An owner-run factory is a genuinely good design. One person holds the whole thing, decisions happen in seconds, and exceptions get handled by someone with the authority and the context to handle them. A ₹70 crore plant with 300 people can run this way and run well. I've seen it work better than some far more systematic operations.
It has one property that eventually ends it: it does not survive the owner being unavailable, and it does not copy to a second site. At some point you want both of those things.
What does it mean to move from an owner-run factory to a process-run one?
Documenting is not the same as systemising
The usual first attempt is a folder of standard operating procedures. Somebody writes down how things are done, everyone signs off, and the folder goes into a drawer where it ages. Six months later the procedures describe a plant that has moved on, and the real process is still what the supervisor does.
A procedure that lives in a document competes with habit and loses. A procedure embedded in the thing people already have to touch, the job card they fill or the issue slip they raise, is just how the work happens. That's the difference between writing down a process and installing one.
The four decisions that have to move first
What to make next. In an owner-run plant this is a judgement made each morning weighing customer pressure, material availability and which machine is free. It's a good judgement. Written as a rule (due date, then customer priority, then setup similarity) it's about eighty percent as good and it works when the owner is on a flight.
When to reorder. Ordering when stock "looks low" produces a range rather than a level, and the range is wide. A reorder point per item is arithmetic that a system watches continuously, which no person does. We've set out the formula in reorder-level automation to prevent stockouts.
What counts as ready to dispatch. Usually undefined, and usually a disagreement waiting to happen between production and quality. A named checklist attached to the despatch turns an argument into a step.
What a job cost. The owner's estimate is often close, and close is not enough once the product range widens. Actual material, hours and bought-outs against the specific job is the version that tells you which work to stop taking.
The test that tells you where you are
Take two weeks off with your phone off. Whatever breaks is the part still running on you. Most owners can't do this, which is itself the finding, so the smaller version works nearly as well: for one week, refuse to answer any operational question and tell people to decide. Write down every question you refuse. That list is your backlog, in priority order, and it costs nothing to generate.
"The rule is worse than my father's decision. It's also there at eleven at night when he isn't, and on both lines at once. That trade took me a while to accept and it was the right one."Machine tool manufacturer, Rajkot
Accept that the rule is worse than the judgement
This is the part owners struggle with, and they're right to, because it's true. Any rule you write will be worse than the decision a good owner makes with full context. The rule wins anyway on availability, consistency and the fact that it can be examined and improved, while a judgement can only be repeated.
So the goal isn't to replace the judgement. It's to handle the routine eighty percent with rules so the owner's attention goes to the twenty percent that genuinely needs it. That's also what makes a second plant possible and what gives a professional hire something to manage with. The scale at which this stops being optional is covered in what changes when a factory crosses 100 crore.
Facto is built to carry these rules where the work already happens rather than in a parallel document, and we deploy it ourselves one module at a time. If you want to map which of the four decisions to move first, talk to our team.
