A 114 cm silk roll takes two 53 cm squares side by side, and only one 65. That step costs the 65 scarf 2.45 times the cloth for 50% more finished area, and it moves the fabric bill further than momme weight does.
A custom scarf’s landed unit cost, meaning the factory invoice plus freight and duty, has seven lines in it. Only two of them fall when you order more.
Build the number from the roll width up rather than from a quoted price down, and it tells you which input to change when the margin comes up short.
Cloth cost per scarf steps at the roll width instead of sloping with area. Two squares fit across a 114 cm roll up to a 57 cm cut, and past that the count drops to one, doubling the cloth on the next centimetre.
Cut-size yield on a nominal 114 cm width:
| Finished size | m² per scarf | vs 53 x 53 |
|---|---|---|
| 53 x 53 cm | 0.302 | 1.00x |
| 57 x 57 cm | 0.325 | 1.08x |
| 65 x 65 cm | 0.741 | 2.45x |
| 110 x 110 cm | 1.254 | 4.15x |
When you look at the numbers, the big scarf is the efficient one. A 110 x 110 wastes 3% of its cloth against 43% for a 65 x 65, so it is expensive because it is large, not because it is wasteful.
Four assumptions sit under those numbers:
Ask the mill for usable width before the artwork size is fixed. Nudging a 65 down to 57 saves more than dropping 16 mm twill to 12 mm habotai, which are about a third apart in fibre.
Screen printing carries a per-colour setup charge, so its unit cost falls as the run grows. Digital has almost nothing to amortise, and the first scarf costs close to what the thousandth costs.
The crossover is not a number worth memorising. Divide your screen charge by the run, add the per-unit screen rate, and set that against the digital rate per metre. Colour count moves it: screen setup scales with colours, digital does not.
The rest of the quote sorts the same way:
| Cost line | Behaviour as the run grows |
|---|---|
| Cloth, hem, trim and packaging | per unit, flat |
| digital flat; screen setup divides by the run | |
| Sampling and strike-offs | fixed, divides by the run |
| Freight and duty | per unit and percentage of value |
Divide the strike-off cost by the run. A run of 10 pieces carries that figure almost whole on each scarf; 300 buries it.
Quantity rungs sit near 10 pieces, 50 to 200, and 500 up. Where a factory takes custom scarf production without a minimum order, the 10-piece rung tests artwork on real cloth. A sampling round takes 7 to 10 days.
The hand-rolled hem is the opposite case. At two to three scarves an hour it never divides by anything: a 300-piece run books about 15 finisher-days, and earns that cost only above roughly $100 retail.
Inbound freight adds roughly $2 to $3 a scarf on goods this small and light. Duty is set by fibre content rather than value, and heading 6214 of the US Harmonized Tariff Schedule spans nearly ten times end to end.
The general rate column, before any trade-remedy layer:
Crossing below 70% silk by weight is a cliff, and the spec decision that crosses it has nothing to do with duty. A blend chosen for hand or price triples the rate on the same value.
The rate follows the fibre content declared on the commercial invoice, so what your silk scarf supplier writes there decides which line you enter under.
Trade-remedy tariffs stack on top of those rates. They are origin-specific and have moved repeatedly, so the rate that counts is the one live on your entry date, from the schedule in force or your customs broker.
Card processing takes its cut on the sell side, not inside landed cost.
A multiple and a margin are the same relationship written two ways: margin = 1 – 1/multiple. A 2x multiple, keystone, is a 50% gross margin, 4x is 75%, and 8x is 87.5%.
Three quantities get called margin in one conversation: the multiple on cost, the margin on the tag, and the margin that survives to the accounts.
Doubling cost to wholesale and doubling again to retail lands at 4x cost-to-retail.
The cost-to-retail ladder in AIMS360’s apparel pricing benchmark puts 4x at the bottom, in the basics tier, and premium at 5.3 to 6.5x. That is a vendor figure rather than research, but a silk line priced like basics gives away its tier.
Run fabric, size and print method through a scarf budget calculator before you fix a retail price.
Full-price sell-through across fashion retail has fallen to roughly 50% on McKinsey’s numbers, down from a 70 to 75% norm, with markdowns taking 20 to 50% of net sales.
Both figures are apparel-wide, and scarves are a softer case than fitted apparel. Half a first drop selling at full price is the current normal, not a failure case.
American Eagle’s FY2025 gross margin fell 2.7 points, from 39.2% to 36.5%, and operating income nearly halved on markdowns and an inventory write-down. Keystone is a floor, not a target, when that much moves on the gross line.
Cost the cut size before you cost anything else. It is the one input where a centimetre of design intent doubles the fabric bill.
The ranking holds across most briefs: cut size first, then run length, then momme. Hem finish and packaging move cents, and they move lead time far more than margin.
All three levers close at design freeze. Once the artwork is approved and the run is booked, the only variable left is the retail price, and the shelf sets that one.