Hey {{first_name}}
I just deleted 70% of my Initial Message Sherpa prompt.
And it got BETTER.
Not "just as good." Not "acceptable."
Better.
For context: the Initial Message Sherpa is my outreach message system that consistently gets 60% reply rates. Not cold DMs - these are thoughtful, personalised initial conversations that actually work.
The system always performed brilliantly. Once installed, my clients loved the results.
But that installation? It was like assembling furniture with instructions written in ancient Sanskrit.
4000 words of prompt engineering that required a PhD in patience to implement.
Then last week, something shifted.
I was learning new prompt engineering techniques in a coaching program when the instructor said something that stopped me cold:
"Most people don't have prompt problems. They have understanding problems."
That night, I opened my Initial Messa
ge Sherpa with fresh eyes.
And I saw it immediately.
Contradictions hiding in my examples. Instructions fighting each other. Clarifications that actually created confusion.
It was like looking at a recipe where you're simultaneously told to bake at 180°C and 220°C. Or worse - like oversalting a dish. Once it's in there, you can't take it back out.
The AI wasn't struggling because it lacked instructions.
It was struggling because I'd given it too many conflicting ones.
So I started cutting.
First the redundancies. Then the examples that contradicted my core principles. Then the "helpful" clarifications that were anything but.
2800 words hit the cutting room floor.
What remained: 1200 words of pure signal.
The result shocked me.
The same 60% reply rates. The same quality of conversation starters. But now it performs even better - faster responses, more consistent outputs, less compute required.
But here's what really got me: it became instantly installable for my clients.
Before, installing the Sherpa required a support call and crossed fingers. Now it takes five minutes and just works.
This is the paradox of mastery.
When you truly understand something, you can remove instead of add.
It's like cooking - anyone can throw twenty ingredients in a pot. But once you've added that seventeenth spice, you can't unmix it. You can't uncomplicate it. You're stuck with the chaos.
A master chef doesn't need 47 ingredients. They need 5 perfect ones that sing together.
And apparently, a master prompt doesn't need every possible instruction.
It needs the right ones, with nothing fighting against them.
I'm now going through every prompt in my library with a machete.
Not adding. Subtracting.
Finding the contradictions I couldn't see before. Removing the "helpful" parts that were actually harmful.
Each prompt is getting shorter. Each one is getting better.
Because simplicity isn't about starting simple.
It's about understanding something so deeply that you finally see what's truly essential.
Everything else is just noise you were too inexperienced to recognise.
The best prompt isn't the one where you can't add anything more.
It's the one where you can't take anything away.
P.S. Want the simplified Initial Message Sherpa plus the entire LinkedIn system and support to actually implement it? Join our community where you get everything - all my simplified prompts, systems, and real help making them work. Reply with SYSTEM and I'll send you the details. $47/month until September 14th, then the price increases.
