Size Curve Calculator
Split a buy across sizes without losing units.
Distribute an order across sizes by ratio, with the rounding handled so the size counts always add back to the order quantity.
Units allocated
500
M is the largest at 166
Rounding is handled so the sizes always add back to the order quantity exactly — leftover units go to the sizes with the largest fractional parts. A curve that does not sum to the buy is the commonest spreadsheet error in a buying office, and it surfaces at the factory rather than the desk. Build your curve from your own sales history, not an industry average.
How the Size Curve Calculator works
A size curve turns an order quantity into units per size, and the arithmetic is a ratio split with one trap: rounding each size independently gains or loses units against the total. A curve that does not sum to the order quantity is the most common spreadsheet error in a buying office, and it is discovered at the factory rather than the desk.
Also known as: size ratio calculator · size breakdown calculator · size distribution calculator
The rounding problem
A size curve is a ratio: 1:3:4:3:1 across XS to XL means the M gets four twelfths of the buy. Multiplying the order quantity by each size's share gives the units, and almost never gives whole numbers.
Rounding each size independently is where it goes wrong. Round everything to nearest and a 500 unit buy can come out at 499 or 502 — a discrepancy that surfaces at the factory rather than at the desk, and that has to be resolved by somebody guessing.
The correct method floors every size first, then hands the leftover units to the sizes with the largest fractional parts until the total matches exactly. It is the same largest-remainder method used for allocating seats in proportional voting, and for the same reason.
A worked example
A 137 unit buy on a 1:3:4:3:1 curve. The exact shares are 11.42, 34.25, 45.67, 34.25 and 11.42. Flooring gives 11, 34, 45, 34, 11 — a total of 135, two short.
The two largest fractional parts are the M at .67 and one of the .42s, so those two sizes get an extra unit each: 11, 34, 46, 34, 12. That sums to 137 exactly, and no size is more than one unit from its ideal share.
Where the answer misleads
The curve itself matters far more than the rounding. Inheriting an industry-standard curve rather than building one from your own sales history is how a business ends up with a rail of unsold XS — size demand varies enormously by brand, category and market.
One curve rarely fits a whole range. Fitted styles skew smaller, oversized styles larger, and a colour that attracts a different customer can shift the mix noticeably. Applying a single curve across a range is convenient rather than correct.
Finally, the curve should reflect demand, not past sales, and those differ when sizes sold out early. A size that sold through 100% in week three was under-bought, and its share of last year's sales understates what it should have been.
Not financial advice. Marketplace fees change, and they vary by country, plan and seller status. Every rate here is an editable default, not a quoted price, check the platform's current fee schedule before you price a product against it. This is not tax or business advice.
Frequently asked questions
What is a size curve?
The proportion of a buy allocated to each size, usually written as a ratio like 1:3:4:3:1 for XS to XL. Multiplying the order quantity by each size's share of the total gives the units per size.
How do I stop the sizes not adding up?
Round every size down first, then hand the leftover units to the sizes with the largest fractional parts until the total matches. That is what this calculator does, and it guarantees the split sums to the order exactly.
Where should a size curve come from?
Your own sales history by size, not an industry average. Size demand varies enormously by brand, category and market, and inheriting someone else's curve is how a business ends up with a rail of unsold XS.
Should the curve change by colour or style?
Often it should. Fitted styles skew smaller and oversized styles larger, and some colours attract a different customer entirely. Applying one curve across a range is convenient rather than correct.