Most of Building SellerProof Has Been Saying No

There is a file on my desk with more in it than the build plan has.

It is the list of things SellerProof is never going to do.

I started it early. It keeps growing faster than the feature list. And I have come around to thinking it is the most important document in the whole project.

Where the list came from

Every builder I know catches the same disease.

You start with one clean idea. Somebody says "you know what would be cool." Eighteen months later you are maintaining a product nobody asked for that you can no longer explain in a sentence.

I have watched it happen to good tools in the seller world. A tool starts as one honest thing. Then it adds a repricer. Then a scraper. Then an AI that writes your listings. Then a dashboard with forty numbers on it. By the end it does everything, and the seller still cannot answer the only question that has ever mattered on a bad day.

What actually happened on my account, and can I prove it?

So before I wrote much of anything, I wrote the no list.

What is on it

I am going to give you the real entries, because a philosophy you will not put in writing is not a philosophy. It is a mood.

No repricer. The second software moves your price on its own, the software owns that decision and you cannot explain it afterward. "The tool did it" is not an answer you want to give a supplier, a customer, or yourself at two in the morning. Pricing is a judgment call. Judgment stays with the human.

Nothing pulled off another seller's account. If I would not want it scraped off mine, it does not go in. That rule has already cost me some obvious, marketable features. I am fine with it.

No software that writes your words for you. A record is worth exactly as much as it is true. The minute a machine starts generating your explanation, the record stops being evidence and turns into marketing. There is a whole industry built on making sellers sound more organized than they are. I do not want to be in it.

Nothing about your business locked behind a login. Your records export, plain format, always. That is a requirement, not an upsell. A tool that holds your own history hostage is not a tool, it is a hook, and I have been on the receiving end of enough platforms that work that way.

No vanity dashboard. Forty numbers on a screen is a screensaver. If a number does not change what you do on Monday, it is decoration and it is coming off.

The test every idea has to pass

I have one question I run everything through, and it is not a product question.

Would the seller I was ten years ago have cared about this on a bad Tuesday?

Not on a good day. Anybody will click a shiny feature on a good day. A bad Tuesday is when something goes sideways, the money is stuck, nobody will tell you why, and the whole thing comes down to what you can put in front of somebody.

Most feature ideas die on that question. They are fun on a good day and worthless on a bad one. That is the whole filter.

I have written before about the through line here, that I am building software that keeps its own receipts. The no list is the other half of that. Keeping receipts only works if the tool stays small enough that the receipts still mean something.

Why saying no is the actual work

Every yes is a bill.

Not a one time bill. A subscription. Every feature you ship has to be maintained, supported, documented, secured, and explained for as long as the thing exists. You pay that bill in the middle of the night, years from now, when the fun of building it is long gone.

Nobody puts that on the roadmap. The roadmap only shows the ship date. It never shows the twelve years of carrying it.

So the no list is not me being precious about design. It is me refusing to sign contracts on behalf of a version of myself who has to honor them later and will not remember why.

Scope is not a limitation. Scope is a promise you make to your future self, in writing, while you are still calm enough to mean it.

The part any business owner can use today

You do not need to be building software for this.

Write your not-doing list before you write your doing list. Then put it somewhere you will actually see it, because the whole value of the list is that it exists before you get excited.

Three prompts to get you started.

One. What is the job you keep taking that you hate every time, and what would it cost you to stop taking it? Write the number. It is usually smaller than the dread.

Two. What is the service everybody says you should offer that you have no interest in being good at? That belongs on the list, not the roadmap.

Three. What would a competitor do that you would never do, even if it worked? That is your line. Write it down while nobody is pressuring you, because somebody eventually will.

The list is not about being noble. It is about speed. A man who already knows his answer to eleven bad ideas can move very fast on the good one.

What I want out of this

I am not trying to build the biggest thing in this category.

I am trying to build the one you can still explain in a single sentence five years from now, that still does the same thing it said it would do, and that hands you your own records on the way out if you decide to leave.

Everything on the no list is there so that sentence stays short.

Small thing. Also the hardest part of the job.