Menu
About This ExerciseFor ParticipantsFor FacilitatorsFAQ
Facilitator Guide

Adjudication Rules

Adjudication Rules#

Individual Participant Decision Model (V7.4)#

Critical Change: Each round, each participant makes individual decisions for their assigned industries (1 or more industries per participant; recommend 2). Unlike V6 (team deliberation), V7.4 has individual participants deciding for their industries directly.

Scoring Logic:

  • Explicit Industry Decisions (If Submitted): Scored normally using banded framework ({-2, 0, +2} per dimension, +/-3 if red-flag)
  • Industries Without Explicit Actions: Receive pre-defined base case fallback scores from fallback bank (small fixed deltas: +/-1 per dimension, deterministic, plausible). Fallbacks are automatic and deterministic.

Each industry decision is scored independently on the same 3 dimensions. There is no "sector-level" decision; all scoring happens at the industry level. The 11 industries are:

  1. Retail — scored using retail-specific context (omnichannel, demand forecasting, inventory)
  2. CPG — scored using CPG-specific context (brand management, R&D cycles, DTC)
  3. Healthcare Provider — scored using provider-specific context (clinical AI, FDA, patient safety)
  4. Healthcare Payer — scored using payer-specific context (claims automation, fraud detection, MLR)
  5. Finance — scored using finance-specific context (underwriting, trading, fair lending, fraud)
  6. Consulting — scored using consulting-specific context (copilot adoption, vertical AI expertise, pricing models, junior talent pipeline)
  7. Law — scored using law-specific context (billable hour model, bar rule compliance, malpractice liability, associate leverage)
  8. Manufacturing — scored using manufacturing-specific context (predictive maintenance, OT/IT integration, union relations)
  9. Logistics — scored using logistics-specific context (route optimization, autonomous vehicles, driver adoption)
  10. Big Tech — scored using Big Tech-specific context (cloud, ads, devices, enterprise software; excludes AI lab/model development)
  11. B2B/B2C SaaS — scored using SaaS-specific context (AI feature integration, pricing pressure, startup disruption)

Participants decide individually; facilitators score each industry decision on its own merits, in its own context. Fallback industries do not require participant justification; facilitator applies the pre-defined score from the fallback bank.


Scoring Framework#

Each industry decision is scored on three dimensions:

DimensionDefinitionRange
Strategic FitDoes the decision align with industry fundamentals, competitive positioning, and the macro scenario?-3 to +3
Execution RiskCan the organization execute this decision within the timeframe? Are there organizational, technical, or capital constraints?-3 to +3
Tail RiskDoes this decision expose the organization to downside scenarios (regulatory, competitive, reputational, systemic)?-3 to +3

Total per decision: Sum of three dimensions (range: -6 to +6).


Banded Scoring During Play#

Default scoring bands during live rounds:

DimensionTypical Bands
Strategic Fit{-2, 0, +2}
Execution Risk{-2, 0, +2}
Tail Risk{-2, 0, +2}

Why banded? Prevents false precision. Participants understand three clear buckets per dimension, not nine.

Exception: If a red-flag/plausibility trigger fires (see Section 2), unlock +/-3 scoring.


Scoring Guidance by Dimension#

Strategic Fit (-3 to +3)#

ScoreInterpretationExample
+2Decision directly captures the core competitive opportunity or mitigates the core threat in this scenario.Retail deploys AI demand forecasting when competitors doing same; Finance deploys AI underwriting when profitability depends on speed; Consulting builds vertical AI expertise when clients demand it.
0Decision is aligned with industry fundamentals; reasonable strategic move but not game-changing.CPG launches direct-to-consumer AI-driven personalization; Manufacturing invests in predictive maintenance; Law pilots AI-assisted research tools.
-2Decision is misaligned with scenario; exposes participant to opportunity cost.Big Tech holds on AI product integration during copilot adoption wave; Healthcare delays diagnostic AI investment during regulatory uncertainty; Law delays all AI adoption waiting for bar rule clarity.

Execution Risk (-3 to +3)#

Execution Risk maps from banded inputs: Spend/Commitment, Time-to-Impact, Execution Complexity, Dependency, and Scale.

