scenario-war-room（情景作战室）

Scenario War Room

Tier: POWERFUL Category: C-Level Advisory Tags: scenario planning, war room, risk modeling, cascade effects, contingency planning, pre-mortem, crisis simulation

Overview

The Scenario War Room models cascading what-if scenarios across all business functions. Not single-assumption stress tests -- compound adversity that shows how one problem creates the next, and where the cascade can be interrupted. Every scenario produces concrete hedges with costs, owners, and deadlines.

When to Use

• A major risk has probability above 15% and impact above 20% of ARR
• Two or more threats could plausibly co-occur
• A strategic decision has significant downside if wrong
• Board or investors are asking "what's the worst case?"
• Pre-mortem before a major commitment (fundraise, acquisition, market entry)
• Quarterly risk review for leadership team

When NOT to Use

• Single-variable financial sensitivity analysis (use CFO Advisor stress testing)
• Routine project risk assessment (use project management risk frameworks)
• Technical failure mode analysis (use engineering incident planning)

The 6-Step Cascade Model

Step 1: Define Scenario Variables (Maximum 3)

More than 3 variables creates analysis paralysis, not insight. Choose the 3 that actually keep leadership awake at night.

For each variable, specify:

Variable Template:

Variable A: [Specific change]
  Probability: [X]%  |  Timeline: [When]
  Detection: [Early warning signal]
  First-order impact: [Immediate consequence]

Variable B: [Specific change]
  Probability: [X]%  |  Timeline: [When]
  Detection: [Early warning signal]
  First-order impact: [Immediate consequence]

Variable C: [Specific change]
  Probability: [X]%  |  Timeline: [When]
  Detection: [Early warning signal]
  First-order impact: [Immediate consequence]

Step 2: Domain Impact Mapping

For each variable, assess impact across every business function:

Step 3: Cascade Mapping (The Core)

This is the most valuable step. Map how Variable A triggers consequences that amplify Variable B.

Cascade Diagram:

TRIGGER: Customer churn ($560K ARR)
  │
  ├──▶ CFO: Runway drops 14 → 8 months
  │     │
  │     └──▶ CHRO: Hiring freeze imposed
  │           │
  │           └──▶ CTO: 3 open engineering reqs frozen, roadmap slips 2 months
  │                 │
  │                 └──▶ CPO: Q4 feature launch delayed → 2 more customers at risk
  │                       │
  │                       └──▶ CRO: NRR drops → additional churn risk (DEATH SPIRAL ENTRY)
  │
  └──▶ CRO: Revenue concentration increases (next largest = 22%)
        │
        └──▶ Investors: Concentration risk flagged → Series A terms worsen

Name the cascades explicitly. Common cascade patterns:

Step 4: Severity Matrix

Model three scenarios with increasing severity:

For each severity level, quantify:

BASE SCENARIO (Variable A only):
  Runway impact: [X] months → [Y] months
  ARR impact: -$[X] ([Y]% of total)
  Headcount impact: [freeze / reduction / none]
  Timeline to critical: [X] months
  Recovery plan: [specific actions]

STRESS SCENARIO (Variables A + B):
  Runway impact: [X] months → [Y] months
  ARR impact: -$[X] ([Y]% of total)
  Headcount impact: [specifics]
  Timeline to critical: [X] months
  Recovery plan: [specific actions]

SEVERE SCENARIO (All three):
  Runway impact: [X] months → [Y] months
  ARR impact: -$[X] ([Y]% of total)
  Headcount impact: [specifics]
  Timeline to critical: [X] months
  Existential: [yes/no]
  Emergency plan: [specific actions requiring board approval]

Step 5: Early Warning Signals (Trigger Points)

Define measurable signals that tell you a scenario is unfolding BEFORE it is confirmed. The value of this exercise is acting early, not reacting late.

Signal Design Criteria:

• Observable (you can actually measure it)
• Leading (appears before the full impact)
• Specific (not just "things feel off")
• Actionable (triggers a specific response)

Step 6: Hedging Strategies

For each scenario: actions to take NOW (before the scenario materializes) that reduce impact if it does. Hedges have costs -- the goal is cheap insurance, not paranoia.

