Made-to-order vs buying stock: which one your business can afford

Most people frame this as a philosophical choice about how a brand should operate. It is not. It is an arithmetic question about where your cash sits, and the answer changes by product line rather than by company.

Buying stock means paying for garments months before anyone has agreed to buy them, in exchange for a lower unit cost and a garment that is ready the moment a customer wants it. Made-to-order means producing after the sale, in exchange for a higher unit cost and a customer who has to wait. Everything else in this argument follows from those two sentences.

What does made-to-order actually mean?

Made-to-order means the garment is manufactured after a customer has committed to buying it. You hold fitting samples rather than inventory, take the order, and production is triggered by that order.

It is worth separating this from two things it gets confused with, because the confusion is what makes the decision hard.

It is not print-on-demand. Print-on-demand puts your artwork on a blank that somebody else has already manufactured and is holding in a warehouse. The cut, the cloth and the construction are fixed, and the thing being made to order is the print. Made-to-order manufacturing makes the garment itself — your pattern, your cloth, your finishing.

It is not made-to-measure either, though they overlap. Made-to-measure means the garment is cut to one person's measurements. Made-to-order describes when production happens, not whose body it is cut for. You can produce standard sizes made-to-order, and most brands that move to this model do exactly that.

Where does the money actually go in each model?

This is the part worth being concrete about, because it is where the decision is really made.

Under a stock model your capital is committed at the point of production. You pay for cloth, making and freight, then you wait. The money is now sitting in boxes. Whatever sells at full price returns your margin; whatever does not gets marked down, and the markdown comes directly out of the same margin. A brand that sells 70% of a range at full price and clears the rest at 40% off is not earning 70% of its expected margin — it is earning considerably less, because the discounted third still cost full price to make.

Under made-to-order your capital is committed at the point of sale, and the customer's payment usually arrives first. What you fund up front is the sample, not the run. The unit cost is higher, sometimes materially so, but it is a known cost applied to a garment that is already sold.

So the real comparison is not "cheaper unit" against "dearer unit". It is a lower unit cost carrying markdown risk and tied-up cash, against a higher unit cost carrying neither.

When is buying stock the right answer?

It is right more often than made-to-order advocates admit, and the honest test is whether your demand is predictable enough that the forecast is not really a guess.

Buy stock when the product is a proven repeat seller with stable volume, when the customer expects immediate delivery and will go elsewhere rather than wait, or when unit cost is genuinely the binding constraint on your margin. A core white shirt that has sold the same quantity for three seasons is not a forecasting problem. Buying it in bulk is simply cheaper, and the risk you are carrying is small and well understood.

Made-to-order does not help you there. It raises your unit cost to insure against a risk you were not really running.

Where does made-to-order fail?

Three places, and it is worth knowing them before committing a range to it.

When the customer will not wait. This is the binding constraint, and it is not negotiable by being clever about it. Gifting, occasion wear bought late, and any replacement purchase are all decided by availability. If your customer needs it this week, a production lead time loses the sale no matter how good the garment is.

When the price cannot carry the unit cost. Made-to-order production costs more per piece. If your retail price was set against a bulk unit cost and there is no headroom, moving the product to made-to-order compresses a margin that was already thin. The fix is usually to move a different product, not to absorb it.

When you have not sold the product before at all. Made-to-order removes inventory risk; it does not tell you whether anyone wants the thing. A product nobody has bought yet is a demand question, and neither model answers it.

How do you decide for a given product line?

Work through it product by product rather than committing the whole business to one model. Most ranges end up split, and that is the correct outcome rather than a compromise.

Ask, in order: Is this a proven repeat seller with stable volume? If yes, stock is probably cheaper and the risk is small. Will the customer wait two to several weeks for it? If no, stock is the only option that works. Does the price carry a higher unit cost? If no, leave it in stock and move something else. Is this range where your markdowns actually happen? If yes, that is the range to move first — it is where the arithmetic changes most.

Most retailers who adopt this start with the part of the range that discounts hardest, keep their core sellers in stock, and expand once the workflow is familiar at the counter. That sequence is deliberate: it puts the model against the product where it has the most to prove.

What changes operationally at the counter?

Less than people expect, and this is the part worth rehearsing before you commit.

The sequence goes from forecast, buy, stock, discount to sample, sell, produce, deliver. In store that means you carry fitting samples in the sizes that let a customer see and feel the garment, the sale happens against the sample, and the piece that gets produced is the exact size and specification the customer chose.

Two things change for staff. They are selling from a sample rather than from a rail, which means the conversation is about specification rather than availability — and they have to give a delivery date and hold it, which means the date has to be real. The second is the one that decides whether this works. A model that produces after the sale lives or dies on whether the lead time you quote at the counter is the lead time the customer actually gets.

The short version

Made-to-order is not better than buying stock. It moves your risk from unsold inventory to customer patience, and raises your unit cost to do it. That is an excellent trade for a range that discounts heavily, a wide set of options, or a product where breadth matters more than immediacy — and a poor one for a proven repeat seller your customer expects to take home today.

Split the range, start with whatever you mark down most, and quote a date you can hold.