Complete firstDecision occasionnot in catalog
A decision with 2-7 bounded options is available.
Without: Without comparable options, the Decision Matrix becomes a wish list instead of a comparison.
RunsheetDecision Making
Decision Matrix
ComplexityMedium
Time45-90 min
Participants2-8
FormatWorkshop + async
MaturityEstablished
01
Prerequisite
What needs to be finished first
Complete firstOption Framing
Options are described with similar depth so they can be compared.
Without: Without equal detail depth, the best-presented option wins and bias increases.
02
Preparation
What needs to be ready before start
Materials
Whiteboard, Miro, or table with rows for options and columns for criteria plus weighting; dot voting or digital voting; scoring scale (for example 1-5 or 1-10).
People / roles
A decider with authority; 3-7 evaluators with relevant knowledge; one facilitator; one scribe; optionally a reviewer for consistency.
Pre-read
Option descriptions; criteria list with per-criterion definitions; weighting approach; known preferences or objections.
Time needed
60-120 min
Setup
Create matrix. Define and weight criteria (sum 100). Agree on scale (1-5 or 1-10). Rule: every score has short rationale.
03
Core question
The one question this method answers
Which option achieves the highest weighted score, and which sensitivity cases influence the recommendation?
04
Flow
Marker: Phase
| Step | Duration | Action | Hint |
|---|---|---|---|
1Phase 1: Clarify options | 15 min | Briefly present options, ensure all evaluators understand them equally. Record assumptions for each option. | If options have different detail, align them before scoring. Otherwise the best-written option wins. |
2Phase 2: Criteria and weighting | 20 min | Criteria per decision (3-7). For each criterion define what counts as good. Weighting (sum 100, or pairwise comparison). | Too many criteria dilute quality. Maximal 7. If all criteria are equally important, the team has not prioritized. |
3Phase 3: Scoring | 30 min | Each evaluator scores each option and criterion (1-5 or 1-10). Add brief rationale. Then calculate average or median per cell. | Score anonymously or simultaneously to avoid anchoring effects. High variance means discussion is required. |
4Phase 4: Score and sensitivity | 20 min | Calculate weighted score per option. Sensitivity: vary weights by ±10% and check whether top option stays stable. | If recommendation flips with small weight changes, it is not robust. The decider must then set weights explicitly. |
5Phase 5: Decision and ADR | 15 min | Decider announces decision with rationale. Write an ADR or decision note. Add follow-up actions with owner and due date. | Decider may override the matrix (for example due to unmeasured factors), but must provide rationale. Otherwise the matrix loses credibility. |
05
Artifact
What comes out at the end
Form
Decision Matrix table with options, criteria, weights, scores, and weighted total score. Supporting ADR with decision, rationale, sensitivity notes, assumptions, and follow-up actions.
Versioning / ownership
Archive snapshot with date. If assumptions change, run a new matrix instead of overwriting. ADR references the matrix snapshot.
Tool alternatives
- Google Sheets or Excel with formulas
- Notion or Confluence table
- Airtable with evaluation forms
- Miro with voting feature
- Specialized tools such as DecideAct or PairWise
trade-off-matrix-markdown.md
Markdown matrix for comparing options across competing criteria and accepted trade-offs.
Trade-off Matrix: Decision
| Criterion | Weight | Option A | Option B | Option C |
|---|---|---|---|---|
| Benefit | 30 | |||
| Cost | 20 | |||
| Risk | 20 | |||
| Time-to-Learn | 15 | |||
| Reversibility | 15 |
Accepted trade-offs
- Option A accepts:
- Option B accepts:
- Option C accepts:
Recommendation
Which option is recommended and why?
Sensitivity
Which small change in weighting or assumption would flip the recommendation?
06
Example output
Concrete filled scenario, fictional example
decision-matrix-beispiel.md
Concrete filled scenario, fictional example
Decision Matrix - Tax-client communication tool selection at Bertram law firm, May 2026
Options
- A: DATEV Meine Steuern (built in, limited)
- B: Hubdoc + Email
- C: AppShake.tax (industry solution)
Criteria and weighting
| Criterion | Weight |
|---|---|
| GDPR compliance | 30 |
| DATEV integration | 25 |
| Client usability | 20 |
| Monthly costs | 15 |
| Onboarding effort | 10 |
Scores (1-5)
| Criterion | A | B | C |
|---|---|---|---|
| GDPR | 5 | 3 | 4 |
| DATEV | 5 | 2 | 4 |
| Usability | 2 | 4 | 4 |
| Cost | 5 | 3 | 3 |
| Onboarding | 4 | 4 | 3 |
Weighted score
- A: 50.3 + 50.25 + 20.2 + 50.15 + 4*0.1 = 4.30
- B: 30.3 + 20.25 + 40.2 + 30.15 + 4*0.1 = 3.05
- C: 40.3 + 40.25 + 40.2 + 30.15 + 3*0.1 = 3.75
Sensitivity: With usability weight +10pp (to 30), score shifts to A 4.15, C 3.85. A remains top, C gets closer.
Decider (Sabine): Option A (DATEV Meine Steuern), GDPR and DATEV integration outweigh usability, which can be compensated with client training.
Follow-up actions: Client training webinar for top-20 clients on 15.06. (Sabine), create help video (Jana, by 30.06.).
07
Pitfalls
Recognize symptoms and steer against them
Trap
Criterion inflation
Symptom
12 criteria are evaluated and each option can appear near the top.
What to do
Keep max 7 criteria. Combine similar criteria. If hierarchy is needed, decide in two stages.
Trap
Anchoring effect
Symptom
First score influences all others, and variance becomes artificially low.
What to do
Score anonymously and simultaneously. Compare averages only afterward. Discuss when variance is high, otherwise do not force consensus.
Trap
Decider follows blindly
Symptom
Recommendation is accepted without sensitivity checks, and an unstable result wins.
What to do
Sensitivity is mandatory. Vary weighting. If recommendation is unstable, restart with new rationale or defer decision.
Trap
Unevenly detailed options
Symptom
One option has a five-page description, another only two sentences.
What to do
Run Option Framing before Decision Matrix. Align detail and assumptions. Otherwise presentation quality dominates.
Trap
Score over discussion
Symptom
Rationales are missing and numbers look arbitrary.
What to do
Add a short rationale (one sentence) per score. When scores diverge, compare rationales; often different assumptions are behind them.
08
Stop criteria
Done signals checkable in under a minute
Fewer than two real options, no comparison basis.
Criteria are not measurable or cannot be defined.
Decider not in the room, recommendation remains floating.
Option detail differs so much that comparisons are distorted.
Sensitivity shows total instability and decision is not mature.
Decision is already politically made; the matrix only legitimizes it.
Finished the runsheet?
Go to the profile for purpose, similar methods, and sources or continue to the next method in the catalog.