Size inclusivity

Offering sizes is not the same as supporting them

The distinction that matters is between extending a size list and extending a product. Grading a pattern up mathematically produces garments that are technically the stated size and frequently fit badly, because bodies do not scale uniformly — proportions change, not just measurements.

Doing it properly touches most of the development process, which is why it is a real commitment rather than a merchandising toggle.

• Grading — proportional adjustment rather than uniform scaling
• Fit sessions — a fit model at the upper end of the range, not only at sample size
• Construction — support, seaming and fabric weight that work at scale
• Pricing — charging more for larger sizes is a visible and frequently criticised decision
• Imagery — the range shown on bodies within it

Why imagery is the most common gap

It is the step with the clearest economics against it. Showing a garment on four body types means casting and booking four models for the same SKU, which multiplies the most expensive line item in a shoot — so most brands photograph the sample size and let the size chart carry everything else.

The consequence for the shopper is concrete. A buyer at the top of the range is looking at a garment on a body unlike theirs and guessing how it translates, which is the single most reliable driver of size-related returns in apparel. The size chart answers what the measurements are; it does not answer how it will look.

What actually helps a shopper decide

The useful signals are all about removing guesswork rather than about reassurance. Measurements alone are necessary but not sufficient, because most shoppers do not know their own garment measurements and are translating from a size they usually wear.

• Garment measurements, not only size labels — and stated as garment or body measurements, explicitly
• Model reference — the model's height and the size they are wearing, on every image
• The same garment at more than one size, which is the signal a size chart cannot give
• Fit intent — whether a style is meant to be close, relaxed or oversized
• Fabric behaviour — stretch and recovery, which change how a size tolerance feels

Where imagery cannot help, and where it can mislead

No image is a fit prediction. Showing a garment on a range of bodies reduces guesswork about appearance; it says nothing about whether a specific shopper's measurements will work, and presenting rendered imagery as sizing guidance is the way this goes wrong.

There is also a credibility trap. Generating diverse bodies while the underlying garment has not actually been graded or fit-tested for them produces images that promise a fit the product does not deliver — which is worse than not showing it, because the return comes with a broken expectation attached. Imagery is the last step of size inclusivity, not a substitute for the development work.

• An image is not a measurement — sizing guidance belongs in the size chart
• Showing sizes you have not graded or fit-tested sets up a return
• Rendered bodies must be plausible for the garment's actual construction
• Disclosure rules for synthetic imagery of people apply here too — see AI fashion model

Common questions

See also

Doing this with TinTin