Management & Thinking
ProCynefin Framework
Cynefin 複雜度框架 · Source: Dave Snowden
Diagnosing the problem type before deciding — clear, complicated, complex, and chaotic situations demand fundamentally different responses
Core Concept
Snowden's decision framework from IBM: classify situations into five types — Clear (known answer, just execute), Complicated (needs expert analysis), Complex (cause-effect only visible in retrospect, requires experimentation), Chaotic (stabilize first), Disorder (uncategorized). Treating Complex as Complicated is the classic organizational disaster — you assume more experts will solve it, but complex systems reject predictive management.
✓ When to use this
When facing an unfamiliar problem and unsure which method applies. Before diving in, pause to classify — Clear / Complicated / Complex / Chaotic — because misclassification means wrong method.
✗ When not to use this
Don't use it on already-clear execution tasks — classification becomes overhead. Skip for routine operational decisions too; classification itself has a time cost.
Questions you will be asked
Using this framework, you will work through —
- 1.Describe the situation or problem you're facing.
- 2.Can the cause-effect relationships be seen in advance?
- 3.Do best practices or SOPs already exist?
- …and 3 more
Worked example
Expand to see what a filled-in run looks like
›
Worked example
Expand to see what a filled-in run looks like
Situation
公司營收連 3 個月衰退 15%,老闆要求 2 週內提出復興計畫。團隊吵了 5 天還沒共識——有人主張砍成本、有人主張開新市場、有人主張加 PR 預算。
1. Describe the situation or problem you're facing.
公司 3 個月衰退 15%,要決定復興方向。
2. Can the cause-effect relationships be seen in advance?
事前能不能看清因果?衰退原因不止一個——可能是市場、競品、產品、團隊綜合。連症狀和原因都還搞不清。
3. Do best practices or SOPs already exist?
有沒有最佳實踐?衰退時的標準動作(砍成本、聚焦核心)是繁雜層級的解;但這次衰退原因不明,標準動作可能對症也可能不對。
4. Based on the above, which category does this fit?
判斷:這是複雜情境(不是繁雜)——因果只能透過實驗看清,不能直接套用過往案例。
5. What's the right response for this category? Does your current approach match?
對應方式:複雜要 probe → sense → respond——做小規模實驗(A:聚焦既有客戶 vs B:開新通路),2 週內看數據再決定。不是一次決定一個方向。
6. Which misclassification are you most prone to?
我們最容易誤判的是「把複雜當繁雜」——找專家給答案。複雜系統沒人知道答案,要用實驗去找。
Use it inside ChatGPT / Claude
Paste the prompt below and the AI will walk you through this framework, one question at a time.
你現在是引導使用者用 Cynefin 框架做情境分類的教練(Dave Snowden)。 依序問: 1) 你正面對的情境或問題是什麼? 2) 因果關係事前能看清嗎?(能 → Clear/Complicated;只能事後看清 → Complex;完全混亂 → Chaotic) 3) 已有最佳實踐或標準作業可循嗎?(有 → Clear;沒有但專家能分析 → Complicated;連專家都沒答案 → Complex) 4) 基於前兩題,這是什麼類別?(明確 / 繁雜 / 複雜 / 混沌) 5) 對應的處理方式是什麼?(明確:直接做|繁雜:找專家|複雜:小實驗 + 學習 + 放大|混沌:先穩定) 6) 你目前的做法是哪一種?符合該類別嗎? 特別注意:當使用者把複雜情境當繁雜情境(找專家給答案),明確指出來——複雜系統沒有事前答案,只能透過實驗找路。 互動規則: 1. 一次只問一題,等使用者回答後再進入下一題。 2. 使用者答完所有題目前,不要做總結或下結論。 3. 若答案太抽象、太籠統,請追問一次具體例子或數字後再繼續。 4. 全部答完後,輸出三段:(a) 摘要使用者的關鍵判斷;(b) 你看到的盲點或張力;(c) 一個具體下一步行動建議。 5. 不要替使用者做決定,只把判斷攤開讓他自己決定。
Related Frameworks
Investment & Finance
Scenario Planning
Systematically evaluating worst, base, and best outcomes under high uncertainty
Management & Thinking
First Principles Thinking
Breaking "we've always done it this way" inertia — rebuilding solutions from the ground up
Management & Thinking
OODA Loop
Making fast, effective decisions in rapidly changing or highly uncertain situations
Related studies