A microgreens week starts before the orders are firm

The week did not start with an order. It started with me deciding how many trays to sow, days before I actually knew what anyone wanted.

That is the part people miss about growing to trade. The crop has a clock. Microgreens take days from sowing to a good cut, and that lead time does not care that half your orders will change on Thursday. So the first real decision of the week is a guess dressed up as a plan: you commit seed and tray space now, and you find out later whether you guessed right.

I grew microgreens during covid and sold them to kitchens, so this is my own week I am describing, not a case study I read somewhere. One honest thing first. I never kept a proper ledger of the numbers, and I am not going to invent them here to make the story tidier. What I can give you is the shape of the week, because the shape was the same every week even when the quantities were not.

Sow day, and the count that was really a bet

On sow day I would work out how much of each thing to put in. Some of that count was solid. A kitchen that took the same tray every week, near enough, was something I could sow against with a straight face.

The rest was reading tea leaves. I was supplying somewhere between five and ten accounts, and the orders behind that count did not arrive as one list. They arrived as scattered messages: a text here, a note in a longer conversation there, a "same as last week" that turned out not to mean last week at all. So before I sowed a single tray I was already reconstructing demand from memory and a phone full of half-conversations.

Standing orders were the worst offenders, because "standing" was a fiction I kept up by hand. A recurring order that changes most weeks is not recurring. It is a new order wearing last week's label, and if you sow against the label you are sowing against something that no longer exists.

The grow, then the window that does not wait

Once it was in, the grow largely looked after itself. That is the deceptive calm in the middle of the week. The trays are doing their thing and there is nothing to decide, which is exactly when you stop thinking about the orders and go and do the twenty other jobs.

Then the cut window arrives, and it does not wait for you. Microgreens are good for a stretch and then they are past it. Cut too early and you are giving away yield you grew. Leave it and the quality goes, and quality is the whole reason a chef is buying from a small grower instead of a box off a lorry.

So the middle of the week ends with a hard edge: this is ready, it needs cutting, and whatever I decided on sow day is now fixed. There is no topping up a tray that was under-sown five days ago.

Packing and delivery, against the orders as they actually were

Cutting and packing is where the sow-day bet meets reality. Now the real orders are in, or as in as they were going to get, and I am packing against them.

This is the moment the two errors show up. Some weeks I had trays I had grown for an order that had quietly shrunk or vanished, and there was nothing to do with them. Other weeks a kitchen wanted more than I had sown for, and there was no honest way to conjure a grown tray on the spot. I got it wrong in both directions across the same run of weeks: waste one week, short the next.

Delivery itself was the straightforward part. Familiar ground, actually, because I had spent about a decade as a chef before this and I had stood on the other side of that handover, receiving from producers. I knew what a kitchen wants at the back door: the right thing, when they were told, without a fuss.

The job that slipped: invoicing

Then the delivery is done and the trays are gone and it feels like the week is finished. It is not. There is an invoice for every drop, and invoicing was the job that slid off the end of every week I had.

I wrote them late. I sent them late. Some I forgot entirely until I noticed the gap, and then I was chasing money from people I needed to keep sweet, because a chef you have annoyed about a late invoice is a chef who remembers it next time you want the order. That is a bad trade to make for the sake of a job I could have done on Friday and didn't.

Why a baker's week looks the same

Swap microgreens for sourdough and the calendar barely changes. You are still committing to a bake before the orders are locked, because dough has a clock the way a tray does. You are still reading demand off scattered messages. You are still cutting or baking to a window that will not move for you, and you are still writing the invoice last, if at all.

The single thing that would have changed my week was not a bigger van or a better crop. It was one list. Not five threads and a memory, but one place where every account's order for the week sat together, so the count I worked to on Monday was the count I actually had to fill on Thursday.

That gap, every account's order in one place instead of scattered across messages, is the one makeweek is trying to close. It would not grow the trays for me, and it would not tell me how many to sow. But sowing against a list I could actually see, instead of a reconstruction, would have made the bet a smaller one.

If you run a week like this one, do the boring test. Look back at last week and count how many separate places an order or a change to it arrived. That number is the size of the bet you are making every time you commit before the orders are firm.