Print-on-Demand vs Custom Manufacturing: Where the Margin Crosses Over

The two cost structures, side by side

Print-on-demand charges you a fulfilment price per unit that already contains the blank garment, the print, the labour, the platform margin and usually the shipping. On a standard heavyweight tee this typically lands somewhere around $18–24 delivered, depending on print size and destination. You pay nothing until a customer buys, and you never hold stock.

Custom manufacturing splits the same money differently. You pay a development cost once, then a unit price that falls as quantity rises. At our indicative CIF pricing an adult regular-sleeve tee starts from $10.75 a piece and a heavyweight oversized tee from $11.25, with screen printing adding $1.50 per piece and embroidery $1.65. Freight is inside that number; import duty and VAT are not.

The structural difference is what the price assumes. POD prices every unit as though it were the only one. A production run prices the whole batch and divides. That is the entire crossover.

Where the crossover actually falls

Take a heavyweight tee with a one-colour front print, sold at $45. On print-on-demand at roughly $20 landed, you keep about $25 a unit and carry no risk. On a 50-piece custom run at $11.25 plus $1.50 screen printing, your unit cost before duty is $12.75 — you keep about $32.

That $7 gap does not pay for itself immediately, because you also spent $85 on a branded sample plus around $45 shipping, and you tied up roughly $640 in stock. At 50 units the sample cost alone eats $1.70 a piece. You are ahead, but barely, and only once you have sold through.

At 300 pieces the picture changes shape. The sample cost spreads to $0.28 a unit, the per-piece price improves with quantity, and the $7-plus margin gap applies to three hundred units rather than fifty. That is roughly $2,100 of additional gross margin on a single run — and it is why the honest crossover sits nearer 200–300 units per design than the 50-piece minimum suggests.

• Under ~100 units per design per quarter: print-on-demand almost always wins
• 100–250 units: genuinely marginal, and depends on how tight your cash is
• 250+ units per design: custom production wins clearly, and the gap widens fast
• Any volume, if you need a garment POD does not stock: custom is the only option

The costs each side hides

Both models have costs that do not appear in the headline comparison, and both sets are usually discovered late.

• POD hides garment control. You cannot change the blank, so fabric weight, fit and neck construction are decided by a supplier optimising for their catalogue, not your brand.
• POD hides reorder risk. When your chosen blank is discontinued or reformulated, your product changes without your involvement and your returning customers notice.
• POD hides the label. Most services print inside a garment that is not yours, and the customer can see whose it is.
• Custom hides duty. Our unit price includes freight to your door, but import VAT and duty are billed to you at clearance — we are not REX registered, so budget standard rates rather than a preference.
• Custom hides working capital. A 300-piece run at $12.75 is about $3,800 committed before a single sale, split 50% before production and 50% before shipment.
• Custom hides lead time. Production runs 12–15 days for orders up to 2,000 pieces from an approved sample, but allow 25–35 days end to end because sampling and approval come first.

The hybrid most brands should actually run

Treating this as a binary is the mistake. The strongest pattern we see is brands running both at once, deliberately: print-on-demand as the design laboratory, custom production for anything proven.

New designs launch on POD, where the cost of being wrong is zero. Sell-through data accumulates. Anything clearing 250 units a quarter moves to a custom run — usually as an upgraded product, on heavier fabric with a proper neck label, at a higher price point. The POD listing either retires or stays as the size-and-colour long tail.

This gives you the risk profile of POD on unproven designs and the margin of production on proven ones, and it means your first custom order is placed against real sales data rather than a hunch.

The reason brands stay on print-on-demand past the crossover is usually the minimum waiting on the other side — 1,000 pieces at most factories. Ours is 50 per style, which is what makes crossing at 300 possible at all; the mechanics are on our low MOQ manufacturing page.

When to stay on print-on-demand

We would rather say this plainly than take an order that does not make sense. Stay on POD if you are selling under about 100 units per design per quarter, if the working capital for a run would be money you cannot afford to have tied up for two months, or if your catalogue is wide and shallow — forty designs selling a handful each is a POD business, not a manufacturing one.

Stay on POD if your product is genuinely a print rather than a garment: if the artwork is the entire proposition and the blank underneath is immaterial to your customer, you are paying for control you will not use.

The tell that you have outgrown it is usually not a spreadsheet. It is the point where you start wanting to change the garment — a heavier fabric, a boxier fit, your own neck label — and cannot.