Decision Matrix Calculator
Score several options against several criteria, weight the criteria by how much each one matters, and get a ranking instead of an argument.
Results
What this tool does
When a choice has more than one dimension — price and quality and delivery and support, all at once — holding it in your head stops working at about three options. A decision matrix is the standard way out: list what matters and how much, mark each option on each, and let the weighted average do the comparing. The ranking comes back in order, with the gap between first and second shown, because a narrow win is information too.
Formula
score of each option = Σ(mark × criterion weight) ÷ Σ weights
Variables
| Symbol | Meaning | Unit |
|---|---|---|
criteria | What matters, one per line, with a weight | — |
options | Each option, with a mark for each criterion | — |
scale | Top mark | — |
OUT | Comes out on top | — |
TB | The ranking | — |
SC | Its weighted score | — |
PC | Out of the top mark | % |
RU | Second | — |
LD | It leads by | — |
NC | Criteria | — |
HV | The criterion that weighed most | — |
Worked example
- What matters, one per line, with a weightPrice = 3 Quality = 5 Delivery time = 2 Support = 4
- Each option, with a mark for each criterionSupplier A = 8, 6, 9, 5 Supplier B = 5, 9, 7, 8 Supplier C = 7, 7, 7, 7
- Top mark10
- Comes out on topSupplier B
- The ranking1. Supplier B 7.57 76% 2. Supplier C 7.00 70% 3. Supplier A 6.57 66%
- Its weighted score7.57
- Out of the top mark75.7 %
- SecondSupplier C
- It leads by0.57
- Criteria4
- The criterion that weighed mostQuality
Limitations
- The calculation runs entirely in your browser. The values you type are never sent to a server.
Frequently asked questions
How do I write the options?
A name, an equals sign, then one mark for each criterion in the same order you listed them, separated by commas. If you list four criteria, each option needs four marks; any you leave out count as zero. The scale is yours — out of ten is usual, but out of five or out of a hundred work exactly as well, as long as every option is marked on the same one.
The winner is barely ahead. Does that matter?
Very much. A lead of two tenths of a point means the marks and weights would have to be accurate to a tenth for the ranking to be trustworthy, and nobody's judgement of "how good is their support, out of ten" is that precise. When the top two are close the honest reading is that the matrix cannot separate them and something outside it — who you would rather work with, what your instinct says — is allowed to decide.