Meta Scenario Planning Agent Skill

Scenario Planning

Framework

IRON LAW: Scenarios Are Not Predictions

Scenarios are PLAUSIBLE futures, not forecasts. The goal is NOT to predict
which future will happen, but to prepare strategies that work across
MULTIPLE possible futures. A strategy that only works in one scenario
is fragile.

The 2×2 Scenario Matrix Method

• Identify driving forces: What macro-forces will most shape the future? (technology, regulation, economy, demographics, competition)
• Select two critical uncertainties: The two most impactful forces with the most uncertain outcomes
• Build the 2×2 matrix: Each axis is one uncertainty with two endpoints (e.g., "regulation: strict vs lax")
• Name and describe four scenarios: Each quadrant is a distinct plausible future
• Test strategies against all four: Which strategies work in most/all scenarios? Which only work in one?

Process

Step 1: Driving Forces (brainstorm 10-15)

• Political, economic, social, technological, environmental, competitive
• Rate each on: Impact (H/M/L) × Uncertainty (H/M/L)
• High Impact + High Uncertainty → candidate for axes

Step 2: Select Two Axes

• Choose two forces that are both high-impact AND high-uncertainty
• They should be independent of each other (not correlated)

Step 3: Build Four Scenarios

• Give each scenario a memorable name (not "Scenario 1")
• Write a 1-paragraph narrative for each: what does this world look like in 5-10 years?

Step 4: Strategy Testing

• For each strategy option, assess: does it work in this scenario? (Yes / Partial / No)
• Robust strategies work in 3-4 scenarios. Fragile strategies work in only 1.

Output Format

# Scenario Planning: {Context}

## Driving Forces
| Force | Impact | Uncertainty | Selected? |
|-------|--------|------------|-----------|
| {force} | H/M/L | H/M/L | ✓/— |

## Scenario Matrix
- Axis 1: {Uncertainty A} — {endpoint 1} vs {endpoint 2}
- Axis 2: {Uncertainty B} — {endpoint 1} vs {endpoint 2}

| | {A: endpoint 1} | {A: endpoint 2} |
|---|---|---|
| **{B: endpoint 1}** | **"{Scenario Name}"**: {narrative} | **"{Scenario Name}"**: {narrative} |
| **{B: endpoint 2}** | **"{Scenario Name}"**: {narrative} | **"{Scenario Name}"**: {narrative} |

## Strategy Robustness Test
| Strategy | Scenario 1 | Scenario 2 | Scenario 3 | Scenario 4 |
|----------|-----------|-----------|-----------|-----------|
| {strategy A} | ✓/△/✗ | ✓/△/✗ | ✓/△/✗ | ✓/△/✗ |

## Robust Strategies
{Strategies that work in most scenarios}

## Contingency Triggers
- If {early signal}, activate {contingency plan for scenario X}

Gotchas

• Scenarios should be uncomfortable: If all four scenarios are comfortable, you haven't explored enough uncertainty. Include at least one scenario you'd rather not think about.
• Avoid "good/bad" framing: Scenarios aren't optimistic vs pessimistic. Each scenario has opportunities AND threats. A "strict regulation" world is bad for some and good for others.
• Early warning signals: Identify observable indicators that signal which scenario is unfolding. This converts scenarios into actionable intelligence.
• Two axes is a simplification: Reality has many uncertainties. The 2×2 is a tool for clarity, not completeness. Consider additional driving forces as variations within scenarios.

References

• For Shell's original scenario planning methodology, see references/shell-method.md