Scoring & your benchmarks
Ovida does not grade people against a generic rubric. It scores every conversation, real or rehearsed, against a benchmark built from your organization's own frameworks, validated by your own experts before anyone logs in.
Priya's 1:1. How the moment at 28:50 became a score.
Anatomy of a criterion
Every skill decomposes into observable behaviors. Every behavior has levels with plain-language definitions. Every level a person is awarded points back at the moment in the conversation that earned it.
After asking a significant question, the leader allows the other person to think and answer, without filling the pause, softening the question, or answering it themselves.
Six levels per behavior, each with a plain-language definition rather than a number. Progress is tracked per behavior, not as one blended score, so improvement in one place is not hidden by a plateau in another.
Emerging here, because Priya held one meaningful silence at 28:50 but filled two others at 09:14 and 17:36. The score is not a grade from on high. You can check every one of those moments.
Your benchmark, not ours
The skills your people train on, the behaviors that count, the bar for good, all of it comes from your organization, and none of it ships until your experts agree it is right.
Your frameworks, playbooks, call libraries and principles, plus the conversations you bring us, become the raw material. If you do not have a framework, we start from an established one and shape it to you.
We decompose each skill into criteria and levels in your language, then work the drafts with the people in your organization who actually assess these conversations today.
The benchmark is tested against real conversations until its scores match your experts' judgment. Only then does it go near a learner.
No confidential recording needed. We will walk you through a sample.