Skill measured, not guessed
Every attempt is scored 0–10 across four dimensions — each with written reasoning — plus hallucination detection, pass/fail gating and analytics your managers can act on.
See exactly who's ready to sell and who needs coaching
A single "good job" score tells a manager nothing. Rehearse breaks every conversation into the four skills that win or lose deals and conversations — and scores each one independently, with the reasoning behind the number.
- 1Product Knowledge — were the facts right and on-point?
- 2Needs Analysis — did they uncover the real need?
- 3Objection Handling — did they address resistance well?
- 4Closing & Coverage — did they advance and cover what mattered?
Why a buyer cares: dimension-level scores tell you what to coach, not just who failed — so improvement is targeted, fast and measurable.
Built so the number holds up to scrutiny
If you're going to gate progression on a score, that score has to be trustworthy. Here's how Rehearse earns that trust.
Hallucination detection
Rehearse flags claims a rep made that contradict your knowledge base — the confident-but-wrong answers that lose deals and create compliance risk. Why it matters: a fluent rep who invents facts is worse than a hesitant one. This catches it.
Retry-verified scoring
When a score looks contradictory, Rehearse re-checks it before it's shown. Why it matters: consistency. Managers and reps need to trust that the same performance always earns the same grade.
Pass/fail gating
Set a pass mark per scenario; reps must clear it to be signed off. Performance categories run from EXEMPLARY down to HALLUCINATION. Why it matters: a consistent bar means no one reaches a customer before they're ready.
False-memory guard
Contested turns are verified before display, so learners never see the AI contradict itself. Why it matters: a single visible AI mistake destroys credibility — and the value of the score.
From one attempt to whole-team visibility
Roll the scores up and you get the picture L&D and sales leaders have been missing.
Per-scenario
Attempts, average score, pass rate and dimension averages — instantly see which scenarios trip people up.
Per-learner progression
First attempt versus latest — proof that practice is producing real improvement, person by person.
Drill-downs
Open any attempt to read the transcript, the per-dimension reasoning and any flags raised.
See a scorecard from your own playbook
Book a demo and we'll score a live roleplay against your real products and standards.