From the workshop, Sunday afternoon #9

This one picks up where the last newsletter left off - decentralisation, and what Kohutt is becoming. 

"Cottage" used to mean something specific: a cotter's home, a peasant farmer's hut, nothing to do with the lifestyle-magazine version of the word today. A cottage industry, literally, was production happening inside that hut. A merchant dropped off raw material, a family worked it up in whatever hours the household could spare, and the merchant collected the finished goods and paid by the piece. This ran European textile manufacturing for the better part of two centuries. It wasn't a side hustle. It was how cloth got made.

It didn't decline because factories were better at the craft. It declined because machines got big. Once production needed a water wheel or a steam engine, you couldn't fit the machine in a cottage, so the cottage had to come to the machine. And when the workers followed the machines into one building, someone else started setting the pace. Clocked hours, supervised shifts, punctuality as a virtue rather than a preference. Paid for time sat, not work finished. Conditions got bad enough that Parliament spent the first half of the 1800s passing Factory Acts just to regulate them. Nobody legislates a system that's obviously working for the people inside it.

The part I keep coming back to is that this constraint never applied to sewing. A sewing machine fits in a spare room. It always has. No water wheel, no steam engine, no reason the work has to happen under one roof beyond habit and management convenience. The industry centralised anyway, because everything centralised, and two centuries later we still assume that's what production looks like.

Kohutt is small enough to ignore the assumption. A business like mine doesn't need volume - the remaining oilskin is finite and the runs are numbered, so a factory's output would be pointless even if I wanted it. What it needs is consistency where the brand lives, flexibility everywhere else, and some way of surviving a bad month.

So here's how the work actually splits now. Design, cutting and quality control stay with me, in the main workshop, because that's where the Kohutt look gets decided - the seam allowances, the hardware, the finish that makes a piece recognisable from across a paddock. A few of the core pieces are sewn away from the workshop, by the piece, on someone else's machine and someone else's schedule - the arrangement I wrote about last time. And the smaller runs, the limited editions, the one-offs and the finishing work still happen here, under this roof, where Hugo has now started alongside me. He works on the limited pieces, the smaller items, and finishes what comes back in from outside - the pressing, the hardware, the final checks before anything gets wrapped. Some of what's on his bench right now I can't say much about yet, except that the pieces are numbered, there aren't many of them, and once the fabric they're cut from is gone, that's that.

The workshop isn't emptying out. It's changing jobs. Less production line, more the place where materials go out, finished work comes back, standards get held and the fiddly, low-volume pieces get made. The old cottage model was never one thing or the other either. The mill and the loom needed each other.

The piece rate matters more than it looks. Paying by the hour means buying someone's time and then supervising how they spend it, which is how you end up needing floor managers and shift schedules whether you wanted them or not. Paying by the piece means buying the work itself. The standard is in the finished article, not the clock. Whether a seam went in at midnight or 6am, around a school run or boat maintenance, isn't my business. My check happens when the work comes back, the same way Harris Tweed's inspector checks the cloth after pickup rather than watching the weaver at the loom. I've never worked well with someone looking over my shoulder - I learn by tinkering, making mistakes, figuring things out alone - and I'm not inclined to build a business that does to someone else what I could never stand myself. It helps that the people who want to work this way tend to be exactly the people who don't need watching.

There's a sharper reason this structure suits Kohutt, and it's the one I keep turning over. A one-person business is fragile in a very specific way, the kind you don't feel day to day because everything is working, right up until the morning it isn't. If I break a wrist, get the flu badly, or lose a month to family - and eventually one of those happens to everyone - production stops. Not slows. Stops. Every order, every launch, every promise on the website depends on one set of hands staying functional. I've priced and planned around my own hours carefully enough to know exactly how thin that margin is.

There's a name for the opposite of this problem: antifragility. Centralised systems, optimised until there's no slack left, break all at once, because every part depends on every other part. Decentralised systems with a bit of redundancy contain a failure to one corner, and sometimes come out better for it, since a small failure in one part becomes a lesson for the whole. A supply chain with one factory in it is a single point of failure. A handful of makers - some out there on their own machines, some here in the workshop - doesn't fail like that. One person has a bad month, the work shifts around it. Things slow. They don't stop.

The romantic argument for all this - craft, autonomy, work that fits around a life instead of consuming it - happens to be true, and it's why the people I want to work with would want to work like this. But the structural argument is colder and stronger: a small network of skilled people, loosely coupled, paid fairly for what they make, is harder to break than anything I could build under one roof. Harder to break than me alone, too.

The factories won the nineteenth century because the machines demanded it. The machines we use don't. So Kohutt gets to run the older model with modern tools, and I get to find out whether a cottage industry, run deliberately rather than by default, holds up the way I think it will. I may be wrong. The next few years will say.