Run-sheet for the facilitator (fictional client, not yet run): minute-level plan, facilitation moves, and recovery lines for hard moments. Written to be executable by a second facilitator without me in the room — the handoff test.
Moves: Stand, no slides. Frame: "This is a practice space. Anything clumsy here stays here. We optimize for attempts, not polish." Norms on flipchart: Phones face-up (we'll use them at the end). Pass is always allowed. Feedback on the feedback is by rubric, not taste.
Recovery line if energy is low: "Raise a hand if you've ever given feedback that changed nothing. … That's our shared project for 90 minutes."
Deck: 12 statement cards ("You're lazy" / "Report was 2 days late" / "You don't respect deadlines" / …). Pairs sort into two piles. Debrief by asking a pair for a contested card — contesting IS the learning.
Facilitation move: never confirm answers yourself first; route to the room: "Which pile, and what makes you sure?" Watch for: the person who sorts by intent — that's your Round-1 role-play volunteer later (they have the most to gain, gently).
Micro-lecture (4 min, flipchart): Situation (time+place, specific) → Behavior (observable, camera-test) → Impact (on you/team/work, owned feeling) → Commitment (their words, their date). Drill (11 min): 3 prompts on table cards; each person rewrites one judgment into SBI-C, pairs swap and camera-test each other ("Could a phone camera have recorded it?").
Hard moment prep: someone asks "what if the behavior is attitude?" → "Attitude is our interpretation. What did the person DO that you interpreted as attitude? Start there."
One volunteer pair in the middle (ask for volunteers; say aloud that anyone can pass). Everyone else scores line 1 only, Specificity, silently. Then all observers show their number at once. Calibration move: if scores spread more than one point, ask the highest and the lowest scorer what they heard. Agreeing within a point is what makes the round-two delta worth reporting. Recovery line if the giver freezes: pause the clock, not the pair: "Let's hold there. This is the most useful ten seconds of the round. What did you actually see, when, on which report?" Do not skip them, and do not take over the sentence yourself first.
Scenario cards (realistic, fictionalized):
Structure per round: 4-min conversation → 2-min observer rubric feedback (1–5 on five lines: specificity, camera test, impact ownership, commitment quality, tone) → 1-min doer self-score. Your moves: float, time-box loudly ("30 seconds — land the commitment"), harvest one rubric line per triad on the wall chart.
Two scripted interruptions played by co-facilitator or volunteer:
Teach two moves only: Bridge (acknowledge → return: "Workload might be real — AND the demo slipped. Let's hold both.") and Refocus on the record ("The past isn't on trial here; the next demo is. What would make it different?"). Whole-group: 3 rapid volleys, room suggests responses, you play the defender harder each time.
Purpose: measurable delta. Same rubric. Observers watch specifically for the clinic moves.
If someone takes the floor with a long story: acknowledge and park it on the board ("Hold that, it's item two, one minute"). Shutting it down flat saves the time but costs the room's warmth; letting it run costs the commitment cards. Commitment card: One real conversation · one real name (privately) · one date within 5 working days · which SBI-C line they'll open with. Then: "Phones out. Open your calendar. Book it now — 30 minutes, before Friday." Wait. The room goes quiet and slightly horrified. That silence is the workshop working.
Close: exit poll QR on screen (3 questions, 40 seconds), thank the observers by name.
Prefer the source? Download the raw markdown.