Last night I opened my Substack dashboard.

Eight hundred and sixty-two followers.

I sat there and looked at it, and my first thought was that it had to be a lie.

So I went and checked. It is not a lie. Every one of those is a real account. The list is public. I can scroll it and see the profiles.

The number is true.

Here is what I learned when I looked closer.

On Substack, a follower and a subscriber are not the same thing. When somebody subscribes to my newsletter, the platform follows me for them automatically. That is part of the deal. So every subscriber I have is already inside that eight hundred and sixty-two.

The rest are follows. A follow costs one tap. It puts my notes in a feed somewhere. It does not put anything in anybody’s inbox. It does not mean a person read a word I wrote.

Then I looked at the shape of the line on the graph.

Flat along the bottom for five months. Then a wall. Straight up, from about thirty to over three hundred, in what looks like a matter of days. Another step up after that. Then a long climb.

And then flat again. Since late March. Five months of writing nearly every day and that line has not moved.

Real readers do not arrive in a vertical wall. That was the platform putting me in front of people in a batch. Some feature, some recommendation, some list. Not people finding me.

Now here is the part that got me.

I have a second measurement. I run analytics on my own site. That tells me how many actual human beings came to intelligent-people.org and how long they stayed.

[CURRENT FATHOM FIGURE — visitors, average time on site, bounce rate]

Put those two numbers side by side.

Almost nine hundred accounts carry my name in a feed. A handful of people walk through the door.

Neither number is wrong. Substack counted correctly. My analytics counted correctly. Nobody made anything up. There is no fraud anywhere in this.

And yet the picture those two numbers make together is false, and I believed the false version for a good while before I checked.

That is the thing I want to put in front of you today.

You are going to spend the rest of your life reading numbers that a machine assembled for you.

Not numbers somebody lied about. Numbers that are correct.

A dashboard does not have to make anything up to leave you with the wrong idea. It only has to count the right thing and put it under a label that means something different in your head than it means in the database.

Followers. That word means something to you. It means people who chose you. Somebody who reads what you write and waits for the next one.

That is not what the field counts. The field counts accounts with a relationship flag set. Some of those are readers. Most of them are a tap somebody made once in an app.

Both of those things get printed with the same word.

Nobody had to lie to me. I did the rest myself.

This is why my framework has a rule that says you check a claim against the source before it goes out the door, and not against the number that looks about right. It is why there is another rule that says a figure has to be labeled with what it actually measures. Not the friendly word. The real scope.

I did not write those rules because I thought machines would start lying to us.

I wrote them because machines do not need to lie to leave us wrong.

Think about where this is headed.

Your bank is going to hand you a summary of your money. Your doctor’s office is going to hand you a page of results with a machine’s read on top. Your insurance company is going to hand you a number. Your employer is going to hand you a performance figure with an AI’s fingerprints on it.

Every one of those will be assembled out of true things.

And every one of them will carry a label chosen by somebody who was not sitting where you are sitting when you read it.

You will not be able to catch that by reading harder. I read that dashboard three or four times before I checked it. The number never changed and it never got any less convincing.

The only thing that caught it was a second source that measured a different way.

That is the whole trick. It is not clever. You need two measurements that do not come from the same place, and you have to be willing to believe the one you like less.

I liked eight hundred and sixty-two. I did not like thirteen.

Thirteen was the true one.

I want to be careful here, because I am not telling you the platform did something wrong to me. It did not. It counted what it said it counted, and the definition is written down where anyone can find it. I just never went and read it, because the number felt good and there was no reason to.

That is the ordinary way this happens. Not a scandal. Just a true number, a friendly word, and a man who did not check.

A true number is not a verified claim.

Write that one down somewhere. You are going to need it more than I did.

This post was drafted with AI governed assistance and reviewed and directed by Michael S. Faust Sr. before publication.

Post Library – Intelligent People Assume Nothing

I post four a day. Leave your email and it comes to you.

Contact: micvicfaust@gmail.com

© 2026 The Faust Baseline LLC | All Rights Reserved

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *