Direct answer: Choose useful criteria, weight them consistently, score evidence, and test sensitivity.
Define distinct criteria
Write criteria so a higher score always means a better outcome. Avoid overlapping ideas such as affordability, price, and cost unless they measure genuinely different things.
Set weights before scoring
Assign relative importance without looking at which option will benefit. Setting weights first reduces the temptation to tune the model toward a preferred answer.
Score from evidence
Use the same 0–10 scale and the same evidence standard for every option. Add short notes outside the matrix when a score depends on an assumption.
Test sensitivity and judgment
Change uncertain scores and important weights. If a small reasonable change reverses the leader, treat the result as close. The matrix clarifies judgment; it does not replace it.
Related calculators and tools
Editorial note
This guide follows our editorial standards. Send corrections through Report an error.