How to Score a Sales Call: A 7-Behaviour Rubric

By Kavyashree SB, Technical Writer November 9, 2026 8 min read

A practical perspective for connecting this idea to your revenue team’s next move.

Editorial calibration bench measuring one sales call across seven behaviour instruments, each tied to an evidence slip and aligned between reviewers for coaching.

To score a sales call, rate it on a small set of selling behaviours, give each a score from 0 to 100, and back every score with a quote from the call. A good rubric covers discovery, active listening, value articulation, objection handling, product and message accuracy, next steps and qualification. The quotes are what make a score coachable: a manager and a rep can argue about a number, but not about what was said.

This guide gives you the seven-behaviour rubric, what strong and weak look like for each, the call metrics worth reading beside the scores, and a calibration routine so two managers score the same call the same way. It suits sales managers, enablement teams and reps who review their own calls.

Last updated: 9 November 2026

Why score sales calls at all?

Most coaching starts from a manager’s memory of one call. Scoring makes it consistent: every call is judged against the same behaviours, so you can see which skills a rep has, which ones the team lacks, and whether training changed anything. A scorecard also makes call review faster, because the reviewer knows what to listen for.

Score behaviours the rep controls, not outcomes. Whether the deal closed depends on many things; whether the rep confirmed a dated next step does not.

The seven-behaviour sales call scoring rubric

Score each behaviour from 0 to 100 and write down the line from the call that justifies it. Skip a behaviour that did not apply (a first discovery call may have no objections) rather than scoring it zero.

Behaviour	The question to answer	Strong evidence sounds like	Weak evidence looks like
Discovery	Did the seller uncover the buyer’s situation, problem and impact with open questions, and go a level deeper than the first answer?	Asks why the problem matters now and what it costs; follows up on vague answers.	Pitches before asking; accepts “we want to improve efficiency” without a follow-up.
Active listening	Did the seller build on what the buyer said, summarise it back and leave room for the buyer to talk?	Paraphrases the buyer’s words; next question uses something the buyer just said.	Interrupts; asks a question the buyer already answered; long monologues.
Value articulation	Did the seller connect capabilities to the outcomes this buyer named, in the buyer’s terms?	“You said onboarding takes a quarter; here is how teams shorten that.”	Feature tour unrelated to anything the buyer raised.
Objection handling	Did the seller acknowledge the concern, ask what sits behind it, answer it, and check it landed?	Clarifies the objection before answering; confirms “does that address it?”	Argues, discounts immediately or moves on without answering.
Product and message accuracy	Were claims about the product, pricing and roadmap correct and on-message?	States what the product does today and says what it does not.	Promises roadmap items, misstates pricing or contradicts approved messaging.
Next steps	Did the call end with a dated, owned next step that the buyer agreed to?	Books the next meeting on the call with named attendees and a purpose.	“I’ll send some info over” with no date or owner.
Qualification	Did the seller confirm the facts your framework needs, such as MEDDICC or BANT: decision process, economic buyer, criteria, timing?	Asks who else signs off and how decisions are made; confirms the timeline is real.	Leaves the call without knowing who decides or by when.

How to turn evidence into a 0 to 100 score

Use four anchors so scores mean the same thing across reviewers:

0 to 25: the behaviour was missing or done in a way that hurt the call.
26 to 50: attempted, but shallow: one question where three were needed, an objection acknowledged but not answered.
51 to 75: done well enough to move the deal, with one clear thing to improve.
76 to 100: done in a way you would play to the team as an example.

Write one strength and one thing to improve per call, not per behaviour. Reps act on one change at a time.

Weight the behaviours by call type

Not every behaviour matters equally on every call. On a discovery call, discovery, active listening and qualification carry most of the weight; on a demo, value articulation and product accuracy; on a pricing or negotiation call, objection handling and next steps. Tag each call with its type before scoring so you compare like with like.

Call metrics to read beside the scores

A transcript also gives you metrics that need no judgement. They do not score the call on their own, but they point at where to listen:

Talk ratio: how much of the call the seller spoke. A seller who dominates a discovery call is usually not discovering.
Longest monologue: the longest stretch the seller spoke without a pause for the buyer. Long monologues often hide a feature tour.
Longest customer story: the longest stretch the buyer spoke. Buyers talk longest when the question was good.
Questions per hour: a rough guide to how much the seller asked, read together with whether the questions were open.
Speaking pace, filler words and speaker switches: signs of nerves, rambling or a conversation that never found a rhythm.

Set your own targets from your best reps’ calls rather than from generic benchmarks: the right talk ratio for a technical demo is not the right one for a first call.

How to calibrate call scores between managers

Pick three recorded calls of different quality and have every manager score them alone, with quotes.
Compare scores behaviour by behaviour. Where managers are more than one anchor band apart, discuss the quotes each one used.
Write down the decisions as examples in the rubric: “this line is a 40 on objection handling, because the concern was acknowledged but not answered.”
Repeat monthly with new calls, and whenever you change the rubric or add an AI score.

If you use AI scoring, calibrate it the same way: score a few calls by hand, compare with the AI scores and quotes, and check that the evidence it cites is the evidence a manager would cite.

Turning call scores into coaching and training

Scores are only useful if something changes. Review the lowest-scoring behaviour per rep each week, play the quoted moment in a one-to-one, and assign practice on that behaviour before the next live call. Across the team, a behaviour that scores low for most reps is a training problem, not a coaching one: build a short course or role-play for it and check the scores move on later calls.

Scoring calls with software

Manual scoring works for a handful of calls a week. Beyond that, conversation intelligence software records, transcribes and scores calls automatically. Revspire Conversation Intelligence scores every call from 0 to 100 on the seven behaviours above, with the quotes behind each score, and reports the transcript metrics listed here; it records through a Mac and Windows desktop app with no meeting bot, and the scores feed Revspire’s sales training, where reps practise the weak behaviour with an AI buyer. Notes stay in Revspire, with no CRM sync today. For other options, see our guides to AI call scoring software and conversation intelligence software.

Free templates: the sales call scorecard template puts this rubric in Word, Excel and PDF, and the sales call notes template covers what to capture during the call.

Frequently asked questions

How do you score a sales call?

Rate the call on a fixed set of selling behaviours, such as discovery, active listening, value articulation, objection handling, product accuracy, next steps and qualification. Give each a score from 0 to 100 using shared anchors, quote the line from the call behind each score, and note one strength and one thing to improve.

What should a sales call scorecard include?

The behaviours you score, a question that defines each one, anchors for what low and high scores mean, space for the quote behind each score, the call type, and one strength and one improvement. Keep it to about seven behaviours so reviewers can score a call in a few minutes.

What is a good talk ratio on a sales call?

It depends on the call type. Buyers should do most of the talking on discovery calls, while sellers talk more on demos. Set targets from your own best reps’ calls rather than a generic figure, and read talk ratio beside the discovery and active listening scores.

Can AI score sales calls accurately?

AI can score calls consistently and at scale, and the useful tools show the quote behind each score so you can check it. Calibrate it like a new manager: score a few calls by hand, compare, and adjust the rubric where the evidence differs.

How many calls should a manager score each week?

Enough to coach every rep on at least one call a week. With manual scoring that usually means a sample; with automatic scoring, every call is scored and the manager reviews the lowest-scoring moments.