Cynefin Sense-making

What The Framework Offers

Cynefin helps you pause before choosing a decision method. It asks a simple but powerful question: what kind of situation are we actually in?

Some situations reward a known process. Some need expertise and analysis. Some only reveal themselves through careful experiments. Some are unstable enough that the first job is to act and create safety. Cynefin gives names to those differences so teams stop using the same meeting, checklist or spreadsheet for every problem.

Why Cynefin Matters

Bad decisions often begin before the decision is discussed. The team quietly assumes the situation is clear, complicated, complex or chaotic, then chooses a method that fits that assumption.

If the assumption is wrong, the method can make things worse. A checklist can flatten a messy customer behaviour problem. An expert can over-diagnose a market that is still emerging. A workshop can waste precious time during an operational crisis. A founder can call something chaotic because it feels stressful, when it is actually complicated and needs calm diagnosis.

The framework is not there to make the business sound clever. It is there to stop avoidable method mistakes.

The Five Domains In Plain English

Clear

Cause and effect are obvious. The right response is known and repeatable.

How it feels: Routine, familiar and slightly boring. People know what good looks like.
Decision process: Sense what is happening, categorise it, then respond using the known process.
Common mistake: Overcomplicating something that should simply be done properly.
Example: A customer requests a refund inside the published policy. Follow the policy, record it and learn later if patterns appear.

Complicated

Cause and effect exist, but they are not obvious. Expertise and analysis can usually find a good answer.

How it feels: Detailed, technical or multi-option. The team may need someone who has seen this before.
Decision process: Sense what is happening, analyse options, compare evidence, then respond.
Common mistake: Letting experts optimise one part while missing the business trade-off around it.
Example: Choosing between payment providers, fulfilment options or accounting software when the criteria are known.

Complex

Cause and effect can only be understood after something happens. The pattern is emerging, not waiting to be calculated.

How it feels: Uncertain, social and alive. Customers, staff, suppliers or markets keep reacting back.
Decision process: Probe safely, sense what changes, then respond by amplifying what works and dampening what does not.
Common mistake: Demanding a full plan before the system has produced enough evidence.
Example: Testing whether a new audience understands the offer, trusts the price and behaves like real buyers.

Chaotic

The situation is unstable. There is no useful cause-and-effect conversation until something has been stabilised.

How it feels: Urgent, noisy and risky. Customers, cash, safety, reputation or delivery may be actively exposed.
Decision process: Act first to contain damage, sense what has changed, then respond and reassess the domain.
Common mistake: Holding a long analysis session while the business is still leaking trust, money or operational control.
Example: A major outage, unsafe product issue, missed event delivery or public customer complaint that needs immediate containment.

Disordered Or Confused

The team has not agreed what kind of situation it is in. Different people are applying different rules without saying so.

How it feels: Meetings loop. One person wants a checklist, another wants an expert, another wants experiments and another wants urgent action.
Decision process: Split the situation into parts, classify each part and stop pretending one method fits the whole mess.
Common mistake: Trying to win the argument about the solution before agreeing what kind of problem is being solved.
Example: A growth problem that includes clear invoicing errors, complicated supplier options, complex customer behaviour and chaotic cash pressure.

Practical Comparison Table

Domain	What you can know	Best response	Business example	Failure mode
Clear	The right answer is known or easy to categorise.	Use the known process.	Process a standard refund, issue an invoice or follow a compliance checklist.	Turning routine work into debate.
Complicated	A good answer can be found through expertise and comparison.	Bring expertise, analyse options and decide against criteria.	Select a fulfilment partner, review platform architecture or compare funding terms.	Letting expert confidence hide business consequences.
Complex	The answer will emerge through interaction with the market, team or system.	Run safe-to-fail probes and learn quickly.	Test a new market, pricing model, onboarding flow or organisation design.	Writing a grand plan for behaviour that has not happened yet.
Chaotic	Very little is useful until the situation is made safer.	Stabilise first, then reassess.	Contain an outage, supplier collapse, cash emergency or public trust problem.	Holding workshops while damage continues.
Disordered	The team does not yet know which domain applies.	Break the situation into parts and classify each part.	A messy growth problem with finance, operations, customer and people issues mixed together.	Using one method for a multi-domain situation.

How To Classify The Situation

Describe the situation without naming the solution. For example: "refunds are rising", not "we need a better policy".

Ask what kind of cause and effect exists. Is it obvious, knowable through expertise, emerging through behaviour or currently unstable?

Choose the response mode. Use process, analysis, probes, stabilisation or decomposition depending on the domain.

Review whether the domain has shifted. A chaotic issue should become complex or complicated once stabilised. A complex pattern may become clear enough to standardise later.

Decision Flow: Analyse, Probe Or Stabilise

When to analyse

Analyse when the answer is knowable but not obvious. Use this for technical choices, supplier comparisons, legal interpretation, finance options or operational design where expertise can reduce uncertainty.

When to probe

Probe when behaviour matters and prediction is weak. Use this for customer response, new markets, culture change, positioning, product adoption or any decision where the system reacts back.

When to stabilise

Stabilise when continuing discussion allows damage to grow. Protect customers, cash, safety, data, reputation or delivery first. Then move the situation into a domain where better thinking becomes possible.

When to split the problem

Split it when people are arguing from different assumptions. One part may be clear, another complicated, another complex and another chaotic. Treating the whole thing as one domain is often the real mistake.

How Situations Move Between Domains

Domains are not fixed labels. A supplier issue may begin as chaotic because customers cannot be served today. Once orders are paused, customers are contacted and cash exposure is known, it may become complicated: compare alternative suppliers, costs and lead times. Later, it may become clear: add a backup supplier rule and reorder threshold.

