Kirkpatrick at workshop scale for a single live event. Fictional client; nothing below has been run or measured. Targets are proposals to agree with the client before the first cohort.
L1 · Reaction (exit poll, 3 items, 40 seconds)
| Item | Proposed target |
|---|
| "Worth my time" (1–5) | most participants 4–5 (to agree; e.g. 80%) |
| "I will hold the conversation" (yes/maybe/no) | most "yes" |
| One thing to improve (open) | themed into the next cohort's changes |
L2 · Learning (rubric delta)
- Observer rubric scores captured per round (
03-observer-rubric.html: five lines, 1–5 each, total /25)
- Metric: mean round-2 total minus mean round-1 total. Proposed check: +1.0 or more
- Calibration first: in the fishbowl round, observers' line-1 scores should agree within one point; if they spread wider, the delta is noise, not learning
- Difficulty confound: if the round-2 scenario is rated harder, report it next to the delta
- Item-level read: the rubric line that lifts least becomes the next cohort's clinic focus
L3 · Behaviour (the commitment card does the work)
- T+1 pocket card with the two clinic moves
- T+10 manager pulse: one question, two clicks ("conversation held?" + "any visible change?")
- Proposed target: most booked conversations held within 10 working days (e.g. 70%), to agree with the client
- T+30 two-question pulse to team members (anonymous): "I get clear, timely feedback" trend
L4 · Results (deliberately modest)
A single workshop cannot claim movement in the engagement survey. The plan only commits to reporting the feedback-quality survey item alongside the cohorts that ran; if several cohorts run and the item rises, that is an association, never causation.
What the simulator adds (and does not)
simulator.html lets a facilitator rehearse four calls (the pair, the freeze, their own line-1 score, the derail) and see their consequence on scripted numbers. It is facilitator practice, not evidence of the workshop's effect; its numbers are invented for the simulation.
Reporting artifacts
- One page per cohort: L1 poll · L2 delta (with calibration spread) · L3 completion
- Facilitator retro notes → next version of the run-sheet