Criteria Weighting Decision Matrix

Overview

Also known as: Weighted scoring model, Decision matrix, Pugh matrix.

Complex decisions become manageable when broken into:

1. Independent criteria
2. Importance weights
3. Option ratings
4. Mechanical aggregation

Each individual judgment is simple; the math does the combining.

Goal

Make decisions by decomposing into criteria, weighting by importance, rating options on each criterion, and calculating weighted scores. Decomposition makes evaluation tractable; aggregation is mechanical.

Steps

Step 1: Define the Decision

Clearly state what you’re deciding. What choice are you making? What’s the goal?

Output: Decision statement

Step 2: List All Options

Enumerate all options being considered. Include “do nothing” or “status quo” if relevant.

Output: Option list

Step 3: Identify Criteria

List all factors that matter for this decision.

Good criteria are:

• Relevant (actually affects decision quality)
• Measurable (can rate options on it)
• Independent (not redundant with other criteria)

Common criteria types:

• Cost/price
• Quality/performance
• Time/speed
• Risk
• Ease/convenience
• Alignment with goals
• Stakeholder preference

Output: Criteria list

Step 4: Assign Weights

Distribute 100 points across criteria by importance. Higher weight = more important in decision.

Methods:

• Direct allocation: Just assign points
• Pairwise: Compare criteria pairs to derive weights
• Ranking: Rank criteria, assign points by rank

Weights should sum to 100 (or 1.0).

Output: Weighted criteria

Step 5: Rate Options on Each Criterion

For each option, for each criterion: Rate on scale of 1-10 (or 1-5).

1 = Worst possible on this criterion 10 = Best possible on this criterion

Rate consistently across options.

Output: Rating matrix

Step 6: Calculate Weighted Scores

For each option: Weighted Score = Σ (Weight_i × Rating_i)

If weights sum to 100 and ratings are 1-10, max possible score = 1000.

Output: Weighted scores

Step 7: Rank Options

Sort options by weighted score (descending). Highest score = recommended option.

Output: Final ranking

Step 8: Sensitivity Analysis

Test how robust the ranking is:

• What if weights changed?
• What if ratings changed?
• How close are the top options?

If top two are very close, decision is sensitive. If top option wins by large margin, decision is robust.

Output: Sensitivity assessment

Step 9: Sanity Check

Does the recommended option feel right? If not, examine:

• Are criteria missing?
• Are weights wrong?
• Are ratings inaccurate?

The matrix is a tool to structure thinking, not replace judgment.

Output: Final decision

When to Use

• Choosing between multiple options
• Making decisions with multiple criteria
• Need to justify decision transparently
• Want to reduce bias from single-factor focus
• Team needs to align on priorities

Verification

• All relevant criteria are included
• Criteria are independent (not double-counting)
• Weights sum to 100 (or 1.0)
• Ratings are consistent across options
• Calculations are correct
• Sensitivity analysis performed
• Result passes sanity check

Input: $ARGUMENTS

Apply this procedure to the input provided.