Cynefin Framework

Theory & origin

Dave Snowden developed Cynefin, Welsh for habitat, at IBM around 1999 as a sense-making framework, not a categorization scheme, meaning its point is to help you decide how to act, not to file problems into boxes. It names five domains. Clear, once called Obvious or Simple, is the domain of the known, where cause and effect are obvious to everyone and the response is Sense, Categorize, Respond, applying best practice. Complicated is the domain of the knowable, where cause and effect exist but require expertise to see, so the response is Sense, Analyze, Respond, and there is good practice rather than one best practice. Complex is the domain of the emergent, where cause and effect are only clear in hindsight, so you cannot analyze your way forward and must instead Probe, Sense, Respond, running safe-to-fail experiments and amplifying what works. Chaotic is the domain of the novel, where there is no discernible cause and effect and the only move is to Act, Sense, Respond, doing something to establish order before you can think. Disorder, the center, is not knowing which domain you are in, and it is the most dangerous, because people default to whichever domain they are most comfortable with rather than the one they are actually in. The framework's teeth are in that last point. Most management failure is a domain error: treating a complex problem, a culture change, a market shift, as if it were merely complicated, solvable by enough analysis and a detailed plan, when it actually needs experiments. And the cliff between Clear and Chaotic is real: complacency in the obvious domain leads to sudden collapse into chaos.

Key components

The parts at a glance. Click any term for the full definition, a field example, and the common failure, in the model below.

How a consultant runs it

1. 01 Diagnose the domain before choosing a method. Most failures are domain errors, using best-practice playbooks on complex problems that need experiments.
2. 02 In the Complex domain, stop planning and start probing. Run small safe-to-fail experiments, amplify what works, dampen what does not, because the answer only emerges from action.
3. 03 Do not confuse complicated with complex. A complicated problem yields to expertise and analysis, a complex one does not, and treating the second like the first is the classic expensive mistake.
4. 04 In Chaos, act first to establish order, then move the situation into complex or complicated. Deliberation in a genuine crisis is its own failure.
5. 05 Beware the cliff and the center. Complacency in the Clear domain drops you into Chaos, and Disorder, not knowing your domain, makes everyone default to their comfort zone.

Worked example

A multifinance lender wants to become the fast one in SME lending and runs the change as a big detailed program, a complicated-domain response, and it stalls. The Cynefin read shows the domain error. Parts of the work truly are Clear: standard applications against fixed rules just need best practice and clean execution. Other parts are Complicated: designing the new credit-scoring logic is hard but knowable, and the right move is expert analysis. But the heart of the ambition, shifting a deeply risk-averse culture toward speed, is Complex: cause and effect only show up in hindsight, and no amount of planning will predict what actually changes behavior. Treated as complicated, it was being run with a fixed plan and a gantt chart, which is why it kept missing. Reframed as complex, the team switches to Probe-Sense-Respond: several small safe-to-fail experiments, a fast-track lane in one branch, a speed metric on one team, a revised approval limit for one segment, then amplify whatever moves the numbers and drop the rest. Meanwhile a fraud scare that erupts mid-program is correctly handled as Chaotic, act first to contain it, analyze later. The program starts working once each part is matched to its real domain instead of all of it being forced through one.

Common pitfalls

1. 01 Treating a complex problem as merely complicated, betting on one analyzed plan reality ignores
2. 02 Living in Disorder, defaulting to a comfortable method instead of the situation's real domain
3. 03 Complacency in the Clear domain, which drops you off the cliff into chaos
4. 04 Deliberating for consensus in a genuine crisis, when only action helps
5. 05 Using best practice everywhere, when it only fits the Clear domain

Sample deliverable

One real engagement, start to finish. Watch the numbers travel from raw input, onto the chart, into the finished artifact.