ScoreInterpretationBand Combination Example
+2Decision is straightforward to execute; minimal organizational change; available capital and talent.Absorbable spend + 0-3mo + Low complexity + Mostly internal + Pilot -> +2 Execution Risk
0Decision is feasible but requires some organizational change and talent.Material spend + 3-12mo + Medium complexity + Mostly internal + Regional -> 0 Execution Risk
-2Decision faces material execution challenges; significant change management, talent gaps, or capex.Material/Transformational spend + 1-2 years + High complexity + Vendor/Partner + National -> -2 Execution Risk
-3 (Exception)Decision is extremely difficult to execute; severe constraint likely failure.Transformational/Existential spend + 2+ years + Very high complexity + Ecosystem shift required + Global -> -3 Execution Risk

Examples:

  • Deploying proven AI copilots (Absorbable, 0-3mo, Low, Internal, Pilot) = +2
  • Enterprise-wide AI deployment with vendor integration (Material, 3-12mo, Medium, Vendor, National) = 0
  • Building proprietary AI from scratch (Transformational, 2+ years, Very high, Ecosystem, Global) = -2 to -3

Tail Risk (-3 to +3)#

ScoreInterpretationExample
+2Decision includes hedges or defensive measures; upside potential with limited downside.AI trading with circuit breakers; diagnostic system with human review; pilot-phase deployment; Law AI with human attorney review of all output.
0Decision is neutral on tail risk; no clear upside or downside hedges.Phased deployment with standard governance; moderate automation with retention plan.
-2Decision exposes organization to material tail risk; likely to backfire if scenario changes.Aggressive headcount cuts without severance; autonomous systems with no rollback; all-in bet on unproven technology; Law firm deploys AI-generated briefs without attorney review (malpractice exposure).

Band-to-Score Translation Reference#

Use this table to quickly map banded inputs to expected Execution Risk scores:

SpendTimeComplexityDependencyScaleTypical Exec Risk
Absorbable0-3moLowInternalPilot+2
Absorbable0-3moMediumInternalRegional+1
Material0-3moLowInternalRegional+1
Material3-12moMediumInternalRegional0
Material3-12moHighVendorNational-1
Transformational3-12moMediumVendorNational-1
Transformational1-2yrHighRegulator/UnionNational-2
Transformational2+yrVery HighEcosystemGlobal-3
Existential2+yrVery HighEcosystemGlobal-3

Quick rule: Absorbable + 0-3mo + Low = +2. Each band downgrade (Material, Transformational, Existential) or timeline extension (3-12mo, 1-2yr, 2+yr) or complexity increase (Medium, High, Very High) or dependency widening (Vendor, Regulator, Ecosystem) shifts score down by roughly 0.5-1.0 per dimension.


Red-Flag Triggers (Band Combinations)#

Red-flag triggers now reference band combinations instead of granular details:

CategoryRed-Flag BandsExample
Timeline MisalignmentGlobal scale + 0-3mo + Absorbable/Material spendLaunching globally in 3 months without pilot
Overcommitted ComplexityTransformational spend + Very High complexity + Ecosystem shift + 0-12moBuilding proprietary AI with ecosystem shift in <1 year
Regulatory ConstraintAny deployment requiring Regulator/Union buy-in without pre-defined pathwayHealthcare AI without FDA engagement plan; Law AI without bar rule compliance assessment
Unhedged ScaleGlobal/National scale + No pilot phase + High/Very High complexityFull enterprise deployment of unproven tech without pilot
Existential BetExistential spend + 2+ years + Very High complexity + Ecosystem shiftBet-the-company M&A or full business model pivot

When a red-flag band combination appears: Challenge the participant. Offer a narrower scope, longer timeline, lower complexity, or lower spend. Unlock +/-3 exception scoring only if participant acknowledges and addresses the constraint.


Decision Specificity Checklist#

Before scoring ANY decision, verify participant specified:

ItemExample
WHOWhich team owner? Committed sponsor? (Not "we'll figure it out later")
WHATWhat specific capability/action? (e.g., "GitHub Copilot," not "deploy AI")
WHEREScope: pilot / function / geography / enterprise? (Not "across the company")
WHENTimeline realistic for scope? Regulatory approval needed? (Not "as soon as possible")
HOW MUCHHeadcount, capex, revenue impact clear? (Not "minimal budget")
HOWTalent plan? Integration detail? Rollback plan? (Not "we'll manage it")
RISKParticipant acknowledges execution and tail risk? (Not "this is foolproof")

Scoring rule: If >2 items are missing -> ask for specificity before scoring.


