Glossary

Behavioural evidence, defined

Stable definitions for the language used across GuruNow's moments, research and measurement.

Current as of 30 September 2026

Moment: A moment is a specific, recurring, uncomfortable conversation, defined tightly enough that you can walk into it, fail it and be observed doing either. It is not the Five Moments of Need learning-support model; here, the unit is the conversation a person must be able to handle.
Walk-into test: The walk-into test separates a moment from a competency. A moment passes if you could walk into it: a room, another party who wants something, a point where you have to act, an outcome you would recognise and a way to fail. "Telling a high performer their work has slipped when you know they've been trying" passes; "demonstrates emotional intelligence" fails.
Behavioural marker: A behavioural marker is a specific, observable feature of what a person said in a conversation, written so that it can be checked against a timestamped transcript.
Evidence classes: Evidence classes distinguish the kind of observation being made, so a practice attempt, a delayed attempt and a real conversation are not treated as equivalent evidence.
Capability and evidence axes: The capability axis describes how consistently the behaviours are demonstrated; the evidence axis describes the conditions under which that demonstration was observed. They are reported separately, never collapsed into one score.
Evidence rung: An evidence rung states how demanding the observation conditions were, progressing from briefed practice towards delayed, unbriefed and real-work evidence.
Held-out assessment: A held-out assessment is an unbriefed attempt using a new counterpart and behaviours that were not disclosed to the learner, used to test whether capability transfers beyond rehearsal.
Conditional credential: A conditional credential states the conditions under which performance was evidenced, including delay, counterpart difficulty and the behaviours observed; it is not presented as an unconditional claim of mastery.
Primary share finding: The primary share finding is that the problem leaders name is often not the one that recurs. A moment can appear in many cases while being the named problem in only a small share of them.
Head–mouth gap: The head–mouth gap is the difference between understanding what should be said and producing it in the conversation itself. Knowledge can be present while the observable behaviour is absent.
Role-play simulation: A role-play simulation is a digital practice environment in which a learner responds to a realistic counterpart and receives feedback on the resulting conversation.
Human-in-the-loop training: Human-in-the-loop training combines scalable digital practice with human judgement and support at the points where context, escalation or professional oversight requires it.

How can this help you now?

Giving open & honest feedbackPrivate to you

You0:02“Have you got a few minutes? I want to talk about Tuesday's client call.”

You named a specific occasion, not a pattern

You0:08“When the client asked about the delay, you said it wasn't our problem.”

You described what was done, not what the person is

You0:19“The client emailed afterwards asking for a new account lead.”

You said what the effect was, and on whom

For you

Learn the skill. Practise that tough conversation privately, now.

Try a simulation free
See the sentence behind every result

Practise, Measurement, Feedback, Confidence. It's in your grasp.

Try it now!

For your organisation

Honest data, because nobody performs for an audience

Find which Moments your leaders actually struggle with
Evidence mapped to your competency framework
Aggregate-only reporting

Psychological safety leads to better outcomes

Book a team demo