The same movement can happen in the other direction. A clear process can become chaotic if ignored for long enough. A complicated platform choice can become complex if staff behaviour, customer adoption and vendor incentives start interacting in unexpected ways.

Real-World Examples

Supplier failure

A supplier misses a delivery. If the replacement process is known, this is clear: follow it. If the team must compare alternative suppliers, quality, lead time and margin, it is complicated. If the failure exposes customer promises, founder capacity, cash timing and trust all interacting, it is complex. If customers are already affected and orders cannot be fulfilled, stabilise first.

New market behaviour

A product works with one customer group, then behaves strangely with another. Competitor research may help, but only so far. If buyers do not understand the offer, trust the claims or use different buying cues, the situation is complex. Run small market probes rather than pretending the old playbook will transfer unchanged.

Operational crisis

A fulfilment mistake becomes a public complaint. The first move is chaotic-domain work: pause the affected process, contact customers, stop further harm and protect trust. After that, the work may become complicated: diagnose process failure, training gaps, supplier issues or system limits. Later, it may become clear: update the checklist and ownership.

Product or strategy uncertainty

A team is deciding whether to reposition a product. If the data clearly shows one segment converting and retaining better, the decision may be complicated. If nobody knows how customers will respond to a new promise, it is complex. Use small messaging tests, sales conversations and reversible offers before committing the whole brand.

Hiring and organisation design

Writing a job advert can be clear. Benchmarking salary and role options can be complicated. Knowing whether the founder is ready to delegate, whether the team will accept new ownership and whether the role changes customer experience is complex. A short contractor brief or trial project may reveal more than a perfect org chart.

Technology or platform choice

Comparing features and costs is complicated. Predicting how the team will actually use the tool is complex. If the current platform is down and orders cannot be taken, the immediate situation is chaotic: restore service first, then analyse whether the platform should change.

Misclassification: Where Decisions Go Wrong

Checklist abuse in complex situations The team applies a tidy process to customer behaviour, culture or market response before the pattern has emerged.
Expert overreach A specialist treats a complex situation as merely complicated because their tools are built for diagnosis, not emergence.
Endless workshops during chaos People keep sense-making while the business is still losing money, trust, data, delivery control or customer safety.
Premature certainty The team decides too early that it has found the answer, then stops noticing contradictory signals.
Treating emergence as failure A complex situation produces surprises and the team assumes planning failed. Sometimes the real failure is expecting planning to remove emergence.

How To Use This In A Real Decision

Pause before choosing the method. Do not default to a workshop, spreadsheet, expert call or checklist. First classify the situation.

Ask what kind of cause and effect exists. Obvious means clear. Knowable means complicated. Emerging means complex. Absent or unstable means chaotic.

Choose the response mode. Use known process, expert analysis, safe-to-fail probes, stabilisation or decomposition.

Set a review point. Ask whether the domain has changed after action. Stabilised chaos often becomes complex or complicated. Repeated complex learning can become clear process.

Watch for false clarity. If stakeholders keep disagreeing, customers keep behaving unexpectedly or the same problem returns in new forms, the situation may not be clear yet.

Quick Diagnostic Questions

Cause and effect

Is cause and effect obvious, or are we pretending it is because we want a quick answer?

Expertise

Could an expert analyse this to a good answer, or would the expert still need to see how people respond?

Emergence

Will the answer only emerge through trying something small and watching what changes?

Stability

Is the situation unstable enough that we must act first and analyse properly once damage is contained?

Disagreement

Are stakeholders disagreeing because they are seeing different domains inside the same situation?

Review

What would tell us the situation has shifted from chaotic to complex, or from complex to complicated?

Good Actions By Domain

Clear: use the known process, keep standards visible and do not turn routine work into theatre.
Complicated: bring expertise, compare options, check assumptions and make the trade-offs explicit.
Complex: run safe-to-fail probes, limit downside, learn quickly and amplify what starts working.
Chaotic: stabilise first, communicate clearly, stop further damage and reassess once the ground is firmer.
Disordered: split the situation into parts, classify each part and stop arguing as if one method fits everything.

Connected Patterns And Decisions To Use With This

Cynefin pairs well with Speed vs Certainty when the team is unsure whether more evidence is possible before action. Use Project Premortem once a plan is forming and you want to expose failure points before committing.

If the situation keeps being deferred, check Analysis Paralysis Loop and Decision Debt. If the question is whether a step is safe enough to try, use the Reversibility Check. If the team is testing a market, pair Cynefin with the Validation Signal Ladder.

In DecisionForge, this connects strongly to Something Is Breaking and You Take a Step Back, where the useful move is not always to push harder.

Connected Patterns And Decisions

Insight The Speed vs Certainty Dilemma

Cynefin helps decide whether certainty is achievable before action.

Framework Project Premortem

Useful once the context is clearer and a plan is forming.

Insight Decision Debt

Use when repeated deferrals are making the situation harder to classify.

Tool Reversibility Check

Useful for designing safe-to-fail probes in complex situations.

Tool Validation Signal Ladder

Use when customer behaviour needs to be tested rather than assumed.

Pattern Analysis Paralysis Loop

Use when complicated analysis has become avoidance.

Archetype Optimistic Operator

The archetype that often treats complex operations as simple effort problems.

DecisionForge Something Is Breaking

A situation where the domain may shift from complicated to complex or chaotic.

DecisionForge You Take a Step Back

The book moment where sense-making becomes more valuable than motion.