The garage is the thing that disappeared

The left half of that picture is what a small clothing business used to look like. Not the design part, not the selling part — the storage part, which was most of it by volume and nearly all of it by risk.

The right half is the same business. The difference isn't tidiness.

What actually changed?

The guess moved from before the sale to after it.

Under the old model you had to predict three things in advance: which designs would sell, in which sizes, in which colours. Then you paid for all of it up front and found out whether you were right. Every wrong prediction became a box.

Made to order inverts the sequence. Somebody orders a specific design in a specific size and colour, and then it gets made. Nothing is predicted, so nothing can be wrong.

Is inventory really the main cost?

The cash is one thing. The decision pressure is the bigger one.

Money tied up in stock is obvious. What's less obvious is what it does to design choices: when a wrong guess costs a garage, you stop making unusual things. You print the safe design, in the safe colours, in a size curve borrowed from someone else, because the specific and strange one might not move seventy-two units.

Which is exactly backwards from what actually sells — specific beats broad, and the old economics punished specificity hardest.

So what's the catch?

Per-unit cost, and it's a real trade rather than a footnote.

Bulk production is cheaper per garment. That's the whole reason minimum orders existed, and it hasn't stopped being true. Made to order costs more per shirt and buys you the elimination of risk, storage, fulfilment and the guess.

For a small operation that trade is overwhelmingly worth it. For someone shipping thousands of a proven design, it stops being obvious. Being honest about where the line is matters more than pretending there isn't one.

Does anything still need predicting?

Almost nothing on the supply side, which relocates the problem.

You no longer predict quantities, sizes or colours. What you still have to know is who this is for — and that was always the hard part, just previously hidden behind logistics that were loud enough to feel like the job.

That's the uncomfortable part of the change. When the warehouse disappears, what's left is distribution and taste, and neither of those was ever solved by working harder at the boxes.

What about returns and unsold stock?

There's no unsold stock, which removes a category of loss entirely.

A design that nobody orders costs nothing — it just sits there being available. Which means the correct number of designs to have published is all of them, and a design that sells twice a year is not a failure, it's a design that sells twice a year at no carrying cost.

That's a genuinely different mental model from retail, and it takes a while to stop pruning things out of habit.

Does this work for a one-off?

That's the extreme case and it works identically.

One shirt, one person, one address. No minimum, no setup, no leftover. Which means the same system serves a company of four, a family reunion, and someone making a single thing for one friend — and none of them are edge cases.

The old model had a floor, and everything below the floor simply didn't get made. That's the category that opened up.

What does the empty room actually need?

A design, a page, and somebody who wants it.

The infrastructure that used to define the business — stock, storage, fulfilment, packing — isn't your problem anymore, and reproducing it as a "workflow" you manage is the trap most of the advice sells.

The laptop on the desk is not an aspirational image. It's an accurate inventory.

How do I get one made?

Describe the design at JustOG and publish it, or just make one for yourself. Pick a direction, drag the crop frame, see it composited on the real garment, and it's made to order and shipped.

Designs other people have published are in the shop.

The boxes were never the business. They were the cost of not knowing.