The 3D commerce procurement narrative, in 2026 as in 2022, presents the technology as a general-purpose conversion lever applicable across retail categories with broadly equivalent benefits. The narrative is supported by a regular cadence of vendor-published case studies in which a deployment of 3D commerce technology produced conversion uplift of some headline figure — twenty-five per cent, forty per cent, sometimes considerably higher — on a specific catalogue. The case studies are, in their isolated form, accurate; the conversion uplifts they describe are real, in the catalogues to which they refer.
The procurement narrative’s mistake is the implicit generalisation from the case studies to the broader retail landscape. The case studies are, almost without exception, drawn from a small set of categories whose properties make them particularly suited to the technology; the categories outside this set produce conversion uplifts that are, on the deployments I have shipped, considerably smaller than the case studies imply. The procurement decisions that proceed on the basis of the headline figures, against catalogues outside the suitable set, produce projects whose costs exceed the conversion benefits by margins that are, in many cases, decisive for the project’s commercial viability. This post is an account of which categories are genuinely suited to 3D commerce, which are not, and what the distinguishing properties are.
The categories that genuinely benefit
The conversion uplift from 3D commerce, on the data I have collected across approximately fifteen 3D-commerce deployments, is concentrated in catalogues whose products satisfy three properties in combination.
The first property is high purchase consideration. The product is, in the visitor’s purchase context, a considered purchase rather than an impulse purchase; the visitor is willing to invest several minutes of session time in the purchase decision and is, accordingly, willing to engage with a 3D representation of the product to support the decision. Furniture, automotive accessories, footwear in higher price points, jewellery, and large-format homewares all fit this property; fast-fashion apparel, food and beverage, and the great majority of small-format consumables do not.
The second property is product differentiation that is principally visual. The visitor’s purchase decision turns, in non-trivial part, on aspects of the product’s appearance that are difficult to communicate via static photography — the way a sofa looks from the angle the visitor’s living room would view it from, the way a watch face appears under varying lighting conditions, the way the colour of a particular fabric responds to environmental change. Product differentiation that is principally functional, principally driven by brand, or principally driven by price does not produce the same uplift from 3D representation; the visitor’s decision in those cases is not constrained by the photographic limitations the 3D representation is addressing.
The third property is sufficient catalogue scale that the per-SKU cost of producing the 3D assets can be amortised across enough orders to recover the cost. The catalogues that satisfy this property are, in most cases, the catalogues with at least several hundred SKUs; smaller catalogues produce per-order asset costs that are, on the implementation pricing I have observed, prohibitive even for products that satisfy the other two properties.
Catalogues that satisfy all three properties produce conversion uplifts from 3D commerce that are, on the data, broadly consistent with the case-study figures the procurement materials present. Catalogues that satisfy two of the three produce uplifts that are smaller but still positive. Catalogues that satisfy only one produce uplifts that are, on most deployments, indistinguishable from zero.
The categories that do not benefit
The categories outside the three-properties intersection are, in 2026, considerably more numerous than the procurement narrative implies. The fast-fashion apparel category, despite its substantial SKU count, fails the considered-purchase property; the visitor in fast-fashion makes purchase decisions in seconds rather than minutes, and the 3D representation does not, on the conversion data, alter the decision pattern in any measurable way. The food-and-beverage category fails the visual-differentiation property; the visitor’s purchase decision in this category is principally driven by brand and price, with the visual representation of the product playing a minor role. The single-product or very-small-catalogue categories fail the scale property; the per-SKU asset cost cannot be amortised across enough orders to recover the implementation investment.
The procurement decisions that have proceeded against these categories, against the procurement narrative’s general endorsement, have produced projects whose cost was, in retrospect, not commercially defensible. The pattern is, in my observation, sufficiently consistent that the procurement narrative’s continued promotion of 3D commerce as a general-purpose intervention is doing, in commercial terms, more harm than good; the retailers who would benefit from the technology continue to benefit, while a substantial fraction of the procurement budget is, on the data, being committed against catalogues for which the benefit is not present.
Why the case studies are misleading
The procurement materials’ case studies are, as noted, accurate descriptions of specific deployments; the case studies are nevertheless misleading as a general guide to the technology’s applicability because the case studies are, almost without exception, drawn from the suitable set. The vendors publishing the case studies select for the deployments whose figures support the procurement narrative; the deployments whose figures do not support the narrative are not, by hypothesis, included in the case-study collection.
The selection bias is not, in any moral sense, dishonest; the vendors are presenting their best work, and the practice of presenting best work is universal in commercial procurement materials. The bias is, however, structurally consequential because the materials do not, in their dominant presentation, indicate the conditions under which the headline figures are achievable. The retailer reading the materials is invited to project the figures onto the retailer’s own catalogue without reference to whether the catalogue satisfies the underlying conditions; the projection is, for catalogues outside the suitable set, substantially incorrect.
What the alternative looks like
For retailers whose catalogues do not satisfy the three properties, the alternative to 3D commerce procurement is, in operational terms, the continued investment in conventional product photography. The conventional photography has, in the period of the 3D-commerce hype, been somewhat underinvested-in by the broader retail community on the assumption that 3D would supersede it; the photography work that has been completed in the period continues to perform broadly as it always has, and the catalogues whose photography is competent continue to convert at rates that the 3D commerce uplift, where applicable, is built on top of rather than replacing.
The recommendation that has emerged from the analysis is approximately the following. For catalogues in the suitable set, 3D commerce procurement is, on the data, justifiable and produces the case-study-consistent uplifts the procurement materials claim. For catalogues outside the suitable set, the procurement budget is, on the data, more productively allocated to the conventional product photography that the catalogues actually depend on; the photography produces measurable improvement against the catalogues’ existing baseline, while the 3D commerce against the same catalogues produces, in many cases, no measurable improvement at all.
(For the practitioner view of how 3D commerce, when correctly deployed, is implemented at a scale that justifies the procurement, see Configurators that pay for themselves and Building a 3D product configurator.)
An advisory close
The 3D commerce procurement narrative is, in 2026, presenting a real opportunity in a manner that systematically overstates its applicability. The retailers who proceed on the basis of the narrative without examining their catalogue’s fit against the underlying conditions are, on the data, the retailers most likely to find the procurement investment unrecoverable; the retailers who first examine the fit and proceed only against catalogues that satisfy the three properties are, on the same data, the retailers most likely to recover the procurement investment with the case-study-consistent margin to spare.
It is recommended that retailers evaluating 3D commerce procurement first audit their catalogue against the three suitability properties — high purchase consideration, principally visual differentiation, and sufficient scale to amortise the per-SKU asset cost. The audit takes minutes against the existing catalogue data and is, in my experience, decisive for whether the procurement will produce the conversion outcomes the materials project or the unrecoverable cost the procurement narrative does not warn against.