Which AI writes the best sales follow-up?

Generic follow-ups are why most prospects ghost. In this lesson you'll watch four AIs attempt the same task, see exactly why their first drafts all sound the same, then learn the two patterns that fix it — specific friction and the constraint stack.

You'll see real outputs, learn why they come out that way, then build a sharper prompt yourself. About 12 minutes.

01 · The prompt everyone writes

Almost everyone's first instinct looks like this — clear, reasonable, and the prompt 90% of people would write:

Before you scroll — what do you think four different AIs send back?

02 · Why they're all the same

This is the most important idea in the lesson. A language model predicts the most statistically likely next words. Give it a generic prompt and it returns the average of everything it has read — and the average follow-up email is, by definition, the most generic possible follow-up email.

So the sameness you just saw isn't the AI being lazy. It's the prompt asking for the average.

Your AI's answer comes out generic. Where's the real fix?

03 · Pattern one — specific friction

The single highest-leverage change to any follow-up: anchor it to one specific moment from the conversation. One sentence is enough. Watch the same task with and without it:

The model didn't get smarter. You gave it an anchor — one concrete detail — and it stopped generating from the average. That's the whole move.

You just saw the difference

Same goal, two prompts. Which one is the anchored version that produced the sharper email?

04 · Pattern two — the constraint stack

Specificity gets you most of the way. Constraints get the rest — the tighter the box you draw, the sharper the output. Here's the email we're aiming for:

Four layers got it there. Tap each one below and watch your own prompt sharpen:

The one line to keep: before you hit send, give the model Context, Specific friction, Goal, and Constraints. Specificity gets you ~70% of the gain; constraints get the last 30%.