3D for home decor: when a lamp needs a model and when a photo is enough

Home decor is an awkward category for 3D. The products are cheap, there are thousands of them, and a good photograph already does most of the job. Nobody needs to place a 20 euro vase in their living room in augmented reality before buying it. At the same time, some decor products are returned constantly for reasons a photograph causes, and those are worth modelling immediately.

The useful question is not whether 3D works for home decor. It is which decor products earn a model. Here is the test we use with clients.

The test: does the photo leave a question that costs money

Run each product line through three questions.

One. Does the shopper misjudge the size? If the answer is yes, a model earns its cost, because scale is the thing a 3D model fixes and a photograph cannot. Rugs are the clearest example in the whole category. A rug photographed in a styled room looks like whatever the room implies, and a 160 by 230 rug arriving in a living room is routinely a surprise in both directions. Pendant lamps are next: the difference between a 30 centimetre shade and a 50 centimetre shade over a dining table is the difference between right and comical, and no photograph communicates it.

Two. Does the material carry the purchase? Ceramic glaze, mouth-blown glass, brushed brass, wool pile. These read very differently to the eye under changing light, and a still photograph freezes one lighting condition. A model the shopper can rotate shows them how a glaze catches light, which is often the actual reason they wanted the object.

Three. Are there many variants? Twelve colourways of the same cushion means twelve photo shoots, or one model and twelve materials. This is where 3D pays on cost rather than on conversion, and it is the argument that usually convinces a finance director.

If a product line scores no on all three, a good photograph is enough. Say so and move on. The fastest way to waste a 3D budget in home decor is to model the whole catalogue evenly.

Where the money actually is

The pattern is that 3D pays in decor wherever the object interacts with the architecture of the room. Things that sit on a table are fine as photographs. Things that go on a floor, a wall or a ceiling are dimensional decisions, and dimensional decisions are what a model is for.

The production cost argument, which is the strong one

For a furniture brand, the case for 3D is usually conversion and returns. For home decor, it is frequently photography.

A decor catalogue refreshes every season. Each new colourway needs product shots and at least one styled lifestyle image. That is a studio day, a stylist, a photographer and retouching, repeated every season for every variant. A 3D model is built once and then recoloured, relit and rephotographed inside software, at close to zero marginal cost per variant.

This changes the arithmetic completely for a brand with a lot of variants and a fast refresh cycle. The model does not have to lift conversion at all to pay for itself, because it replaced a line item that already existed. Any conversion effect is then upside.

It also changes the timeline. A new colourway can have imagery before it has physical stock, which matters for pre-orders and for getting the product live on launch day rather than three weeks later.

What still has to be right

The same rules apply as for large furniture, and in one respect decor is harder.

Scale must be exact, because scale is the only reason the shopper is using the model. This is not negotiable and it is the first thing to check on delivery.

Materials have to survive a close-up, because decor shoppers zoom. A ceramic glaze needs the slight unevenness real glaze has. Wool pile needs depth, or a rug reads as printed vinyl. Metal needs the right sheen. This is the part that takes the time, and the part where cheap modelling shows.

Lighting products have an extra requirement, which people forget. A lamp is bought partly for the light it gives. A model that shows the fixture but not the quality of light it throws has answered half the question. This can be handled with renders of the lamp lit in a room, produced from the same model.

What the evidence says

The published figures for 3D and AR content are not decor-specific, but they point the same way. Shopify reported in 2020 that products with 3D or AR content converted 94 percent more often than comparable products without. Its case material includes Rebecca Minkoff, where shoppers viewing a product as a 3D model were 44 percent more likely to add it to cart, and Gunner Kennels, which cut return rates by 5 percent. Against a market where the NRF put the online return rate at 17.6 percent of sales in 2023, a few points of return reduction is a real number, and for bulky decor like rugs it is a large one.

A sensible first step

Take your worst return rate in decor. It is probably rugs, and if it is not rugs it is lighting. Model five products in that line, leave the rest of the catalogue alone, and compare return rates over a quarter against the same period last year. That is a clean experiment and it costs very little.

We will do the first one for nothing. Send us a product with its dimensions and some close-up material photos, and you get the model to keep. The solutions page explains what the service covers and the prices page shows the monthly cost per product once you decide to carry on.

Sources

• Shopify, merchant data on 3D and AR content, September 2020.
• Shopify, AR shopping: Rebecca Minkoff and Gunner Kennels results.
• National Retail Federation and Appriss Retail, 2023 Consumer Returns in the Retail Industry.