Cynefin Framework: The 5 Domains -- Guide and Practical Example

Most companies treat their innovation projects as if they were all the same. A new app is managed with the same methodology as a fundamental business model transformation. A process improvement project goes through the same stage-gate process as an exploratory prototype in an entirely new market. The result: methods that fit one problem perfectly fail with another — not because the method is bad, but because it’s applied in the wrong context.

The Cynefin Framework solves precisely this problem. It is not another project management tool and not an innovation process. It is a sense-making framework — a tool that helps you recognize the nature of your problem before you decide which method to use. Its value lies not in the answer it provides but in the question it forces: What kind of situation am I in?

Where Does Cynefin Come From?

Dave Snowden and the Origin at IBM

Dave Snowden developed the Cynefin Framework in the late 1990s as part of his work at the IBM Cynefin Centre for Organisational Complexity. The name “Cynefin” (pronounced “kuh-NEV-in”) comes from Welsh and roughly means “habitat” or “the place where we belong” — but with the connotation that we cannot fully describe this place because it has shaped us in ways we are not conscious of.¹

The theoretical roots are interdisciplinary: complexity theory (Stuart Kauffman, Santa Fe Institute), social systems theory, cognitive science, and knowledge management. Snowden combined these perspectives into a framework pragmatic enough for managers and theoretically sound enough for researchers.

In 2007, Snowden and Mary Boone published “A Leader’s Framework for Decision Making” in the Harvard Business Review — the article that brought Cynefin into the management mainstream.² The central thesis: most leaders respond to every situation with the same tools because they don’t diagnose the context. Cynefin provides a diagnostic that precedes the choice of method.

Ontological Foundation: Why Cynefin Is Not a Categorization Tool

A common misconception: Cynefin is treated as a 2x2 matrix into which you sort problems. Snowden explicitly resists this. Cynefin is a sense-making framework, not a categorization scheme.³ The difference is fundamental:

• Categorization: You define the categories first and then assign phenomena to them. The categories are fixed.
• Sense-making: You first observe the properties of the situation and let the assignment emerge. The domains are contexts with different action logics.

In practice, this means: you don’t diagnose once that “we are in the Complicated domain” and then work from that assessment for three months. You continuously check whether the nature of the situation has changed — and adjust your approach accordingly.

The Five Domains in Detail

1. Clear (formerly Obvious/Simple)

Decision logic: Sense — Categorize — Respond

In the Clear domain, the relationship between cause and effect is obvious. Known best practices exist that reliably work. The correct approach is clear, and anyone with basic knowledge can recognize it.

Characteristics:

• Repeatable processes with clear rules
• Best practices exist and are documented
• Cause-and-effect relationships are apparent to everyone
• The right answer is unambiguous

Examples: Invoicing according to a fixed template, onboarding checklist for new employees, standard service requests with documented solutions.

Recommended action: Standardize, automate, delegate. Create process manuals, train employees on best practices, implement checklists.

The danger in the Clear domain: Complacency. When a process functions in the Clear domain for a long time, the illusion forms that it will always work this way. Snowden describes this as the “complacent zone”: teams stop questioning their assumptions and are caught off guard by changes that shift the context.² An example: an insurer has had a standardized claims process for 20 years. It works — until a new competitor with digital claims processing shifts customer expectations. Suddenly, the previously clear process is no longer “best practice” but “the way we’ve always done it.”

2. Complicated

Decision logic: Sense — Analyze — Respond

In the Complicated domain, a clear cause-and-effect relationship exists, but it isn’t obvious. You need expertise to recognize it. Multiple correct solutions exist (good practices, not best practices), and the choice between them requires specialist knowledge.

Characteristics:

• Cause-and-effect relationships exist but aren’t obvious
• Expert knowledge is required
• Analysis and diagnosis lead to the right solution
• Multiple valid approaches (good practices)

Examples: Architecture decision for an IT platform, process optimization with Lean methods, market analysis for a known market with established competitors.

Recommended action: Bring in experts, analyze thoroughly, evaluate options. The temptation: simplify problems to solve them without expertise. The consequence: suboptimal solutions that work short-term and produce costs long-term.

Typical mistake in the DACH context: DACH companies with strong engineering cultures tend to treat everything as complicated. Every problem is analyzed, every solution calculated, every decision validated by experts. This works for problems that are genuinely complicated. For complex problems — where cause-and-effect relationships are unpredictable — this approach produces detailed plans for a world that doesn’t follow the plan.

