In partnership with
Cut Lead Review From Hours To Minutes
Sign up for a free trial of Attio, the agentic CRM.
Ask Attio to build a daily workflow that surfaces the deals that need your attention today, like anything with a stage change, a recent reply, or a new signal in the last 24 hours.
Review your pipeline in Claude, synced live from Attio via MCP.
That's it.
GM to the Top 1% ☕
A rep sent me two emails last week and asked which one the AI wrote. One was sharp, specific, and clearly about the buyer. The other was smooth, generic, and clearly about nobody.
The AI wrote both. Same model, same rep, same buyer. The only thing that changed was the prompt behind it.
That is the entire game right now, and almost nobody is playing it on purpose.
💡 THE MODEL IS NOT THE VARIABLE. THE INPUT IS.
Here is the stat that gets misread constantly. Teams using AI are 3.7x more likely to hit quota. Everyone reads that as "buy the tool, hit the number." That is not what it says. The lift shows up only when the input is disciplined. Feed a world class model a lazy prompt and you get world class slop, faster.
This is where the 10 billion dollars in ungoverned AI cost actually leaks out. Not from bad models. From good models pointed at vague instructions by people in a hurry. The output looks confident, reads fine, and quietly says nothing a buyer hasn't heard a hundred times.
Your buyer can smell it instantly. Generic AI outreach is now a signal, and the signal it sends is "this person did not think about me." That is worse than sending nothing.
The reps pulling ahead are not using a secret model. They are using disciplined inputs. They give the machine the deal context, the buyer's actual situation, the specific tension, and a clear job. The prompt is where the judgment lives now. The prompt is the craft.
If the input is generic, the output is a liability. If the input is specific, the output is a weapon. Same tool. Completely different rep.
🔧 THE SIGNAL PROMPT
Turn a slop machine into a signal machine with four inputs.
1. Load the real context: Paste the actual deal notes, the buyer's words, the last call summary. Vague in, vague out. Specific in, specific out.
2. Name the tension: Tell the model the exact friction, the competitor, the objection, the political landmine. Generic prompts skip the tension, which is why they read like everyone else.
3. Give it a job, not a topic: "Write a follow up" is a topic. "Get the CFO to grant fifteen minutes by addressing her cost concern from Tuesday" is a job.
4. Force a human pass: Never send the first draft. Read it as the buyer and cut anything a competitor could have sent to anyone. That last edit is the whole difference.
🎯 THIS WEEK'S HOMEWORK
Take one message you would normally fire off with a lazy prompt, and rebuild the input with real context, the specific tension, and a clear job. Send that one. Then compare the reply you get to your usual. That gap is your reason to fix every prompt you use.
❓ QUESTION OF THE DAY
Would your last AI-assisted message survive the buyer asking, could you have sent this to anyone?
Reply with your honest answer. I read every one.
See you tomorrow.
Edward
Founder, Morning Sales
P.S. Morning Sales goes out every morning to a room full of AI-native enterprise sellers who actually open, read, and act. If you want your brand in front of them, a single placement or a four week run is the highest-signal ad space in B2B sales. Book it here: https://app.beehiiv.com/direct_sponsorships/3760348f-b319-4d3d-a479-f7934571d587