meta-scenario-planning
畫 2x2 情境矩陣，檢驗策略在不同未來下撐不撐得住
中文翻譯
---
name: "meta-scenario-planning"
description: "用驅動力、不確定性軸線和 2x2 情境矩陣做情境規劃，為多種可能的未來做準備。使用者面對高度不確定性、需要拿不同的未來來壓力測試某個策略，或要準備應變計畫時使用，就算他們說的是「萬一出問題怎麼辦」、「未來可能長什麼樣子」、「我們要怎麼為不確定性做準備」或「幫我們的策略做壓力測試」也一樣。"
metadata:
category: "WP-22 跨學科"
tags: ["meta-thinking", "scenario-planning", "strategy", "uncertainty"]
---
# 情境規劃
## 框架
```
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.
```
### 2×2 情境矩陣法
1. **找出驅動力**：哪些宏觀力量最會形塑未來？（技術、法規、經濟、人口結構、競爭）
2. **選出兩個關鍵不確定性**：影響最大、結果又最不確定的兩個力量
3. **建立 2×2 矩陣**：每條軸是一個不確定性，有兩個端點（例如「法規：嚴格 vs 寬鬆」）
4. **為四個情境命名並描述**：每個象限都是一個不同的合理未來
5. **用四個情境全部測試策略**：哪些策略在大部分／所有情境下都行得通？哪些只在一個情境下有用？
### 流程
**Step 1：驅動力（腦力激盪 10-15 個）**
- 政治、經濟、社會、技術、環境、競爭
- 每一個都評：影響（H/M/L）× 不確定性（H/M/L）
- 高影響 + 高不確定性 → 候選的軸線
**Step 2：選兩條軸**
- 選兩個既高影響、又高不確定性的力量
- 它們彼此要獨立（不相關）
**Step 3：建立四個情境**
- 給每個情境一個好記的名字（不要叫「情境 1」）
- 每個寫一段敘事：5-10 年後，這個世界是什麼樣子？
**Step 4：策略測試**
- 針對每個策略選項，評估：它在這個情境下行得通嗎？（Yes / Partial / No）
- 穩健的策略在 3-4 個情境下行得通。脆弱的策略只在 1 個情境下行得通。
## 輸出格式
```markdown
# 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}
```
## 注意事項
- **情境應該讓人不舒服**：如果四個情境都讓人覺得舒服，代表你探索的不確定性還不夠。至少要有一個你寧可不去想的情境。
- **避免「好／壞」的框架**：情境不是樂觀 vs 悲觀。每個情境都同時有機會和威脅。「嚴格法規」的世界對某些人不利，對另一些人有利。
- **早期預警訊號**：找出可觀察的指標，用來判斷哪個情境正在展開。這樣情境才會變成可行動的情報。
- **兩條軸是簡化**：現實有很多不確定性。2×2 是用來讓事情更清楚的工具，不是用來求完整。其他驅動力可以當成情境內部的變化來考慮。
## 參考資料
- Shell 原始的情境規劃方法論請看 `references/shell-method.md`英文原文
---
name: "meta-scenario-planning"
description: "Conduct scenario planning to prepare for multiple plausible futures using driving forces, uncertainty axes, and the 2x2 scenario matrix. Use this skill when the user faces high uncertainty, needs to stress-test a strategy against different futures, or prepare contingency plans — even if they say 'what if things go wrong', 'what could the future look like', 'how do we prepare for uncertainty', or 'stress-test our strategy'."
metadata:
category: "WP-22 跨學科"
tags: ["meta-thinking", "scenario-planning", "strategy", "uncertainty"]
---
# 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
1. **Identify driving forces**: What macro-forces will most shape the future? (technology, regulation, economy, demographics, competition)
2. **Select two critical uncertainties**: The two most impactful forces with the most uncertain outcomes
3. **Build the 2×2 matrix**: Each axis is one uncertainty with two endpoints (e.g., "regulation: strict vs lax")
4. **Name and describe four scenarios**: Each quadrant is a distinct plausible future
5. **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
```markdown
# 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`