When to Challenge vs. When to Accept#

Challenge (Ask for Clarification)#

Ask for specificity if proposal is vague on WHO/WHAT/WHERE/WHEN/HOW/HOW MUCH/RISK.

Template: "I appreciate the direction. I need clarity on [specific issue]. Can you walk me through [detail]? That changes the execution risk profile significantly."

Accept (If Specificity Met)#

Once all 7 items are clear, score the decision per framework above.


Red-Flag Triggers (When to Unlock +/-3)#

Red-flag triggers are plausibility gates based on band combinations and strategic archetypes. If a red-flag fires, unlock +/-3 exception scoring (see Plausibility Decision Trees).

Quantitative Requirements (Only Required When...)#

Only quantify when the move is M&A, major capex build, or regulatory/legal commitment. Otherwise, bands + justification is sufficient.

  • M&A: Deal size, close timeline, integration plan required
  • Major Capex: Capex amount, ROI, payback period required
  • Regulatory/Legal Commitment: Regulatory timeline, legal exposure, severance obligations required
  • Otherwise: Band classification + concise justification suffices

Red-Flag Categories (Band-Based)#

CategoryTriggerExample
Timeline MisalignmentGlobal/National scale + 0-3mo timeline + Absorbable/Material spendLaunching globally in 3 months without pilot phase
Regulator-dependent deployment + <12mo timelineFDA or OCC approval assumed in under a year
Pilot to enterprise scaling + <6mo without proven track recordJumping from 50-store pilot to national rollout in 5 months
No Execution Risk DiscussionMissing talent acquisition plan (esp. Execution Complexity = High/Very High)"We'll hire AI engineers" with no sourcing plan or timeline
No integration roadmap (esp. Dependency = Vendor/Regulator/Ecosystem)Vendor partnership assumed but no contractual or technical plan
Unclear regulatory pathway (esp. regulated industry)Healthcare or Finance AI deployment with no compliance strategy
No data readiness assessment (esp. AI-heavy system)ML model proposed without addressing data quality or availability
Industry Constraint ViolationsHealthcare Provider AI + High/Very High complexity + no FDA engagement planDeploying clinical AI diagnostics without regulatory pathway
Finance/Trading AI + High complexity + no circuit breakers/risk controlsAI trading system with no kill switch or position limits
Retail/CPG + Headcount reduction >15% + no labor transition planMass layoffs without severance, retraining, or redeployment
Finance + AI underwriting + no bias auditing frameworkAutomated lending decisions without fair-lending compliance
Law + AI-generated work product + no bar rule compliance + no attorney reviewAI drafts filed with courts without human attorney review
Law + AI-generated briefs without human review + malpractice unaddressedClient-facing AI legal work with no liability framework
Consulting + AI advisory services + no client confidentiality safeguardsAI tools processing client data without data-handling protocols
Consulting + >40% junior headcount reduction + no talent pipeline planGutting the associate bench with no plan to develop future partners
Unhedged Tail RiskAutonomous systems + National/Global scale + no rollback planEnterprise-wide autonomous process with no manual override
Headcount reduction >30% + no severance/retention plan (M&A/major restructure only)Major restructuring with no workforce transition support
Full deployment + High/Very High complexity + no pilot phaseSkipping test-and-learn on a complex, unproven system
Novel technology in mission-critical domain (Healthcare, Finance, Law) + no human oversightAI making clinical, financial, or legal decisions without human review
Implausible Synergies (M&A only)Deal size >$5B + synergy targets >100% of acquisition costClaiming $6B in synergies on a $5B acquisition
Integration complexity High/Very High + synergy realization <12moExpecting full integration benefits within a year of close

Facilitation Language for Industry-Level Decisions#

When challenging a vague decision, reference the industry explicitly to keep context clear:

Challenging a Vague Decision (Industry-Specific)#

"I want to score this Retail decision, but I need specificity. Right now you've said 'deploy AI across stores and online.' But I need to know:

  • Where are you starting? Which 500 stores? All stores? Which e-commerce functions?
  • How much are you spending on the Retail initiative? $5M or $50M?
  • How will you staff it? Do you have retail tech talent, or do you need to hire?
  • What's the rollback plan if customer trust erodes?

Once you answer those, I can score your Retail decision. Note: If you also submit a Law decision, it will be scored separately on Law-specific metrics (billable hour impact, bar compliance, malpractice exposure, etc.)."

