Investigate root-cause hypotheses with contrary evidence
Treat causes as testable hypotheses with evidence for, evidence against, confidence and explicit rejection or confirmation.
State the hypothesis
- Describe a mechanism, not a person to blame.
- Assign an investigation owner.
Seek both directions
- Retain supporting and contrary evidence.
- Unknown confidence remains unknown.
Decide explicitly
- Confirm or reject from evidence.
- Do not convert repetition into causation.
Retain the problem record
The Problem Management Studio keeps hypotheses, contrary evidence, factors, actions, checks, recurrence observations and decisions connected. It does not change systems, assign blame, spend money or prove effectiveness without observation.