Solo Operations

What runs while you are not watching

A system that runs quietly is not the same as one that runs correctly.

I found it by accident. A product on one of the businesses I run had a pricing bug — the card on the page said one number, the checkout charged another. Not on one item. On most of them. Quietly, for longer than I want to say.

Nobody complained. The gap was not enormous. It was just wrong, in a direction that cost customers money they were not told they would spend.

I found it because I happened to look. Not because I had a process for looking.

That is the part I kept returning to yesterday. In a solo operation, you are the QA department. You are customer service. You are the ops team that runs the quarterly audit. Nobody else is checking whether the number on the card matches the number in the cart. Nobody is reviewing whether the data your tools show you has drifted from reality. Nobody flags the timestamp that is eight hours off, the counter that stopped following what actually happens, the two records that should be one.

You built the system. It runs. You moved on. The running feels like success — and it is, right up until it is not.

The map and the territory

In a company with departments, someone's job is to make the map match the territory. In a solo business, the map is whatever you last drew, and the territory keeps moving without informing you.

The specific knowledge that compounds in this life is not knowing how to build things. It is knowing when to stop trusting what you built and go look at it again with fresh eyes. That second look is the one most founders skip, because the thing is running and there are seventeen other things that are not running yet.

I spent most of yesterday closing gaps — small ones, each individually defensible. Timestamps technically correct but practically wrong. Counts accurate at one moment, stale by the next. A filter quietly excluding the right answer. None of these broke anything visible. All of them were making the picture slightly wrong, and I was making decisions from that picture.

The pricing bug is the version of this that has a dollar sign on it. The others have a decision sign on it.

The question I left with is not how to fix the system. I fixed the system. The question is: what else is running that I have not looked at recently.

That question has no bottom. But it is a better question than what should I build next.

Keep going

Daily essay

Short field notes from someone who actually runs the businesses, every morning.