3. Complex

Decision logic: Probe — Sense — Respond

In the Complex domain, cause-and-effect relationships are only recognizable in hindsight — they cannot be predicted. The system consists of many interacting elements whose interplay produces emergent patterns. What worked yesterday may not work tomorrow.

Characteristics:

• Cause-and-effect relationships are only recognizable retrospectively
• The system is emergent — patterns arise through interaction
• Prediction is impossible; understanding only possible in hindsight
• Emergent practices rather than good or best practices
• Small changes can have large effects (and vice versa)

Examples: Introducing a new business model in an established company, cultural change in an organization, innovation in a market with unclear demand, service innovation in a new customer segment.

Recommended action: Experiment. Deploy “safe-to-fail probes” — small experiments that quickly yield insights. Amplify what works. Dampen what doesn’t. Accept that you cannot know the path in advance.

Why most innovation lives here: Design thinking is a method for the Complex domain. Prototyping, iterative testing, “fail fast” — these principles only make sense in the Complex domain. In the Complicated domain, “fail fast” would be wasteful — you could find the right solution through analysis. In the Complex domain, failure is the fastest path to learning because the right solution only emerges through interaction with the system.

Safe-to-fail probes: The central instrument in the Complex domain. A safe-to-fail probe is a small experiment that:

• Is fast and cheap
• Causes no systemic damage if it fails
• Delivers clear signals about whether to amplify or dampen
• Runs in parallel with other probes (because you can’t predict which one will work)

A DACH insurer testing a new digital claims service could launch three parallel safe-to-fail probes: (1) A chatbot for standard claims. (2) A video call service for complex claims. (3) A self-service portal with automatic damage recognition via photo. After six weeks, the data shows which approach resonates with customers — and that one is amplified. The others are not judged as “failures” but as information gained.

4. Chaotic

Decision logic: Act — Sense — Respond

In the Chaotic domain, there is no recognizable relationship between cause and effect. The system is unstable. The priority: establish stability, not find the optimal solution.

Characteristics:

• No recognizable order
• Immediate action required
• No room for analysis or experiments
• Goal: move from Chaos into the Complex or Complicated domain
• Novel practices — new approaches never tested before

Examples: Cyberattack on critical infrastructure, massive service outage with unknown cause, sudden market collapse (COVID-19 for travel companies), crisis communication after a scandal.

Recommended action: Act immediately. Communicate clearly. Create stability. Analyze later. In the Chaotic domain, action is more important than perfection — any decision is better than no decision. Only once stability is established do you move the situation into the Complex domain and begin probing.

Deliberate entry into Chaos: Snowden describes a counterintuitive use case: some organizations deliberately create controlled chaos to break entrenched structures.³ An example: a corporation dissolves a department and completely reconstitutes the teams to break ingrained thinking patterns. This is risky, but sometimes the only way to escape a system that has ossified in the Clear domain.

5. Disorder

Decision logic: None — that’s the problem.

Disorder is the central area of the Cynefin Framework. It describes the state where you don’t know which domain you’re in. The natural reaction: you fall back on the action logic most familiar to you — and that is almost always wrong for the current context.

Characteristics:

• Uncertainty about the nature of the situation
• Reversion to preferred action patterns
• Highest risk of wrong decisions

Why Disorder is dangerous: A manager with an engineering background will instinctively treat a problem in Disorder as “complicated” and launch an analysis. A manager with startup experience will treat it as “complex” and start experiments. Both could be right — but both could also be fundamentally wrong if the situation is actually chaotic and requires immediate action.

Recommended action: Break the situation into smaller parts and assign each part to a domain. A large innovation project likely has elements in all four domains: the technical platform is Complicated. Customer behavior is Complex. The budget process is Clear. And the regulatory situation might be Chaotic.

Cynefin and Innovation: Why the Framework Is Indispensable for Innovation Management

The Fundamental Problem: One Methodology for Different Contexts

Most innovation management frameworks — stage-gate, Lean Startup, design thinking — are context-specific but deployed context-universally. Stage-gate works excellently in the Complicated domain (clearly defined problem, expert knowledge available, predictable cause-and-effect chains). In the Complex domain, stage-gate becomes an obstacle because its linear gates hinder iteration.

Cynefin doesn’t solve this problem by proposing a new method but by providing an upstream diagnostic: What kind of problem do you have — and which method fits it?

