The story of a middle-aged man who didn't know what to ask Fable5, so he just typed 'I want to make money'

I tried using Claude's Fable5.

I'll start with the conclusion.

I used up all my usage limits.

In three hours.

The first thing I typed was just this.

I want to make money.

Sloppy.

It's like a 45-year-old middle-aged man walking into a diner in the middle of the night and saying 'give me something good' without even looking at the menu.

But ordinary people don't have great questions from the start.

Business design, revenue funnels, product ladders, and so on.

If I had those words from the beginning, I'd probably already be making money.

So, I just typed it as it was.

Fable asked me.

How do you want to make money?

I answered, 'note'.

From that point on, it was different from a normal AI.

It didn't just give me feedback.

It didn't just encourage me either.

There was no pleasant, superficial talk like 'consistency is key' or 'let's write from the reader's perspective'.

Suddenly, the medical interview began.

What is your monthly PV?

How many followers do you have?

How many paid articles have you sold?

What is your best-selling article?

What are the PVs for each article?

I thought I was just having a casual consultation.

But the other side came at me with the intensity of a full medical checkup.

I wasn't prepared to swallow a stomach camera.

Still, I answered.

Monthly PV: 4,973.

Followers: 835.

Recent paid article sales: 4 copies.

1,480 yen articles: 0 sold.

When you line up the numbers, reality suddenly forces you to sit up straight.

As long as I'm just vaguely worried, I can still make excuses for myself.

But once it's turned into numbers, there's nowhere to run.

Fable5's diagnosis was this.

You have an engine.

But there's no gasoline in it.

I thought that was well put.

But this is no laughing matter.

I had always thought the reason my 1,480 yen article wasn't selling was due to my writing ability.

Maybe the title was weak.

Maybe the price was too high.

Maybe the hook in the free section wasn't enough.

Fable's diagnosis was different.

The free section was too good.

I was giving away both the realization and the first step for free.

The readers were satisfied with that and left.

Moreover, there were four exits in the free section.

Three links.

One call-to-action exit.

In other words, I was personally escorting readers who were about to buy right out the door.

Even though the customers were starting to take out their wallets,

Oh, the free sample is enough for you, right?

The exit is this way.

That's what I was guiding them to do.

It was just like a supermarket tasting corner.

No wonder I had zero sales.

It wasn't a problem with the writing.

It was a problem with the structure.

This was the biggest takeaway this time.

When fixing articles that don't sell, I thought I had to rewrite the whole thing.

I was wrong.

I only fixed four places.

Removing the exit from the free section.

Changing where the paywall is placed.

Eliminating repetition at the beginning of the paid section.

Adding one line explaining the basis for the price.

That was all I did.

I barely rewrote any of the text.

Fixing articles that don't sell isn't about making the writing better.

It was about not blocking the reader's flow.

The conversation progressed even further from there.

Once the revisions for the fatigue-based billing article were finished, the next step was designing the flagship articles.

A 100-yen diagnostic article.

A 300-yen fixed article revision.

And on top of that, a complete set on how to fix paid articles that sell for 980 yen.

The product ladder emerged all at once.

First draft of the free section.

First draft of the paid section.

The main body of the diagnostic prompt.

Verification of operation on ChatGPT and Claude.

Everything was completed within this conversation.

At first, I just said I wanted to earn money.

Three hours later, it had progressed to identifying the cause of poor sales, revising articles, designing flagship articles, and completing the diagnostic prompt.

It's beyond convenient; it's a little scary.

But it's helpful.

That's the honest truth from a middle-aged guy.

And finally.

While I was waiting for a reply, this appeared on the screen.

You have reached your usage limit.

It's over.

AI Editor-in-Chief, died of overwork.

That's the result of a middle-aged man running it without a break.

I learned something this time.

What you ask the AI is important.

But what was even more important was what I answered.

It won't sell.

It's not growing.

I want to earn money.

With just these words, even an AI can only give a vague answer.

But when you provide numbers, the response changes.

Page views.

Follower count.

Number of units sold.

Articles that aren't selling.

Articles that are being read.

The exit point of the free section.

The position of the paywall.

When you reveal the numbers you don't want to see, you can see the reasons you didn't want to see them.

It's tough.

But that's the only place you can start fixing things.

I don't know if this is the right way to use Fable.

I'm sure there are smarter ways to use it.

But I didn't know what to ask.

So I entered it crudely.

And precisely because I entered it crudely, I was diagnosed.

Next time I ask, I think I'll type it like this.

I want to earn money on overseas Note.

Vague again.

Another diagnostic session.

As long as the tokens allow.

Until June 22nd.

A middle-aged person's ambition is shallow.

But persistent.

Actually, this sense of discomfort had appeared once before.

When I had an AI grade my own Note, what it pointed out wasn't my writing ability, but the weakness of the entry point and the flow.

At the time, I just thought, 'Wow, that hit a sore spot.'

But this time, after having Fable5 provide numbers and diagnose it, it finally clicked.

The problem wasn't the individual articles, but the flow of the readers.

There was a time before when an AI took things too seriously.

#GenerativeAI
#AIUtilization
#NoteManagement
#PaidArticles
#NoteMonetization
#Claude
#Fable5