Distinctions
IS
- A sense-making and decision-making framework that helps leaders choose an approach based on the kind of situation they are facing, rather than assuming one best practice fits everything.
- A method to differentiate problems that can be solved with analysis and problems that require experimentation and learning.
- Answers: What kind of problem are you dealing with?
IS Not
Systems
Zooming In: Parts
Contexts
| Context | Characteristics | Examples | Response |
|---|---|---|---|
| Clear (AKA Simple) | - Obvious cause → effect - Repeatable processes - Known best practices | - Payroll processing - Following a recipe - Filling out a tax form | Sense → Categorize → Respond |
| Complicated | - Cause and effect exist but require expertise to see - You need experts or analysis. - There may be multiple correct solutions, not just one best practice. | - Engineering problems - Diagnosing a car engine - Architecture decisions | Sense → Analyze → Respond |
| Complex | - Cause and effect only visible after the fact - Patterns emerge over time - You cannot predict outcomes | - Organizational culture - Innovation - Markets - Ecosystems - Product adoption | Probe → Sense → Respond |
| Chaotic | - No perceivable cause-effect relationship - System is unstable | - Crisis response - Production outage - Natural disaster | Act → Sense → Respond - Stabilize first, then learn |
| Disorder/Confusion | - You don’t yet know which domain you’re in. | - Most organizations | Sense-making (identify one of the other 4 contexts) |
Responses
| Response | Meaning in Cynefin |
|---|---|
| Sense | Observe what is happening in the system and gather signals from reality. |
| Categorize | Match the situation to a known pattern or existing class of problems. |
| Analyze | Examine the situation using expertise, models, or investigation to understand cause and effect. |
| Probe | Run a small safe-to-fail experiment to see what patterns emerge. |
| Act | Take decisive action to stabilize the system or interrupt chaos. |
| Respond | Adjust actions based on what has been learned from sensing. |
| Sense-Making | Interpreting signals collectively to understand what context you are in. |
Failure Modes
| Situation | Mistake | What actually happens |
|---|---|---|
| Complex problem | demanding predictability | Analysis produces false certainty, causing rigid plans that collapse when reality shifts. |
| Complicated problem | running experiments instead of analysis | Trial-and-error wastes time and may introduce risk when expertise could identify the correct solution. |
| Simple problem | overengineering | Extra analysis adds unnecessary complexity and slows down routine work. |
| Chaotic problem | holding meetings | Delay allows instability to spread before the system is stabilized. |
Relationships
| Relationship | Concept | Rationale |
|---|---|---|
| Prevents | Best Practices Overreach | Best practices apply only in Clear contexts; recognizing the context prevents applying them where judgment or experimentation is required. |
| Prevents | treating routine work like a crisis (“Sky is Falling”) | Clear contexts should follow established practices instead of urgent intervention |
| Explains | Deterministic Planning as a failure mode | Occurs when complex problems are treated as complicated ones requiring prediction |
| The Cliff |