Complete firstDecision Framing Workshop
Decision question, options, and relevant conditions are at least roughly clear.
Without: Without framing, the tree becomes a catch-all for everything that could happen.
RunsheetDecision Making
Decision Tree
ComplexityMedium
Time30-90 min
Participants1-6
FormatWorkshop + async
MaturityCanonical
01
Prerequisite
What needs to be finished first
02
Preparation
What needs to be ready before start
Materials
Whiteboard or diagram tool; decision question; known options; criteria, conditions, probabilities, or outcomes; markers for assumptions.
People / roles
Facilitator or decision owner; 1-6 subject experts; decider for final interpretation; scribe for the tree.
Pre-read
Starting question, options, decision criteria, known risks, probability data, and non-negotiable constraints.
Time needed
30-90 min
Setup
Start with one single decision question. Draw tree left to right. Phrase each branch as a question or condition.
03
Core question
The one question this method answers
Which option or rule emerges when conditions, consequences, and uncertainties are explicitly branched?
04
Flow
Marker: Phase
| Step | Duration | Action | Hint |
|---|---|---|---|
1Set root question | 10 min | Formulate the decision question and scope as the root of the tree. | If the root contains multiple questions, the tree becomes unreadable. |
2Branch options | 15-20 min | Add primary options or decision paths as first branches. | Options should be mutually exclusive or clearly combinable. |
3Add conditions | 20-30 min | For each branch, add relevant conditions, risks, probabilities, and follow-up decisions. | Only include conditions that materially change the decision. |
4Evaluate outcomes | 15-25 min | Describe endpoints with benefit, cost, risk, or expected consequence. | Always mark source or assumption for numbers. Avoid pseudo-precision. |
5Select path | 10-15 min | Identify robust paths, mark open assumptions, and define next evidence steps. | The tree supports decision making, but it does not replace the decider's authority. |
05
Artifact
What comes out at the end
Form
Decision Tree with root question, options, conditions, outcomes, assumptions, and a recommended path marked, plus short decision note.
Versioning / ownership
Save tree with date and assumptions. If conditions or probabilities change, create new version with change note.
Tool alternatives
- Miro or FigJam
- Lucidchart
- draw.io
- Whimsical
- Mermaid in Markdown
decision-tree-working-template.md
Compact working template for Decision Tree with decision question, branches, outcomes, and recommendation.
Decision Tree Working Template
Goal
Map options, conditions, and outcomes so the decision becomes explicit.
Context
When and for what do we use this method?
Input
Which data, observations, decisions, or materials are available?
Working area
- Decision question:
- Options:
- Conditions:
- Branches and outcomes:
- Risks and probabilities:
Output artifacts
- Decision tree:
- Option map:
- Assumption list:
Open questions
- ...
Next step
Owner, date, and success signal.
06
Example output
Concrete filled scenario, fictional example
decision-tree-beispiel.md
Concrete filled scenario, fictional example
Decision Tree - Support Chatbot rollout
Root question: Roll out to all customers or only a beta segment?
- Option A: Full rollout
- If intent detection stays above 85%: faster ticket reduction, moderate risk.
- If intent detection falls below 85%: high misrouting, support backlog increases.
- Option B: Beta segment
- If 200 customers participate: reliable learning rate, low risk.
- If fewer than 50 participate: insufficient data, rollout delays.
Recommendation: Option B with minimum criterion of 200 beta customers and review after 14 days.
07
Pitfalls
Recognize symptoms and steer against them
Trap
Tree becomes process model
Symptom
Every operational step is represented as a branch.
What to do
Model only decision points and conditions; put process details in separate artifacts.
Trap
Too many levels
Symptom
Nobody can explain the tree at a glance.
What to do
Limit to 3-4 levels and model subtrees separately.
Trap
Unclear outcomes
Symptom
Endpoints are labeled "good" or "bad" only.
What to do
Describe outcomes with concrete benefit, cost, risk, or next decision.
Trap
Invented probabilities
Symptom
Percent values look precise without source.
What to do
Mark them as assumptions or use qualitative classes.
Trap
No decision
Symptom
Tree is complete but no path is chosen.
What to do
Name a decider and document result as recommendation or decision.
08
Stop criteria
Done signals checkable in under a minute
No concrete decision question is available.
Options are not distinguishable.
Conditions do not influence decision.
Tree grows beyond readable complexity.
Decider or next evidence step is missing.
Finished the runsheet?
Go to the profile for purpose, similar methods, and sources or continue to the next method in the catalog.