Garment measurements vs body measurements

A garment measurement describes a piece of clothing. A body measurement describes you. Garment-first sizing starts from a shirt you already own and looks for other garments whose published numbers sit close to it. Body-first sizing starts from your chest, waist and height and predicts which size label should fit.

Both schools are legitimate. They fail in different places, and most of this page is about where each one fails, because that is the part the marketing on either side tends to skip.

What is the difference between garment measurements and body measurements?

A garment measurement is taken from the clothing, laid flat on a table. Chest width runs from armpit seam to armpit seam. Body length runs from the high point of the shoulder down to the hem. Sleeve length runs along the sleeve seam. Each of those is a fact about fabric, and because the garment lies still on the table, two people measuring the same shirt will usually land on the same figure. How to measure a shirt covers the conventions, and pit to pit is the single most useful of them.

A body measurement is taken from a person, usually as a circumference: chest, waist, hip, inseam. It is a fact about you, and it is harder to take consistently. Tape tension, arm position and whether you have just breathed in can move a chest reading by more than the difference between two adjacent sizes.

The two do not convert into one another cleanly. A flat chest of 20.0" doubles to a 40" garment circumference, but nobody designs a shirt to sit on a 40" chest with nothing to spare. The space between body and garment is ease, and ease is a design decision the brand makes. A boxy heavyweight tee and a trim retail tee can publish the same flat chest and wear as completely different clothes. Turning a body measurement into a size therefore rests on an assumption the brand rarely writes down.

Why does the size label not settle the question?

Because a letter is a name, not a measurement. Below are published chest specifications for five blank styles, all measured flat, all taken from the manufacturers' own spec sheets.

The finding that matters is the last row. A ladies' cut Gildan G500L Medium publishes a 17.5" chest. The unisex Mediums above it publish 20.0" to 20.5". That is a three inch gap on the same letter, and it widens to 3.5" at Large, where 19.0" faces the 22.5" of the Comfort Colors C4017. Medium is not a shared size. It names a different garment depending on the cut, and no amount of shopper education fixes that.

One row above is misquoted more often than any other. The 22.0" chest that gets copied onto the Comfort Colors C4017 Medium row is a size Large figure from the common unisex ladder, and it is not the C4017's own Large either: that publishes 22.5". Read the published sheet and C4017 at Medium is 20.5", half an inch above the 20.0" Gildan G500 publishes at that letter; on these numbers C4017 sits half an inch above the common unisex ladder at every size in the table. Across the unisex Mediums the entire spread is half an inch, from Next Level's 20.5" down to 20.0". Vanity sizing explains why letters drift like this in the first place, and the t-shirt size chart sets these styles side by side.

What does garment-first sizing actually do?

1. Start from something that already fits — you pick a shirt you like wearing, which is the only fit judgement in the whole process and it is yours.
2. Measure it flat — chest, body length, sleeve, taken on a table rather than on a body, so the reading is repeatable.
3. Compare against published specs — those numbers are set against manufacturers' published garment measurements across a catalogue.
4. Return garments, not labels — the output is a list of specific products whose numbers sit near yours, with the differences shown rather than hidden.

The appeal is that every step is checkable by hand. You can re-measure the shirt, and you can open the brand's spec sheet and read the number yourself. Nothing in the chain predicts your body, so nothing in the chain can be wrong about it. Our methodology page sets out how the matching is done.

What does body-measurement sizing do?

1. Collect something about you — a tape measurement, a set of proxies such as height, weight and age, or photographs used for a body scan.
2. Build a body model — those inputs become an estimate of your measurements, often filled out from a much larger dataset of other people.
3. Compare against the brand's grading — the estimate is set against that retailer's size chart and the fit the designer intended.
4. Return a size — usually one letter or number on the product page, sometimes with a note about how confident the system is.

The appeal is real and garment-first cannot match it. Body-first works with an empty wardrobe, and it transfers across categories, because the same body model can be pointed at trousers, dresses and coats without a reference garment in each one.

Who builds body-measurement sizing?

Four companies come up most often. What follows is each company's own public positioning, kept deliberately coarse, because we have not tested any of them.

True Fit

True Fit publicly positions itself as a footwear and apparel personalisation platform that retailers connect to their product pages, drawing on a large connected dataset of garments and shopper behaviour to return a size and fit recommendation.

