Six Days Ago I Wrote About This. Then I Did It Four More Times.

On August 11 I published an article about a problem in my own archive. Too many category names. Forty six of them. One subject filed under four different labels.

I ran the same query this morning.

Fifty categories.

Four new names in six days, every one of them created after I published the warning about creating new names.

The new ones are Ryan's Take, Faith and Service, History and Lessons, and Texas History and Service. I already had a History and Service. I already had a Faith and Today. I already had a Texas and Community.

Twenty of my fifty categories still hold exactly one article.

Writing about a problem is not fixing a problem

This is the part I want any owner reading this to take, because it cost me six days to learn it twice.

I did the hard part. I audited my own work in public. I put the ugly number on the internet with my name on it. I explained the mechanism, I explained why it hurt readers, and I meant every word of it.

Then I changed nothing about the system, and the system kept doing what it was built to do.

Nothing in my publishing pipeline stopped anybody from typing a brand new category name into an open text field. Not a warning, not a list, not a check. So four more got typed. Not out of carelessness. Every one of those four names is a perfectly reasonable name. That is exactly the trap. Reasonable choices, made one at a time, with nothing in the way.

The open text field is the villain

Look at your own operation for this pattern. It is everywhere once you see it.

An open text field where a fixed list belongs.

Category. Tag. Lead source. Job title. Service type. Referral origin. Anywhere your software says "type something here" and your business needed you to say "pick one of these seven."

The field does not feel dangerous. It feels flexible. Flexibility is what you want on day one when you do not know your own shape yet.

By month three it is not flexibility. It is drift. And drift does not announce itself, it just quietly splits your best subject into four piles that do not point at each other, and then a reader who lands on one pile never learns the other three exist.

I have written before about the one hundred thirty three pages on this site that were dead ends. Same disease, different organ. Both come from the same root cause, which is that I built a machine that ships fast and never built the thing that tells it no.

What a real fix looks like

A real fix has three properties. It is in the system, not in your head. It fires at the moment of the mistake, not at review time. And it is annoying to get around.

For this one that means the category field stops being a text box and becomes a list. If a new category genuinely needs to exist, adding it should require a deliberate second action, not a keystroke. Ten seconds of friction at the right moment beats ten hours of cleanup later.

I am not writing this to flog myself. I am writing it because the version of this story you usually read online ends at the audit. Guy finds problem, guy posts about problem, comments say great insight, everyone moves on, problem grows for another six days.

The audit is the cheap part. The constraint is the expensive part. Only one of them changes the number.

Your turn

Go find one open text field in your business that should be a fixed list. Just one.

If you cannot think of one, pull a report and count how many distinct values live in a field you assumed had five. That number is your answer, and I promise it is bigger than you think.

This is most of what I actually do when I build for somebody. Not the pretty part. The part where we find the place your system lets you hurt yourself and we close it.

If your site or your intake is quietly drifting like this and you want somebody to go find it, that is work I take on.

Own your platform. Then build the thing that keeps you honest on it.