How's it going, Reader?

There is a version of business advice that always tells you to add.

• Add a platform
• Add an offer
• Add a launch
• Add another hour to the day

It sounds like progress because it feels like effort, and most of us have followed it for years without asking whether the thing we are adding actually touches the reason we are stuck. 😫

I went the other direction recently.

Instead of asking Claude what I should add, I gave it every ugly number in my business and asked it what was in my way. Not five things, but one thing.

And I told it not to be nice to me about the answer. 😂

It found something I had been looking straight past for over a month.

My registrations had been sliding, and I had been blaming seasonality, blaming my audience, blaming the fact that I had run a few workshops close together.

None of that was it.

What had actually changed was the cost of the traffic feeding the whole thing, which had quietly quadrupled over five runs while I was busy optimizing everything downstream.

Every fix I had been considering was aimed at a room that the real problem was preventing from filling in the first place.

That is the difference between adding and removing.

I could have spent another month writing better emails to a room that was never going to be big enough for better emails to matter.

Here is the prompt, and I want to explain why it is built the way it is.

The new Fable 5.1 model in Claude is noticeably better at holding a large, messy pile of information in its head at once and reasoning across all of it, which is exactly what this kind of question needs.

But it will also give you a shallower answer if you let it, so the prompt tells it to work hard, read everything before it responds, and label how confident it actually is.

That last part is what turns it from a chatbot opinion into something you can act on.

Copy this whole thing:

The step most people are tempted to delete is number four, and it does the heavy lifting.

Once you can see which parts of the answer are grounded in your real numbers and which parts are the model filling a gap, you stop taking confident guesses as findings.

I have overturned my own conclusions four times this way, and each reversal saved me from spending money in the wrong direction.

Give it fifteen minutes and your real numbers, not your tidy ones. The messier the paste, the better the answer.

If this is the kind of thinking you want running your business every day rather than in a one-off chat, that is what I am teaching on Monday, September 14 at 2 pm CST.

It is free, it is live, and I'll show the full setup on screen, including the parts I don't put in emails.

Register now if you haven't yet!

​REGISTER FOR THE WORKSHOP​

Becky

P.S.: Everyone who attends live gets new free gifts for being there. Don't forget to join at 2 pm CST/3 pm EST/12 pm PST/8 pm UK!