Domain	Innovation Type	Suitable Method	Wrong Method
Clear	Incremental improvement	Standard process optimization, Lean	Design thinking (overkill)
Complicated	Sustaining innovation	Stage-gate, Lean Six Sigma, expert reviews	Agile exploration (unnecessary)
Complex	Disruptive innovation, new business models	Design thinking, Lean Startup, safe-to-fail probes	Stage-gate (too linear), detailed business cases (not validatable)
Chaotic	Crisis management, radical pivots	Immediate action, clear communication, taskforce	Participatory processes, long-term planning

Safe-to-Fail in the Innovation Portfolio

An innovation portfolio in a DACH company typically contains projects in different domains. Cynefin helps assign the right management logic to each project:

Practical example: Innovation portfolio of an automotive supplier

A DACH automotive supplier has five innovation projects in its portfolio:

1. Increase production efficiency -> Clear. Apply best practice, standardize.
2. Develop new sensor platform -> Complicated. Bring together expert knowledge, analyze, choose good practice.
3. Test mobility-as-a-service business model -> Complex. Launch safe-to-fail probes, experiment with customers.
4. Respond to semiconductor shortage -> Chaotic (as it was in 2021). Act immediately, rebuild supply chains.
5. Evaluate entry into hydrogen technology -> Disorder. Still unclear whether this is a complicated (technology) or complex (market) problem. Initial diagnostics needed.

Without Cynefin, the supplier would send all five projects through the same stage-gate process. With Cynefin: projects 1-2 go through stage-gate. Project 3 is managed with parallel experiments. Project 4 gets a taskforce. Project 5 starts with a domain diagnostic.

Cynefin vs. Stacey Matrix: What’s the Difference?

The Stacey Matrix (Ralph Stacey, 1996) is often cited as an alternative or complement to Cynefin. Both address complexity but with different approaches:⁴

Dimension	Cynefin	Stacey Matrix
Axes	No explicit axes (contexts, not dimensions)	Two axes: Requirements clarity x Technology certainty
Domains	5 (Clear, Complicated, Complex, Chaotic, Disorder)	4 zones (Simple, Complicated, Complex, Chaos) + “Zone of Agreement/Certainty”
Dynamics	Explicit: domains shift, boundaries are permeable	More static: position on the matrix is determined once
Theoretical basis	Complexity theory, sense-making, cognitive science	Systems theory, contingency theory
Strength	Action logic per domain, dynamic shifts	Intuitive visualization, quick classification
Weakness	Requires practice in diagnostics	Suggests static classification, less differentiated

Recommendation: Use the Stacey Matrix for quick initial diagnostics in workshops (it’s more intuitive). Use Cynefin for ongoing management because it better captures dynamics between domains and provides concrete action logics.

Common Cynefin Mistakes

Mistake 1: Confusing Complex with Complicated

The most common and most expensive mistake. A complicated problem has a correct solution that you can find through analysis. A complex problem has no predictable solution — it emerges through interaction with the system. If you treat a complex problem as complicated, you create a detailed plan that fails against reality. If you treat a complicated problem as complex, you experiment unnecessarily instead of using available expertise.

Test: Can an expert tell you the solution if you give them enough information? If yes: complicated. If no expert can predict the solution because it depends on interactions that haven’t happened yet: complex.

Mistake 2: Using Cynefin as a Static Classification

Situations move between domains. A project that was complex last week can tip into the Chaotic domain this week (e.g., due to a sudden regulatory change). A problem that started as complicated can turn out to be complex when experts disagree and analysis yields no clear solution.

Mistake 3: Classifying Only the Overall Problem

Most real problems have aspects in different domains. A business transformation has clear elements (regulatory requirements), complicated elements (technology selection), complex elements (cultural change), and possibly chaotic elements (market disruption). Applying Cynefin to the overall problem is too coarse. Apply it to the sub-problems.

Mistake 4: Ignoring the Disorder Domain

In practice, leadership teams find themselves in Disorder more often than they admit. The natural reaction — falling back on familiar action logic — is the most dangerous one. A leadership team that admits “We don’t know what kind of situation we’re in” is better positioned than one that acts on instinct.

Mistake 5: Using Cynefin as an Excuse for Passivity

“That’s complex, so we can’t plan” is not an action strategy. Cynefin doesn’t say complex situations are unmanageable. It says they must be managed differently: through parallel experiments, not through detailed plans. Safe-to-fail probes are active interventions, not passive observation.

Cynefin in Practice: Workshop Format for Innovation Teams

Domain Mapping in 90 Minutes

Goal: Diagnose the innovation portfolio by Cynefin domains and adjust the management logic per project.

