Short answer
Start with one style, one colour and one size range. Launch too many styles at a low minimum and you commit your cash to inventory nobody has validated — and the per-garment cost stays high because every style repeats its own setup. Prove one style sells, then widen the range using the same fabric, trims and factory setup, which drops your unit cost on everything that follows.
Key points
- The minimum is per style and per colour, so styles cost you the minimum each time.
- Fewer styles at a slightly higher minimum often produces cheaper inventory than many styles at the floor.
- Your first run is a test, not an inventory position. Treat it like one.
- Reusing fabric and trims across your second run is the single biggest lever on unit cost.
- Decide the sizing and trims before sampling — they are the expensive things to change later.
The advice that circulates about starting a clothing brand is mostly about marketing. The advice that matters is about inventory, and it is less exciting: your first run is a test, and the most common way to fail it is to treat it as a launch.
The arithmetic that decides most of this
The minimum is 30 pieces per style per colour. It is not 30 pieces per order.
| Range | Total commitment | What it tells you |
|---|---|---|
| 1 style, 1 colour, 4 sizes | 30 pieces | Whether the design sells, and which sizes move |
| 2 styles, 1 colour each | 60 pieces | Whether the category has interest |
| 4 styles, 2 colours each | 240 pieces | Almost nothing you can act on |
| 6 styles, 3 colours each | 540 pieces | You have committed to a range you have not tested |
The last two rows are the ones people arrive with. They spend their cash on breadth instead of learning, and when the data comes back it is spread too thin across styles to tell them anything.
There is also a cost consequence. Setup — pattern grading, screens, digitising, trims — repeats for every style. Four styles of one design is cheaper per garment than one style of four designs, which is counter-intuitive and consistently true.
The sequence that works
**1. Pick one style you are confident about.** Not the most experimental one, and not the safest boring one. The one you would put your own money behind.
**2. Choose one colour.** Colourways multiply the minimum. Add them in run two, using what run one taught you.
**3. Use one fabric.** Reusing it in your second run is the biggest single lever on unit cost, because fabric setup, colour matching and shrinkage behaviour are already solved.
**4. Sample properly.** One sample round, properly specified, costs less than three rushed ones. This is not the place to economise.
**5. Run the minimum.** Produce the smallest quantity that gives you a real read. You are buying information, not inventory.
**6. Decide what "it worked" means before you ship.** A sell-through percentage, a return rate, a reorder signal. Deciding afterwards that it "did not work" is how people talk themselves into a reorder.
**7. Build run two around the same platform.** Same fabric, same trims, same fit block, now across more styles and colours — and your unit cost drops because the setup is largely already paid.
What to spend money on and what to cut
**Spend on:** fabric quality, fit, the pre-production sample, and trims. These are visible to the customer and they determine whether the garment survives contact with a washing machine.
**Cut for the first run:** custom decoration, elaborate packaging, wide size ranges, multiple colours, and anything you cannot change once it is on a label.
The parts buyers underestimate
**Sizing.** Your first run tells you the real size distribution for your market. That is one of the most valuable things it produces, and it is worth more than the profit on the garments themselves.
**Returns and exchanges.** The shape of your returns tells you whether it was sizing, quality or expectation. It is the cheapest quality data available.
**Lead time.** Working backward from a date you need the stock is a discipline, and it is where first-time buyers lose their launch window. The timeline article covers this properly.
**Cash timing.** You pay a deposit months before the stock arrives. Model that gap before committing, not after.
Where to go next
Frequently asked questions
What is the smallest realistic first order?
Our starting minimum is 30 pieces per style per colour. What that becomes in practice depends on how many styles and sizes you want — which is why a narrow first range matters more than a low headline number.
How many styles should a first range have?
Fewer than feels comfortable. One or two styles is a normal and sensible first run. The point of the first order is to learn what sells, not to present a full collection.
Should I wait until I know my sizes better?
No. Run one style across the sizes you expect to sell, learn the actual size distribution from it, and use that for the next run. Waiting for certainty before the first order usually means waiting indefinitely.
How do I keep the first run cheap?
One fabric, one colour, minimal trims, a narrow size range and no custom decoration. Each of those is a setup cost that repeats per style, so removing them from a first run is cheaper than removing them later from your margin.
What goes wrong most often?
Launching too wide. A first range of six styles commits the full minimum six times, spreads the cash thin, and means no single style gets enough units to show whether it works.