Showing jewelry variants without confusing the buyer

Every other photography guide on this site is about making one piece look like itself. A variant set asks for something else. By the time a buyer reaches the options, they have already decided they want the piece; what is left is deciding which version, and the only question the images have to answer is which one is which.

That inverts the usual measure of a good frame. A variant photograph can be well exposed, well composed and attractive, and still fail completely, because the frame next to it is equally attractive and nothing visible separates the two. The content that matters is not inside either image. It is the difference between them.

So the set is the unit of work, not the photograph. Shooting one metal option carefully on Monday and the other carefully on Thursday produces two good photographs and one unusable comparison.

A comparison is readable when a single variable moves. If the surface, the camera distance, the angle, the crop and the light position all changed along with the metal, the buyer sees a difference they cannot attribute. They know the two listings are not the same. They cannot tell which part of that will be in the parcel.

The things to set once and leave alone are the surface, the camera position, the light position and size, the crop, and the orientation of the piece. This site has a separate guide on choosing a surface rather than a background; for a variant set the requirement is narrower than that argument — whatever surface you choose has to be the same one in every frame, including the frames you add months later.

That last part is where sets break. Options are rarely all made at once, so the newest frame is the one shot in different conditions. When an option is added, rebuild the setup for the set and judge the new frame beside the existing ones rather than on its own.

A polished metal surface has very little appearance of its own in the sense that matters here. What it shows is a picture of its surroundings: a small curved surface returns a miniature of the source lighting it, edge included, and a broad source spreads into a smooth gradient along the curve.

For a variant set, that has a direct consequence. Where the reflected source covers the part of the piece the buyer is looking at, both options return the same bright shape, and the tone that actually distinguishes them survives only around the edges of that reflection. Two options that are obviously different in the hand can then look nearly the same on screen.

The correction is not exposure. Raising or lowering brightness moves both frames in the same direction and leaves them just as similar. What changes the metal is what the metal is reflecting — the size of the source, its distance and where it sits relative to the piece. This is also the reason the surroundings must stay fixed across the set: the same adjustment that reveals the metal will invent a difference if it happens to only one frame.

When the options differ in lustre rather than in colour, the mechanism is different again. This site's pearl guide sets out why lustre is not brightness, and that page is the one to read before shooting a pearl option against a polished one.

Where a shop presents options as a row of small images, the choice is made before anything has been enlarged. A difference that only appears at full size is not available at the moment the buyer is choosing, however carefully it was captured.

So the test for a variant set is done at small size, not at full size. Reduce the frames, place them beside each other, and look at whether the difference is still there. If it is not, the problem is in the shoot rather than in the file.

How large the file itself is, and why a small frame costs you at the zoom step rather than at the thumbnail, is a separate question covered in the note on generated image resolution. Both matter, at different moments in the same visit.

Frame two sizes so that each fills the frame and you have made the same picture twice. Frame them from a fixed distance and the difference is present but slight, while every other quality of the two frames is now unequal, because the smaller item occupies less of the sensor and less of the buyer's attention.

There is also a technical cost to matching the crop. Moving closer to make the smaller item fill the frame raises the magnification, and depth of field narrows as magnification increases. The matched pair is now unmatched in what is sharp, which is a difference the buyer did not ask about and cannot interpret.

The honest answer is that size belongs in the listing text. Photographs should not be asked to carry it and should not be edited to suggest it. If a reference object is included, it has to be the same object, in the same place, in every frame of the set — otherwise it becomes another thing that moved.