Step 1: Problem Inventory (20 min) Each team member writes their three most important current challenges or projects on cards.

Step 2: Domain Diagnostics (30 min) For each card: discussion along diagnostic questions:

• Is there a known, proven solution? -> Clear
• Are there experts who can analyze the solution? -> Complicated
• Does the solution depend on interactions we cannot predict? -> Complex
• Is the situation unstable and requiring immediate action? -> Chaotic
• Are we uncertain which domain we’re in? -> Disorder

Step 3: Method Assignment (20 min) For each card: Which method fits the diagnosed domain? Where are we currently using a method that doesn’t fit the context?

Step 4: Action Planning (20 min) Identify the three most important adjustments and define concrete next steps.

Frequently Asked Questions

What is the Cynefin Framework in simple terms?

The Cynefin Framework is a decision framework that helps you recognize what kind of situation you’re in — and which approach fits. It distinguishes five domains: Clear (known solution), Complicated (expert solution), Complex (experimental solution), Chaotic (immediate action), and Disorder (uncertainty about the situation). The core: not every problem needs the same method.

How do you pronounce Cynefin?

“Kuh-NEV-in.” The word comes from Welsh and means “habitat” or “place of belonging” — with the connotation that your own context shapes you in ways you’re not conscious of.

What is the difference between complicated and complex?

Complicated: there is a correct solution that an expert can find through analysis (e.g., building a bridge). Complex: the solution is not predictable and only emerges through interaction with the system (e.g., initiating cultural change). You solve complicated problems through expertise. You solve complex problems through experiments.

How does Cynefin help with innovation?

Most innovation happens in the Complex domain — where the solution is not predictable. Cynefin helps you distinguish between projects that need analysis (Complicated) and those that need experiments (Complex). This prevents the most common mistake: managing a complex innovation project with the methodology of a complicated project.

What are safe-to-fail probes?

Small, fast, cheap experiments that run in parallel and yield insights about a complex system. If an experiment works, it is amplified. If it doesn’t, it is dampened — without systemic damage. Safe-to-fail probes are the central action instrument in the Complex domain.

When is a problem chaotic rather than complex?

In the Complex domain, you have time for experiments — the situation is dynamic but not unstable. In the Chaotic domain, you have no time: the situation requires immediate action because the system is unstable. Typical chaotic situations: cyberattack, sudden market collapse, acute crisis.

Methodology & Sources

This article draws on 10 academic and practitioner sources, including the foundational works by Snowden (1999, 2007, 2020), complexity theory (Kauffman, Santa Fe Institute), the Stacey Matrix (1996), and practical applications in innovation management.

SERP finding: The German-language top-10 results for “Cynefin Framework” are superficial introductions (definition + 4/5 domain descriptions). None systematically connects Cynefin to innovation management, explains safe-to-fail probes with a concrete DACH example, offers a structured comparison with the Stacey Matrix, or warns against the most common mistake (confusing Complex with Complicated). This article closes these four gaps.

Limitations: Cynefin is primarily a qualitative sense-making framework. Empirical studies on its effectiveness in corporate contexts are limited. Domain assignment is subjective and requires practice — different teams may assign the same situation to different domains.

Disclosure: SI Labs supports companies in developing service innovation capabilities. Cynefin is one of the diagnostic frameworks we use to determine the right method mix for innovation projects — not a standalone consulting product.

References

Footnotes

1. Snowden, Dave. “Complex Acts of Knowing: Paradox and Descriptive Self-Awareness.” Journal of Knowledge Management 6, no. 2 (2002): 100—111. First academic description of the Cynefin Framework with derivation of the name. ↩

2. Snowden, David J. and Mary E. Boone. “A Leader’s Framework for Decision Making.” Harvard Business Review 85, no. 11 (November 2007): 68—76. The article that established Cynefin in the management mainstream. Describes the four original domains (Simple, Complicated, Complex, Chaotic) plus Disorder. ↩ ↩²

3. Snowden, Dave, Zhen Goh, and Sue Borchardt. Cynefin — Weaving Sense-Making into the Fabric of Our World. Cognitive Edge / The Cynefin Company, 2020. Framework update: renaming “Simple” to “Clear,” expanded depiction of domain dynamics. ↩ ↩²

4. Stacey, Ralph D. Strategic Management and Organisational Dynamics: The Challenge of Complexity. Pitman Publishing, 1996. The Stacey Matrix as predecessor and parallel to Cynefin. ↩