Challenging a Red-Flag#

"I appreciate the ambition here. But this proposal triggers a plausibility concern for me. Let me walk you through:

  • You're proposing FDA approval in 12 months for a novel diagnostic system.
  • FDA approval for novel diagnostics typically takes 18-24 months, even with a complete application.
  • In 12 months, you could submit an application or run a pilot with investigational status, but not full deployment.

Here's what I can score: Proposing a pilot with human oversight and FDA submission (not approval). That's Strategic Fit +1 (good direction), Execution Risk -1 (regulatory timeline is long), Tail Risk 0 (human oversight limits downside). Total: 0/6.

Or, if you want higher execution feasibility, propose a different approach. What would you prefer?"

Accepting a Strong Industry Decision#

"Okay, I have all the specificity I need for your Retail decision. Let me score:

  • Strategic Fit: +2 -- Direct response to inventory margin loss; aligns with omnichannel strategy.
  • Execution Risk: +1 -- Demand forecasting AI is proven; you have experience; 500-store pilot is feasible.
  • Tail Risk: +1 -- Phased pilot allows learning; if brand backlash emerges, you can narrow scope.
  • Total: +4/6 -- This is a strong Retail decision. Expected probability of success: 70%+. If you also submit a Consulting decision, it will be scored separately."

Scoring Reference Table#

Total ScoreInterpretationAction
+5 to +6Strong decision; expect 70%+ success probability.Accept and celebrate. Participant is executing well.
+1 to +4Acceptable decision; execution risk is real; monitor closely.Accept; add to watch list; expect 50-70% success probability.
-2 to 0Weak decision; reframe as narrower scope or defer.Challenge; ask participant to narrow scope or add hedges.
<-2Challenge; likely implausible; ask participant to narrow or add hedges.Reject or reframe significantly. Low probability of success.

Base Case Fallback Scoring#

After scoring all explicit industry decisions, apply base case fallbacks to any industry that did not receive an explicit action from any participant.

Fallback Scoring Process:

  1. Identify industries without explicit actions (variable count based on how many industries each participant covers)
  2. Reference the fallback bank (see Base Case Fallback Bank)
  3. Apply the pre-defined fallback score for that industry (deterministic, small fixed deltas: +/-1 per dimension, plausible)
  4. Fallback industries are applied automatically (no participant input required)
  5. Post fallback score alongside explicit decisions for transparency

Example: Participant covering Retail and CPG submits explicit Retail decision (+2/6). CPG receives no explicit action. Reference fallback bank -> CPG falls back to {0, +1, 0} = +1/6 (defensive but plausible cost-reduction move). Post both scores.

Fallback scoring is deterministic, not participant-controlled. Facilitator applies it directly from the fallback bank.


Industry Health Signals (End of Round)#

  • During play: Score explicit decisions only (3 dimensions, sum per decision).
  • Apply base case fallbacks after explicit scoring (reference fallback bank).
  • End of round: Update cumulative aggregate score for each industry (running total across all rounds). Look up Industry Health condition (Surge/Tailwind/Steady/Headwind/Crisis) from Industry Health Signal Tables.
  • Announce conditions at start of next round (~2 min). Apply Headwind/Crisis constraints as applicable.

Summary Checklist (Per Explicit Decision)#

  • Industry identified (Retail? CPG? Finance? Healthcare Provider? Consulting? Law? etc.)
  • Participant specified all 7 items (WHO/WHAT/WHERE/WHEN/HOW/HOW MUCH/RISK)
  • Band mapping clear (Spend/Commitment, Time-to-Impact, Complexity, Dependency, Scale)
  • No red-flag band combinations (if yes, unlock +/-3 scoring and reference decision trees)
  • Only quantify if M&A, major capex, or regulatory/legal commitment; otherwise bands + justification suffices
  • Score Strategic Fit: {-2, 0, +2} (or +/-3 if red-flag)
  • Score Execution Risk: {-2, 0, +2} (or +/-3 if red-flag) -- reference band translation table
  • Score Tail Risk: {-2, 0, +1} (or +/-3 if red-flag)
  • Sum total (range: -6 to +6)
  • Post score and brief rationale to participant
  • After round: Apply base case fallbacks to industries without explicit actions (reference Base Case Fallback Bank)
  • End of round: Update cumulative scores and look up Industry Health conditions (reference Industry Health Signal Tables)