Hedge Evaluation Criteria:

Hedge Table Template:

Output Format

Every war room session produces this structured output:

SCENARIO: [Name]
DATE: [Date of analysis]
PARTICIPANTS: [Who was involved]

VARIABLES:
  A: [Description] — Probability: [X]%, Timeline: [When]
  B: [Description] — Probability: [X]%, Timeline: [When]
  C: [Description] — Probability: [X]%, Timeline: [When]

MOST LIKELY PATH: [Which combination actually plays out, with reasoning]

SEVERITY LEVELS:
  Base (A only):  Runway [X]→[Y]mo, ARR impact -$[X]
    Recovery: [2-3 specific actions]
  Stress (A+B):   Runway [X]→[Y]mo, ARR impact -$[X]
    Recovery: [3-4 specific actions]
  Severe (A+B+C): Runway [X]→[Y]mo, ARR impact -$[X]
    Existential: [yes/no]
    Emergency: [actions requiring board approval]

CASCADE MAP:
  [A] → [domain impact] → [triggers B amplification] → [domain impact] → [end state]
  Interruption points: [where cascade can be broken]

EARLY WARNING SIGNALS:
  1. [Signal] → indicates [scenario], threshold: [specific]
  2. [Signal] → indicates [scenario], threshold: [specific]
  3. [Signal] → indicates [scenario], threshold: [specific]

HEDGES (implement now):
  1. [Action] — cost: $[X] — protects: [scenarios] — owner: [role] — deadline: [date]
  2. [Action] — cost: $[X] — protects: [scenarios] — owner: [role] — deadline: [date]
  3. [Action] — cost: $[X] — protects: [scenarios] — owner: [role] — deadline: [date]

RECOMMENDED DECISION:
  [One paragraph: what to do, in what order, and why]

REVIEW DATE: [When to re-run this analysis — typically 90 days or after any variable shifts]

Common Scenarios by Company Stage

Seed Stage

• Co-founder departure + product misses launch deadline
• Runway runs out + bridge terms are predatory
• Key technical hire falls through + competitor ships first

Series A

• Miss ARR target + fundraise delayed
• Top customer churns + competitor raises large round
• Key engineer leaves + critical feature deadline

Series B+

• Market contraction + burn multiple spikes above 3x
• Lead investor wants strategic pivot + team resists
• Regulatory change + product requires rearchitecture

War Room Ground Rules

1. Maximum 3 variables per scenario. More is noise. Model the ones that actually matter.
2. Quantify or estimate. "Revenue drops" is not useful. "$420K ARR at risk over 60 days" is. Use ranges if uncertain.
3. Don't stop at first-order effects. The real damage is always in the cascade.
4. Model recovery, not just impact. Every scenario must have a "what we do" path.
5. Separate base case from sensitivity. Don't conflate "what probably happens" with "what could happen."
6. 3-4 scenarios per planning cycle. More creates analysis paralysis.
7. Review every 90 days. Probabilities and variables change. Stale scenarios give false comfort.
8. No judgment-free zone. People must feel safe naming ugly scenarios.

Related Skills

Troubleshooting

Success Criteria

• Each scenario session produces exactly 3 variables, 3 severity levels, and a cascade map with interruption points identified
• Early warning signals are specific enough to be monitored: observable, leading, and actionable with defined thresholds
• Hedges are costed, owned, and have deadlines within 7 days of the war room session
• At least one hedge per scenario is implemented (not just planned) within 30 days
• Scenario review conducted every 90 days with probability updates based on new information
• When an early warning signal fires, the pre-planned response is executed within the defined timeline
• War room output is concise enough for board consumption: one-page summary per scenario

Scope & Limitations

• In scope: Multi-variable scenario construction, cascade modeling across all business functions, severity matrix analysis, early warning signal design, hedge strategy with cost-benefit analysis, scenario review cadence
• Out of scope: Single-variable financial sensitivity analysis (use CFO Advisor stress testing); technical failure mode analysis (use engineering incident planning); routine project risk assessment (use project management frameworks); insurance and risk transfer (use specialized broker)
• Limitation: Scenario probabilities are subjective estimates, not actuarial calculations; value is in preparedness, not prediction accuracy
• Limitation: Framework assumes scenarios are independent or correlated; black swan events by definition are not modelable
• Limitation: Cascade mapping is based on common organizational patterns; unique company structures may have different cascade paths
• Limitation: Maximum 3 variables per scenario is a deliberate constraint; more variables create analysis paralysis, not better insight

Integration Points

Python Tools