Decision Matrix

For each option (row) and each criterion (column), you assign a score from 1 to 10. For each criterion, you also assign a weight from 1 to 10 that reflects how much that criterion matters to you. The tool multiplies the score by the weight for every cell, then sums those products to give the option's total. So if Option A scores 8/10 on Salary (weight 9) and 5/10 on Commute (weight 3), its weighted contribution from those two criteria is (8*9) + (5*3) = 87. The total is summed across all criteria, then the options are sorted by total descending. The top three get gold/silver/bronze styling, and the top option is shown as the recommendation in the winner card. The math is intentionally simple and transparent — there is no fuzzy logic, no hidden normalization, no surprise weighting.

The weights are the most subjective part of the matrix and that is the point. A common approach is "must-have vs nice-to-have": if a criterion is essential (e.g. salary must cover your rent), give it a high weight (8-10). If it is a tie-breaker (e.g. free coffee), give it a low weight (1-3). Another approach is "comparative weighting": if Salary matters twice as much as Commute to you, set Salary weight = 8 and Commute weight = 4 (or 6 and 3, or 10 and 5 — any consistent ratio). The absolute scale (1-10) does not matter as much as the relative ratios between criteria. After you build the matrix, try flipping one weight from 5 to 9 and see if the winner changes — if not, the criterion is not actually doing work and you can drop it.

Every time you click Export JSON, a snapshot of the current matrix (options, criteria, scores, weights) is saved to localStorage under the key decision-matrix-saved as a JSON array, max 10 entries. The "Saved Decisions" card at the bottom shows these snapshots with their name (auto-generated from the option names), timestamp, and a Load / Delete button pair. Load replaces the current matrix with the saved one. Delete removes it from the list. The list is local to this browser — it does not sync anywhere. If you want a permanent backup, use Export JSON and save the file to your device or cloud storage; you can re-import it later with the Import button.

Technically yes, the tool will accept any number of options and criteria. Practically, the matrix becomes hard to read beyond 5-6 options (the table is wide on mobile) and beyond 5-6 criteria (you start losing the ability to differentiate scores consistently). For larger comparisons, consider grouping criteria into categories (e.g. "Financial" weight = sum of 3 sub-criteria, "Lifestyle" weight = sum of 3 sub-criteria) and running the matrix at the category level. Or build multiple matrices — one for "must-haves" filter, one for "trade-off comparison" — and combine the conclusions.

Yes. The matrix is pure client-side JavaScript inlined in the HTML. There are no network calls, no API uploads, no analytics, no CDN. Your options, criteria, scores, and weights stay in localStorage on this device. The Export JSON button uses the standard browser <a download> click to save to your local Downloads folder. Import reads a JSON file from your local file system via FileReader. You can model a personal decision (job choice, relationship, big purchase) without any data being transmitted. Clearing browser data for this site clears the matrix and the saved list.