Hey future-proof friends 💜

Quick question: have you ever given Claude (or ChatGPT) a really clear instruction… and then had it completely ignore you?

Turns out, one tiny word can completely change the result you get!

So this week I want to share the single smallest prompting tweak I've found that makes the biggest difference.

Let's get into it, but before we dive in:

Ready to transform artificial intelligence from a buzzword into your personal revenue generator

HubSpot’s groundbreaking guide "200+ AI-Powered Income Ideas" is your gateway to financial innovation in the digital age.

Inside you'll discover:

Download your guide today and unlock a future where artificial intelligence powers your success. Your next income stream is waiting.

Get Your Guide

TL;DR - today’s lineup:

💌 Your say genuinely shapes this newsletter: I genuinely check for feedback like a maniac after I send this out because I really want to know what you honestly think. So thank you!

Here's a scenario yo might have experienced:

You tell Claude: "Don’t em dashes"

And it listens! No em dashes.

But then you read back what it gives you, and something's still… off. It's now packed with phrases like "it's not just X, it's Y." Every paragraph opens with "In today's fast-paced world."

The em dashes are gone. But it still sounds exactly like AI wrote it.

Here's what actually happened: you banned one specific thing, so Claude stopped doing that thing. But you never explained what you were really trying to fix - which was sounding robotic. So every other AI tell is still fair game.

AI can take your rules incredibly literally.

But there’s one word that makes this better:

Because.

⛔ Weak: "Don't use emojis."

✅ Strong: "Don't use emojis because this is going into a formal proposal for a corporate client who expects a polished, professional tone."

See that?

In the first version, you banned one specific thing - emojis. And Claude will very likely obey that.

In the second version, you explained the goal - a formal, professional document. Now Claude understands that emojis, casual language, exclamation marks, and anything else that undermines a polished tone are ALL off the table.

You didn't have to list every possible mistake. You gave it the content + the reason, and it figured out the rest.

"Because it's important" does absolutely nothing. Claude can't work with it as there’s not much context to that.

Here are a few good examples instead:

For how it sounds:

"Write short sentences because most of my readers open these on their phone and skim while doing three other things."

(Bonus: this will likely also get you short paragraphs and front-loaded points automatically.)

For formatting:

"Don't use tables because I paste this into my email platform, and tables break when it sends."

For protecting trust:

"Don't promise specific results because this goes to clients in a regulated industry, and anything that reads like a guarantee gets flagged."

(Words like "guaranteed" and "proven" will likely vanish from results - without you having to list them.)

For keeping facts straight:

"Tell me plainly when you're not sure because I repeat these facts to clients, and a wrong answer costs me more than a missing one."

(Don’t rely on this though - always fact check results from AI.)

Custom Instructions (for reasons about YOU and your audience)

Click your initials → Settings → General → "Instructions for Claude." Anything you write here gets read at the start of every single conversation.

Project Instructions (for reasons about a specific project)

Inside any Project, click "Set project instructions." These only apply within that Project.

Set these up once, and every future first draft comes back already checked against these criteria.

Hope you enjoyed this hack -tiny tweak, massive difference!

I made $18,228.97 in July, and that's officially my best month ever! Again!

Last month I told you June's $17,066.28 felt surreal. So you can imagine how I'm feeling typing an even bigger number 4 weeks later. 🤯

But July wasn't a fluke, and it wasn't luck. It was compounding.

I had return UGC clients coming back, systems I set up starting to pay off and multiple income streams slowly starting to mature.

And in full transparency, it was also a seriously hard-working (i.e. exhausting) month. I'll dive into that a bit more at the end, because I'm expecting August to be a much lower month, on purpose.

First, let me show you exactly how the $18,228.97 came together.