the corner, buy or find a plain surface, and photograph six items with four views each, at the same hour and without flash. Compare them with your last six posts. You will know immediately whether to call us or simply carry on.