Choose the road.
Show your reasons.
The crawlerway runs one way to the launch pad, and every turn on it was decided long before. Put your options side by side, weight what matters, knock out anything that misses a must-have, and see which choice wins, by how much, and how easily the answer could change. Free, and nothing leaves this browser.
A ranking you can defend in the meeting.
Score each option 1–10 on each criterion. Mark criteria where lower is better (cost, risk, effort) and they're inverted automatically. Weights are rescaled to 100%, so you can think in rough importance and still get exact maths.

Know how solid the answer is.
Sensitivity shows how far each weight must move before the leader changes. A robustness test re-scores the matrix 500 times with every weight varied by up to 25% and reports how often each option still wins.
Must-haves, not just weights
Set a minimum (or, for costs, a maximum) score on any criterion. Options that miss it are listed separately instead of quietly winning on everything else.
Checks every score
Missing scores, out-of-range numbers and duplicate options are flagged and left out of the ranking until fixed, so a typo can't decide for you.
Templates and exports
Start from a software-vendor, business-idea (the ARQS 12-dimension model) or location template, then export a Markdown decision record, a CSV matrix or JSON.
Decision matrix
A sample (choosing a project management tool between four fictional vendors) is loaded. Change any score or weight and the result updates instantly. Saved in this browser only.
Ranking
Score breakdown
| Criterion | Score | Weight | Normalised | Points |
|---|
Must-have check
Score checks
Score each option from 1 (worst) to 10 (best) on its own terms; for "lower is better" criteria enter the raw level (e.g. 9 = very expensive) and it's inverted for you. Half points are fine.
Weight share (always adds to 100%):
Add a criterion
Weight sensitivity
Points one weight would need to move (others rescaled) before a different option leads.
| Criterion | Weight | Leader changes if… |
|---|
Robustness test
500 re-scores with every weight varied by up to ±25% (repeatable). Share of runs each option wins: