AI scoring and rubrics · Autessa Roleplay
Evidence-linked scores
Every AI score in Autessa Roleplay is evidence-linked: open a criterion and you see the scorer's rationale and the exact quotes it was based on, and clicking a quote jumps to that turn in the transcript and highlights it. Trainees can see why they got a score, and coaches can check it in seconds.
The problem it solves
People do not change because of a number
A 2 out of 5 on "Question quality" tells a rep nothing. What they need is the moment it happened: the question they should have asked, or the one they did ask that worked. Scores without evidence also make coaches second-guess the AI.
What you get
- Every criterion on the report expands to its rationale.
- Evidence quotes are the exact words, with who said them.
- Each quote links to its turn in the transcript; the turn scrolls into view and flashes.
- Framework criteria show each part; knowledge checks show each claim with its verdict.
- A criterion that could not be scored says "Not scored" rather than guessing.
- Coach adjustments appear next to the evidence they disagree with.
The run-through
How does evidence-linked scores work?
- SCENE 1
Open a criterion
On the report, expand any row under "Scored against the rubric".
- SCENE 2
Read the reasoning
See the value, the rationale and the quotes.
- SCENE 3
Jump to the moment
Click a quote to scroll the transcript to that turn.
- SCENE 4
Agree or adjust
A coach can adjust the score with a reason if the evidence says otherwise.
The cue sheet
What each scored row carries
| Option | What it does |
|---|---|
| Value | A rating, Achieved or Not achieved, a framework average, or Feedback. |
| Rationale | The scorer's explanation in plain language. |
| Evidence quotes | Exact quotes with the speaker and the turn they came from. |
| Turn link | Scrolls the transcript to the turn and flashes it. |
| Adjustment | A coach value and reason, when one was added. |
The cast
Who can do what?
| Who | Can |
|---|---|
| Trainee | Read the rationale and evidence on their own reports. |
| Coach | Read evidence on everyone's reports and adjust scores. |
What the AI does
Behind the scenes
The scorer is asked to return the quotes it relied on with each score. Delivery numbers like pace and filler words are counted separately, and the scorer is told never to recount them.
Questions
Questions about evidence-linked scores
Can trainees see why they got a score?
Yes. Every criterion opens to its rationale and the exact quotes it was based on, each linked to the transcript.
What if the AI could not score a criterion?
It shows "Not scored" and is left out of the overall, instead of guessing. A coach can retry scoring.
Is the evidence from the full conversation?
Yes. Scoring reads the full two-sided transcript of the session.
100% customizable, code and all
Make Autessa Roleplay fully your own.
Use it out of the box, then change anything. Ask Autobot, Autessa’s AI builder, for a new field, screen, report, workflow or agent and it builds it into your copy of the app. Or open the code and edit it yourself. Nothing is locked: the app, its data and its code are yours.
Start free and make it yours