SALES JUDGMENT CHALLENGE 006
THE SHARED
PICTURE
An AI account-prioritization tool ranks a long-time reference account low, because its objective is expected revenue. The system saw everything — the advisory board seat, the case studies. It just wasn't built to value them.
What would you do?
THE SITUATION
The system saw everything. It just wasn't built to value it.
An AI account-prioritization tool ranks every active account by expected revenue. The system has access to the full account record — it knows this customer sits on the product advisory board, appears in two published case studies, and has been a reference account for years.
But its prioritization objective is expected revenue. On that measure, the account ranks low. The tool recommends deprioritizing it: reduced service tier, no discretionary support hours, off the QBR calendar.
The account manager objects — not because the AI is missing something, but because she believes revenue is the wrong measure for this account entirely.
Both sides are working from the same facts.
MAKE THE CALL
What do you do?
Choose before you continue.
WATCH THE CHALLENGE
Coming November 16.
The video for this challenge publishes November 16. Check back then — or work through the situation and questions below in the meantime.
THE JUDGMENT PROBLEM
A ranking isn't a fact. It's a fact plus an objective.
This isn't a story about someone missing information. The AI didn't miss the advisory board seat or the case studies. It saw all of it.
Its objective simply gave that information no weight. The disagreement isn't about the facts — it's about what the system was built to optimize.
The AI didn't miss the strategic relationship. Its objective gave that relationship no weight. The system worked exactly as designed.
The question is whether the design reflects what the business actually values.
THE OBJECTIVE PROBLEM
A ranking isn't a fact. It's a fact plus an objective.
Shared facts don't guarantee shared conclusions, because agreeing on the data was never the hard part here.
AI can optimize an objective perfectly. Judgment is deciding whether it's the right one.
THE JUDGMENT DIFFERENCE
Accurate is not the same as agreed.
Expected revenue.
Is that still the right thing to optimize?
Same facts, different conclusions. The gap isn't information — it's the objective.
BETTER QUESTIONS
Before you resolve this account, ask:
Which outcomes should this ranking actually be optimizing?
Which of those can the system measure reliably?
Who decided the trade-off between them?
This isn't about who has better information. It's about what the system was built to value — and whether that's still right.
SO, WHAT WOULD I DO?
None of these is free — and none is obviously the responsible one.
A keeps a consistent, measurable process — and formally decides that reference value is worth zero, whether or not anyone meant to decide that. B protects a relationship the model can't see the value of, but every account manager believes their account is the exception. C resolves today's account, but leaves the objective untouched, so the identical argument returns with the next strategic account. D sounds like the responsible fix, but someone now has to decide how much revenue to trade for strategic value — a measure far harder to define consistently than revenue ever was.
Expanding the model doesn't remove the judgment call. It just moves it into the model's design, where it's harder to see and easier to forget was ever made.
THE SALES JUDGMENT TAKEAWAY
Don't ask who's right when both sides see the same facts.
Ask what the system was built to optimize.
Know who owns that trade-off before you change it.
AI can optimize an objective.
Judgment is deciding whether it's the right one.
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 objectives, account prioritization and what a ranking tool is actually optimizing for.