Apparel Sizing: Fix the Garment, Not the Size Chart

Most apparel brands eventually rewrite the size chart. Measurements get checked against a real sample, the fit intent gets stated, the numbers stop being a template. That work usually moves the number. Then the number stops moving, and the chart is no longer what is holding it up.

This is where the work tends to stop, because the next step costs money. We have written before about why fashion returns are born in the catalogue rather than at the courier. This post is about what comes after the catalogue is honest. An accurate chart describing a badly graded garment is still accurate. It just describes the problem precisely.

Why your medium and their medium are different garments

There is no common body standard Indian apparel brands are obliged to grade to. Each brand starts from its own block, fits it on its own fit model, and grades by its own rules. The letters on the label are internal shorthand, never an industry unit.

A repeat buyer solves this with memory. They know your medium. A first-time buyer has nothing to transfer, so your size is an unknown attached to an unfamiliar brand. They do one of three things. They leave. They order the size they wear elsewhere and hope. Or they order two and return one.

Two of those cost money directly and the third costs you the sale. All three land on the first order, the most expensive order you will ever buy. The cost of inconsistent sizing concentrates on acquisition.

Garment measurements and body measurements are not the same chart

This is the quietest failure in Indian apparel listings, and it is almost never deliberate. A body measurement describes the person. A garment measurement describes the item laid flat, and it already contains the ease the designer put there on purpose. The two numbers for the same size differ by design, and neither is wrong.

The damage happens when the chart does not say which one it is. A buyer sees a column headed chest, measures their own, finds the garment number larger, and concludes the garment runs big. They size down, and it arrives fitted when it was drafted relaxed. Nothing in the listing was false. The convention was never stated.

It gets worse at catalogue scale, because the two conventions rarely arrive from the same place. Charts copied from a tech pack are garment measurements. Charts written by a marketing team are body measurements. Run both into one catalogue and your return pattern looks random when it is really two clean patterns overlaid. The fix is boring and permanent. Choose one convention, state it in plain words above the table, explain how to take the measurement, and apply it to every style. It is the kind of detail that separates a catalogue that converts from one that quietly leaks, which we cover in the listing mistakes that kill conversion.

Returns are the cheapest fit research you will ever get

Every returned garment is a funded, unsolicited product test, and most brands read it at the wrong altitude.

Style-level return rate tells you a style has a problem, not what the problem is. Read the same data at size level with the direction of the miss attached, and the diagnosis usually falls out on its own.

• If every size returns as too small at roughly the same rate, the chart is offset. One edit fixes the whole style.
• If only the top of the range returns as too small, the grading is failing as it scales up. That is a pattern problem, not a copy problem.
• If too small and too large come back in similar volume, buyers are guessing, which usually means the chart is unclear rather than wrong.
• If one production lot behaves differently from the rest, the garment changed and nobody told the catalogue.

The most useful field is the intended size on an exchange request. The buyer has handed you direction and magnitude in one click. Free-text comments are second. Both sit in reports almost nobody opens, and both are free.

One boundary worth drawing. This is about returns after a successful delivery, by someone who tried the garment on. It is a different problem from parcels refused at the door, which we handle in cutting RTO on cash orders. Mixing the two into one number hides both.

When one SKU is the outlier, treat it as a product decision

Category return rates get quoted as though they were weather. Rank your own styles by return rate against their own volume and the distribution is almost never flat. A few SKUs carry a disproportionate share, and those are decisions, not conditions.

For an outlier, work in order. Measure a physical unit from a shipped lot against the published chart, not the tech pack, because the tech pack is an intention and the garment is a fact. Measure three units from the same lot to separate a design problem from a consistency problem. Check whether the fabric changed.

Then take a product action. Re-grade the size run. Change the block. Change the size ratio you buy, because a style that fails only at one end of the curve may just be over-bought there. Or discontinue it. Pulling a style that reliably comes back is cheaper than funding it another season. What happens to the units already in the returns bay is a separate discipline, covered in disposition, grading and write-off.

The honest tradeoff on a generous returns policy

A free and easy returns policy does buy conversion. For a first-time buyer with no reason to trust your sizing, it is what makes the order possible at all. That is the mechanism, not a marketing claim.

It also guarantees you pay on the reverse leg, on a garment that comes back creased, sometimes worn, sometimes unsellable at full price. Both are true at once. The mistake is running the policy while pretending only the first one is.

Price it. Work out what one returned unit costs fully loaded, forward freight, reverse freight, inspection, repack and markdown. Hold that against what the policy buys on first-order conversion, and decide per category if the answer differs. What you should not do is quietly tighten the policy to fix a fit problem. That does not remove the cost. It moves it to the review page, where it compounds.

Fit is the largest controllable cost in Indian online apparel. It is controllable at the garment. The chart is where you describe the decision, not where you make it.