A weighted decision matrix scores every option against the same criteria, weights each criterion by how much it matters, and adds it all up. The highest total wins. The calculator above does the arithmetic live; here is what it computes, using the same criteria and weights as the built-in laptop example.
The example
Four criteria, each weighted from 1 (nice to have) to 10 (critical), and three laptops scored 1 (poor) to 10 (excellent) on each criterion:
| Criterion | Weight | Laptop A | Laptop B | Laptop C |
|---|---|---|---|---|
| Price | 9 | 6 | 9 | 4 |
| Performance | 8 | 8 | 5 | 10 |
| Battery life | 7 | 7 | 8 | 5 |
| Portability | 5 | 8 | 6 | 4 |
| Weighted total | 207 | 207 | 171 |
Laptop A: 9×6 + 8×8 + 7×7 + 5×8 = 54 + 64 + 49 + 40 = 207. Laptop B: 9×9 + 8×5 + 7×8 + 5×6 = 81 + 40 + 56 + 30 = 207. Laptop C: 9×4 + 8×10 + 7×5 + 5×4 = 36 + 80 + 35 + 20 = 171.
A and B tie, which is useful information in itself: the decision now hinges on whether you value performance and portability (A) or price and battery (B). Nudge one weight — say performance from 8 to 9 — and A pulls ahead 215 to 212. If a one-point change flips the winner, the options are effectively equal on your criteria, and you can decide on gut feel without guilt.
Choosing weights
- Rank before you weight. Put the criteria in order first, then assign numbers. If you can't decide which criterion matters more, rank them with the pairwise tool and use the order as your weights.
- Use the whole scale. If every criterion gets 7–9, the weights barely change the outcome. Reserve 9–10 for true deal-breakers.
- Keep criteria independent. "Price" and "value for money" measure the same thing; merging them avoids double-counting.
Reading the results
The results panel ranks options by weighted total and shows each one as a percentage of the best score. Export to CSV to keep a record of the decision, or share a link so others can see exactly how you got there.
More detail: how to build a weighted decision matrix, a copyable weighted decision matrix template with examples, and when to use a decision matrix vs pairwise comparison.