Decision Tree Builder
Develop an analytical decision model, assign probabilities and payoffs, compare expected values, and test how sensitive the recommendation is to uncertainty.
Decision trees make uncertainty explicit before a choice is made.
This builder separates alternatives, chance outcomes, probabilities and payoffs, then compares expected monetary value and tests how robust the recommendation is.
Decision Tree Builder — Step by Step
Eight locked stages move from decision context through alternatives, probabilities, payoffs, expected value, sensitivity, an execution shock and management review.
Alternatives, Outcomes and Expected Value
Reveal the decision context to begin.
Interpret value, downside and sensitivity.
Build Your Own Decision Scenario
Experiment Mode is isolated from the guided demonstration. Change the scenario, probabilities, payoffs and sensitivity stress without affecting guided progress.
Test decision-tree judgement.
Decision-tree cues
Decision Node
A square represents a choice between mutually exclusive alternatives controlled by the decision maker.
Chance Node
A circle represents uncertainty. The probabilities of mutually exclusive, exhaustive outcomes should sum to 100%.
Expected Value
EMV = Σ(probability × outcome value). Compare alternatives using the same payoff basis.
Downside Exposure
Expected value can favour an option that still has a severe adverse outcome. Make the downside visible.
Break-Even Probability
The probability at which two alternatives have equal expected value identifies a useful sensitivity threshold.
Professional Judgement
Expected value is risk-neutral. Consider risk appetite, strategic constraints, reversibility and information quality before deciding.
Use expected value to compare alternatives, then examine downside exposure and sensitivity before turning the model output into a management decision.