A call review rubric

What it is

A short scorecard for reviewing sales calls. It lists the moments that matter on a call, what good looks like for each, and a simple score. Managers use it in one-to-ones, and reps use it to review their own calls.

This version is written for first discovery calls. Copy it and change the sections for demos or negotiation calls.

When to use it

In weekly one-to-ones when reviewing a recorded call.
When onboarding new reps, to show what a good call looks like.
When two managers give reps conflicting feedback.
When setting up AI call scoring, to define what the model should look for.

Steps

Agree the sections of the rubric. A common starting set:

Section	What good looks like	Score (1 to 3)
Opening and agenda	The rep confirms time, states a purpose and agenda, and asks the buyer what they want to cover.
Discovery of the problem	Open questions get the buyer describing the problem in their own words, with follow-ups that go a level deeper.
Quantifying impact	The rep asks what the problem costs and leaves with at least one number the buyer gave.
Decision process and people	The rep learns who else is involved, who signs off and what the buying steps are.
Next step agreed with a date	The call ends with a specific next step and date the buyer has agreed to, ideally booked on the call.
Talk and listen balance	The buyer does most of the talking, and the rep summarises back what they heard.

Adjust the “what good looks like” sentences to fit your motion. They describe a score of 3.
Calibrate: two managers score the same call separately, then compare and adjust the wording.
Use the rubric in one-to-ones. Pick the lowest-scoring section and coach only that.
Review the rubric every quarter.

Good example

A manager scores a discovery call with the rep. Discovery and next steps score well, but quantifying impact scores 1. They spend the one-to-one on one moment where the buyer mentioned a problem and the rep moved on without asking about its cost.

Weak example

A manager listens to a call and gives the rep ten pieces of feedback across every part of the call. The rep can’t act on all of them and changes nothing.

How AI helps

Scoring at scale. Once the rubric is written, a model can score calls against it and highlight the moments behind each score.
Calibration. Compare AI scores with manager scores on the same calls to find vague rubric wording.
Keep the conversation human. Use AI scores to choose which call to review. The coaching conversation itself stays with the manager.