Selling at Scale

Everything else in this wing is about permission. This page is about the thing that actually stops people, which is not permission at all.

I am not a lawyer and none of this is legal advice.

The gap is capital, not paperwork

Once you know that cottage food does not reach cheese and that selling means a licensed manufacturing dairy plant, the question stops being may I? and becomes can I afford the room?

That is a different kind of problem, and it is the same one I have already written about from three other directions:

Outfitting a kitchen — what commercial equipment actually costs, and buying used, which is where most of the savings are.
The food truck numbers and its rules and regs — the closest thing I have written to a worked example of a licensed food business at the smallest viable size.
The bakery — the same arithmetic where the product is also a temperature- and time-sensitive craft item with a long unpaid learning curve.

⭐ The reason those pages transfer is that the hard parts are not about the food. A dairy plant and a bakery differ in their inspections and their standards of identity; they do not differ much in the shape of the problem — a room you must build to a specification, a licence, an inspector, a volume you must hit to pay for the room, and a product whose quality depends on the person rather than the machine.

The three routes people actually take

Rent licensed space. A shared-use commercial kitchen or an existing licensed dairy plant with spare capacity. This is overwhelmingly the sane first move, because it converts a capital problem into a per-hour cost and lets you find out whether you like making cheese to a deadline — which is a different activity from making cheese.

⚠ Note the catch specific to cheese: you also need somewhere to age it, for months, at the right numbers. Shared kitchens rent you a vat and a schedule, not a cave. That aging requirement is what makes cheese awkward in a model that works well for bread and sauces, and it is worth solving on paper before anything else.

Build a small licensed plant. The real thing, usually on a farm, usually alongside an existing dairy operation where the milk is already there and already permitted. If you are not already farming, you are proposing to start two businesses.

Don’t sell. I want to put this third option on the page seriously rather than as a joke, because for most people reading this it is the right answer and nobody ever says so.

⭐ The honest arithmetic of not selling

Thirty years in, here is what I think the trade actually is.

Selling requires you to make the same cheese repeatedly, on a schedule, to a specification, and to have an answer when a batch goes wrong in week nine. That is the job. It is a good job and I respect it enormously. But it is emphatically not the thing I enjoy, which is making a cheese I have never made before and finding out what happened.

The moment you sell, variation becomes a defect. Everything this section celebrates — the seasonal milk, the wild rind, the batch that came out differently and taught you something — turns into a customer complaint and a traceability question.

⚠ So the useful question is not how do I get licensed? but which of the two hobbies do I want? They share equipment and share almost no daily experience.

If you are going ahead anyway

The order that seems to work, from watching people do it:

Pick the cheese first, and pick one. Preferably one that ages long enough to be forgiving and short enough to be paid for. Fresh cheeses have brutal shelf lives; three-year wheels are an unpaid loan to yourself.
Solve aging before you solve making. See above. It is the constraint everyone underestimates.
Talk to your regulator early and in person. In Michigan that is MDARD. Regulators are, in my experience, far more helpful before you have built the room than after.
Get the name and label right from the first wheel — and strongly consider naming the cheese yourself rather than reaching for a standard-of-identity name you then have to keep hitting.
Cost the milk honestly, including the milk you will pour away while you learn to make it to a schedule rather than to a mood.

⚠ I have not done this, and I want that stated plainly rather than buried. Everything above is reading, arithmetic and watching other people, not experience of running a licensed plant. Where this section speaks from experience it says so; here it does not.

Sources and further reading

Selling cheese in Michigan — the licensing position.
Kitchen equipment · Food truck · Bakery — the cost and licensing arithmetic worked through for adjacent businesses.
Aging & the cave — the constraint that makes cheese different from every other small food business.