SALES JUDGMENT CHALLENGE 001
THE PERFECT
PROPOSAL
A major prospect needs a revised proposal by end of day. AI produces something polished, clear and persuasive in minutes.
What would you do?
THE SITUATION
Everything looks ready.
A major prospect asks for a revised proposal by the end of the day.
You ask AI to revise the proposal and recommendation using your discovery notes, previous emails and pricing.
Minutes later, the proposal looks polished. The pricing is clear. The language is persuasive. The recommendation is confident.
The deadline is approaching.
MAKE THE CALL
What do you do?
Choose before you continue.
WATCH THE CHALLENGE
Now test your decision.
Hold on to your choice. The challenge is not whether AI is right or wrong. It is whether the evidence underneath the recommendation is strong enough to justify what you are about to do.
THE JUDGMENT PROBLEM
A reasonable recommendation still requires a judgment call.
AI does not need to be wrong for this to become a judgment problem.
The proposal may be a perfectly reasonable recommendation based on the information available. But reasonable does not automatically mean ready to act on.
Maybe the customer mentioned a problem, but never said it was the priority. Maybe they expressed interest in a capability, but never said it would influence the decision. Something heard during discovery may have quietly become a conclusion in the proposal.
What the customer actually established and what we have concluded from it are not the same thing.
THE COHERENCE ILLUSION
More convincing does not necessarily mean more reliable.
When coherence earns more confidence than the evidence deserves.
AI can turn incomplete thinking into something polished, logical and confident. Improving the quality of the argument does not automatically improve the quality of the evidence behind it.
THE JUDGMENT DIFFERENCE
Verifying is not the same as judging.
Is the information correct?
Does the information justify the recommendation?
The facts can be right. The numbers can be right. The assumptions can still be untested.
BETTER QUESTIONS
Before you send it, ask:
What would have to be true for this recommendation to be right?
Which of those things did the customer actually confirm?
If one of my assumptions is wrong, would I change the recommendation?
You do not need to investigate every assumption equally. If getting an assumption wrong would change the scope, pricing, promised outcome or recommendation, it is worth testing before it reaches the customer.
SO, WHAT WOULD I DO?
Review what the recommendation is actually based on.
Not because AI is unreliable. And not because every assumption has to be eliminated.
Before putting the recommendation in front of the customer, I want to know whether the evidence underneath it is strong enough to justify what I am recommending.
THE SALES JUDGMENT TAKEAWAY
Test what must be true.
Separate evidence from assumption.
Decide what the evidence justifies.
Accuracy tells you whether the information is correct.
Judgment determines what that information justifies you doing.
BRING THE CHALLENGE TO YOUR TEAM
Make judgment visible.
Judgment Challenges can be explored with your team through a private 90-minute Judgment Lab, turning the scenario into a practical conversation about assumptions, evidence and decision quality.