Systems archetypes are recurring feedback structures that produce recognizable patterns of behavior. They are diagnostic hypotheses, not labels that prove a cause. Used carefully, they help teams move from isolated events to the structures that repeatedly generate them.
Eight widely used archetypes
- Fixes That Fail: a quick correction reduces a symptom but creates delayed consequences that restore it.
- Shifting the Burden: reliance on symptom relief weakens capacity for a fundamental solution.
- Limits to Growth: reinforcing success eventually strengthens a constraint.
- Drifting Goals: pressure lowers the target instead of improving performance.
- Success to the Successful: early advantage attracts resources and compounds.
- Escalation: each actor responds to the other’s advantage, intensifying competition.
- Tragedy of the Commons: individually beneficial use depletes a shared resource.
- Accidental Adversaries: local corrective actions unintentionally undermine a partner.
How to use them
Start with observed behavior over time, then compare several archetypes. Map the actual variables, delays, goals, and evidence. Ask what would disconfirm the proposed structure. Combining archetypes is often necessary because real systems contain several loops.
Interventions should address structure: reveal delayed effects, protect fundamental capability, change allocation rules, manage shared resources, or remove constraints. Renaming a problem after an archetype is not an intervention.
Common mistakes
- Choosing the first familiar pattern.
- Treating the template as evidence.
- Ignoring power and contested goals.
- Mapping without a time horizon.
- Assuming the same leverage works in every context.
References
- Senge, P. M. (2006). The Fifth Discipline. Doubleday.
- Kim, D. H. (1992). Systems Archetypes I. Pegasus Communications.
- The Systems Thinker, “Systems Archetypes at a Glance”.
From pattern recognition to testable diagnosis
For each candidate archetype, write the loop story in plain language and list the observations it predicts. A Limits to Growth hypothesis predicts that an early reinforcing process will be offset by a strengthening constraint. A Fixes That Fail hypothesis predicts delayed harm produced by the corrective action itself. Compare those predictions with the historical record.
Interview people in different positions because the same structure can look different locally. Frontline staff may observe rework long before it appears in executive measures. Customers may experience costs excluded from internal dashboards.
Intervention questions
- Which loop currently dominates and what could change that dominance?
- Which delay hides the consequence of current action?
- Does the intervention build fundamental capability or only suppress a symptom?
- Could a new rule create gaming or transfer costs?
- Which indicator would reveal that the hypothesis is wrong?
Revisit the map after action. Archetypes are most valuable when they shorten a learning cycle, not when they become permanent stories about why a system behaves as it does.
Discussion