A wicked problem is not merely a very difficult problem. It is a situation in which the definition of the problem is contested, proposed actions change the situation, stakeholders value outcomes differently, and there is no final test proving that the problem has been solved.
Horst Rittel and Melvin Webber introduced the concept in planning and policy. Climate adaptation, housing affordability, public health, educational inequality, and misinformation often display wicked characteristics because technical, political, economic, and cultural dynamics interact.
What makes a problem wicked?
- There is no universally accepted formulation.
- Understanding the problem and proposing action develop together.
- Interventions are judged as better or worse, not simply true or false.
- Consequences may be delayed, distributed, or irreversible.
- Each case has important contextual differences.
- Stakeholders disagree about values, boundaries, and acceptable trade-offs.
These features do not mean “anything goes.” Evidence can reject claims, reveal mechanisms, and compare outcomes. The difficulty is that evidence alone cannot decide all value conflicts or system boundaries.
Why conventional problem solving fails
Linear problem solving assumes the problem can be specified, decomposed, solved, and closed. That works well for many bounded technical tasks. In a wicked situation, decomposition can hide interactions and transfer costs. A transport policy may improve traffic speed while increasing land-use pressure and excluding people without cars.
The search for a single root cause can also mislead. Persistent patterns may be generated by multiple feedback loops, institutional rules, resource constraints, and changing behavior.
A systems approach to framing
- State the current framing and who benefits or disappears within it.
- Map stakeholders, power, incentives, and conflicting goals.
- Describe behavior over time rather than focusing on a recent event.
- Identify accumulations, feedback, delays, and cross-boundary effects.
- Create several plausible framings and compare the actions each makes visible.
- Choose an actionable boundary while documenting what remains outside it.
Act through learning portfolios
Because no plan removes uncertainty, use a portfolio: protective measures for urgent harm, experiments that test mechanisms, structural changes with monitoring, and processes for stakeholder learning. Prefer reversible steps when evidence is weak and consequences could be large.
Agree in advance on signals that would trigger adaptation or withdrawal. Measure distribution as well as aggregate performance. A program can improve the average while worsening outcomes for a vulnerable group.
Avoid analysis paralysis
Systems maps can always expand. The purpose is not to capture everything, but to make assumptions, relationships, and consequences discussable. A practical boundary is justified by the decision at hand, then revisited when new evidence shows that excluded factors matter.
Questions for decision-makers
- Whose definition of success dominates?
- Which costs are shifted across place, group, or time?
- What feedback could offset the intervention?
- Which assumptions can be tested cheaply?
- What would make the action unsafe or ineffective?
- How will affected people influence revision?
References and further reading
- Rittel, H. W. J., & Webber, M. M. (1973). “Dilemmas in a General Theory of Planning.” Policy Sciences, 4, 155–169.
- Head, B. W. (2022). Wicked Problems in Public Policy. Palgrave Macmillan.
- UN Environment Programme, Wicked Problems, Dynamic Solutions.
- Churchman, C. W. (1967). “Wicked Problems.” Management Science, 14(4), B141–B142.
Discussion