Fit Analytics

Fit Analytics publicly documents a product called Fit Finder, a widget placed on a retailer's product page that asks a shopper a short set of questions and returns a recommended size for that product.

Bold Metrics

Bold Metrics publicly positions itself around predicting a person's body measurements from a small number of simple inputs, and supplying those predictions to apparel brands for sizing and product development.

3DLOOK

3DLOOK publicly positions itself around mobile body scanning: a shopper takes photographs of themselves and the service derives body measurements, which are then used for size recommendation and for fit data.

Which approach is closest to garment-first sizing?

Virtusize, and it is worth saying plainly rather than burying. Virtusize's public material describes comparing a product against an item the shopper already owns, shown as an overlay on the product page. That is the same premise garment-first sizing runs on: the reference is a real garment in a real wardrobe, not a model of a body. Of the services named above, it is the methodological neighbour, not the opposite number.

How each fits into a shopping journey is best answered from the vendors' own documentation. The honest summary is that two independent teams arrived at the same starting point.

Where garment-first is worse

You must already own a garment that fits. This is the hard boundary. A first-time buyer with an empty wardrobe has nothing to measure. So does a teenager who has grown out of everything, someone whose body has changed and whose whole wardrobe is now wrong, and anyone buying a category they have never bought before. Body-first sizing answers those people and garment-first sizing does not.

It tells you nothing about your body. A flat chest of 20.0" is a fact about cotton. It will find you another garment with similar numbers, but it cannot tell you your own chest measurement, cannot answer the question a tailor asks, and cannot be carried to a shop that does not publish specs. If what you need is a number about yourself, this method never produces one.

It inherits every error in the retailer's published numbers. The match is only as good as the spec sheet behind it. If a brand publishes a figure copied from an older pattern or simply gets it wrong, the match is wrong in exactly the same direction and gives no signal that anything has gone astray. Even when the spec is right, tolerance of +/-0.3" to +/-0.8" sits underneath every figure.

It cannot work at all where nobody publishes measurements. Footwear and eyewear are the clearest cases. A photograph alone cannot tell you your shoe size; that needs your known size in another shoe or a measurement of your foot. The numbers printed inside a pair of sunglasses are readable by a person, and our guide to what those numbers mean explains them, but ShirtMath does not measure eyewear from a photograph. Where no published measurement exists, garment-first has nothing to match against.

One further limit belongs on this list because it is ours. The ShirtMath scanner reads a shirt laid flat with a scale card in frame. Roughly half of its readings land within an inch of a tape measure. Sleeve opening and bicep are the least reliable dimensions it attempts, and anything it cannot read confidently is left blank rather than guessed. A blank is a better outcome than a confident wrong number, but it is still a gap you have to close with a tape.

Where body-first is worse

It needs a measurement of you that is actually right. Self-measured tapes drift, and estimates built from height, weight and age are population averages wearing your name. The further your proportions sit from the middle of the training data, the weaker the estimate becomes, and the shopper has no way to see that happening.

It returns a size label, and a label is only as good as the brand's grading. A recommendation of Medium inherits whatever Medium means at that brand. The table above shows how far apart two Mediums can sit, so the recommendation carries all the drift that the garment measurement route steps around.

It usually depends on the retailer having integrated the vendor. These are business-to-business products. If a shop has not bought the widget, the shopper cannot use it there, and the profile does not travel to the next shop.

Body data is personal in a way garment data is not. A body scan or a set of body measurements is a sensitive record about a person. A photograph of a shirt on a table is a photograph of a shirt on a table. That matters a great deal to some shoppers and not at all to others, but it is a real difference.

How should you choose between them?

1. Own something that fits? — measure it and shop garment-first. It is the shorter path and every step is verifiable.
2. Empty wardrobe, or a body that has changed? — a body measurement is the only starting point available to you.
3. Buying inside one brand? — read that brand's own size chart first, because their grading beats any general model of them.
4. Want the durable answer? — keep both. Measure your body once with a tape, measure your best-fitting shirt once on a table, and write both sets down.

The two schools are not really competitors at the level of method. One describes clothing, one describes people, and a complete answer to "will this fit" needs both numbers eventually. What garment-first offers is that half of the problem can be settled with facts you can check yourself.

Common questions

Where to go next

The sections below carry the underlying numbers and the per-vendor detail.