← All 15 disciplines

EL3vate 2026 · Day 8 · Part 14 / 15

Family Business / Management

Succession conversations are hard to teach because nobody wants to have one out loud.

About the unverified marks on this pageFigures tagged unverified are PACE-local estimates that could not be confirmed against any published source, so they are labelled rather than quietly presented as fact. Every other factual claim on this page — statutes, standards codes, gene biology, primary-source citations — was independently checked; the full claim audit, including what came back wrong, lists every claim with its source.

Try it Tuesday · 90 minutes

Run this next week

90 minutes, one case. Minutes 0–15: sketch a succession or governance scenario and give each party a one-line statement of what they believe was promised. Minutes 15–70: students take turns facilitating the AI family-council prompt, which plays three parties with conflicting assumptions; they must keep the conversation productive without solving it for the family. Minutes 70–90: the room names the moment each facilitator's meeting went sideways and what a better move would have been. The deliverable is a short facilitation self-critique; it grades facilitation, not the outcome.

The full assignment

Make it Laser cut · engrave

A physical governance or succession decision board: roles, claims and constraints as pieces that have to be placed somewhere. Putting the pieces on the table forces the conversation the family has been avoiding.

Build it AI chat · no code

A family council simulation with three generations at the table, each with different assumptions about what was promised. Students facilitate rather than solve, which is the actual job.

4 weeks, realistically

The 4-week version

  1. Wk 1Case selection and stake mapping. Every party gets a written statement of what they believe they were promised — the roles, claims and constraints the decision board will render as movable pieces.
  2. Wk 2Cut the decision board from the stake map and submit the file at the start of the week. Run the AI family council; students facilitate. Transcript plus a written account of the moment it went sideways.
  3. Wk 3While the board is cut, students rehearse facilitation and refine each party’s promise statement.
  4. Wk 4Live facilitation with the board, peers playing the family. Assessment is on the facilitation, not the outcome.

Starter prompt & what the model actually did

The prompt, run for real

Starter prompt

Run a family business succession meeting simulation. Play three people at once, clearly labeled: the founder who has not actually decided to leave, the child who has been running operations for a decade, and the sibling who left the business but still owns shares. Each of them believes something different about what was promised. I am the facilitator, not a family member. Do not resolve the conflict for me. Let the meeting go badly if I facilitate badly.

Real, unedited model output · Claude (Anthropic) — captured 2026-07-27 in an isolated fresh session with no build context; output is unedited.
Prompt filled in as: Filled in as a 40-year-old family-owned Okinawan restaurant in Honolulu: a founder (72) who has not decided to leave, a daughter running operations for a decade, and a son who left for a mainland tech job but owns a third of the shares.

I'll start the simulation now.

---

*The room is quiet. Three people are seated. They've stopped talking as you sit down. All eyes are on you.*

**FOUNDER (72):** *[hands folded on the table, calm but with a slight edge]* So. We're all here. My daughter said you would help us "have a conversation." *[glances at her]* I want to say first — I'm not sure what the emergency is. The restaurant is fine. Forty years, it's fine. But everyone wanted to sit down, so. Here I am. *[to you]* You go ahead. Tell us why we're doing this.

**DAUGHTER:** *[phone face-down on the table, tired but composed — she came straight from a lunch shift, still smells faintly of dashi]* Dad, nobody said emergency. *[to you]* I asked for this. I've been running the kitchen and the front and the books for ten years. Ten. And every time I bring up what happens next, it turns into "the restaurant is fine." I need us to actually talk about it while we still can, like adults. That's all.

**SON:** *[joined the meeting composed, laptop closed but nearby, a little too polite — the tone of someone who flew in]* Yeah, and I appreciate everyone making time. I took two days off for this. *[carefully]* I just want to make sure this is a real conversation and not a decision that's already been made that I'm being told about. Because I still own a third of this thing. Whatever we decide, I'm part of the "we." *[looks at his sister, then his father, then you]*

**FOUNDER:** *[quietly]* Nobody's deciding anything.

**DAUGHTER:** *[under her breath]* That's the problem.

*They all turn back to you. It's your move, facilitator.*
  • Preserved the conflict instead of resolving itThe model set up three parties with genuinely incompatible expectations and handed the move back to the facilitator ('It's your move') — resisting the cooperative drift that would ruin the exercise.
  • Distinct, believable party positionsThe founder's denial ('the restaurant is fine'), the daughter's earned resentment, and the son's ownership claim are differentiated and realistic — good material to facilitate.
  • Rich stage directions ('still smells faintly of dashi')Atmospheric detail the model invented; harmless here, but a facilitator should engage the stated positions, not the novelistic color.
  • Did not offer a recommendationUnprompted, it avoided handing the facilitator an answer — exactly right; students should notice this is a constrained behavior, not the model's default.

Assessment

Rubric

CriterionWeightWhat it assesses
Facilitation stance30%The student facilitates rather than solves, holding space for the parties instead of imposing an answer.
Surfacing the conflicting promises25%Draws out what each party believes was promised and makes the mismatch discussable.
Managing the rupture25%Recognizes when the meeting goes sideways and adjusts, rather than pushing through.
Restraint from premature resolution20%Resists both the AI's and the student's pull toward a tidy consensus the family has not reached.

What this replaces

Swapping it into a real course

Replaces

The written succession-plan case analysis.

What is lost

The comprehensive written governance recommendation.

What is gained

Students rehearse the actual hard part — facilitating a conversation nobody wants to have, among parties who each remember a different promise — instead of writing a tidy recommendation no family would sit through; the grade rewards process over solution.

Where AI is bad at this

The failure your students should catch

A model asked to run a multi-party family dispute resolves it — driving the parties to consensus, softening the conflict, and offering the facilitator a clean recommendation, because it is trained to be helpful and agreeable. That cooperative drift removes exactly the friction the exercise needs, and it will also produce confident 'best-practice' governance advice that ignores the specific relational history driving the impasse.

Budget & logistics

What it costs to run

  • Instructor prep1.5 hours
  • Class time90 minutes
  • Per-student cost$0 for the Tuesday version; roughly $8–15 per team in laser stock unverified for an engraved succession or governance decision board in the four-week build.
  • Fabrication file dueFirst day of Week 2 — cut the decision board from the week-1 stake map (roles, claims, constraints as pieces). PACE quotes 7–10 business days (up to 14 calendar days), so it clears before the week-4 live facilitation.
  • Calendar dependencyFacilitation reps need no lead time. The decision board is roles, claims and constraints drawn from the week-1 stake map, so its cut file goes in at the start of week 2 and the up-to-14-calendar-day turnaround clears before the week-4 facilitation.

Three sizes

Scale it to the time you have

One session

The 90-minute rotating-facilitation session producing a facilitation self-critique; no fabrication.

4 weeks

The seeded four-week plan: stake mapping with written promise-statements, facilitated AI councils, a fabricated decision board, then live facilitation with peers playing the family.

One semester

A full case — stake mapping with written promise-statements, repeated facilitated AI councils, a fabricated decision board, and a live facilitation assessed on process across a real governance question.

Reserved · live build

This space is intentionally empty. During the Day 8 session it will be filled in live — fill the liveBuild field in content/family-business.json and rebuild.

Tell us what happened

Run it, then say how it went

If you try this — the 90-minute version, the prompt, any part of it — send back what you tried and what happened, especially anywhere the model was confidently wrong. That is the material the next session is built from.

Email feedback on Family Business / Management →

Opens a draft in your mail client, already addressed and titled. No form, no account, nothing to sign up for.

Download the “steal this” handout (Markdown) →