contrast pair / sheet 3 · free workflow
Rank Your Colour Pairs by Which One Fails First — Free
Paste each colour pair with its measured contrast ratio and size, and the sheet ranks them so the two that fail the threshold are the first thing you fix rather than something a report mentions in passing.
What this replaces
An accessibility report lists every failure at equal weight. This ranks your pairs by measured ratio and size, so the two that fail hardest are at the top of the fix list. The page does not compute the ratios — it ranks the ones you measured, and says so.
Tool note. This runs entirely in your browser — no account, no upload, no paid API key. Google NotebookLM is named as the free source-grounded reading path for the material you paste; its published free-tier terms were checked on 2026-09-13 and can change.
The four passes
Collect the smallest useful input
One pair per line with the ratio you measured and the font size in pixels.
Let the sheet rank them
Each gets a score with the two numbers behind it.
Fix the bottom two first
A ratio of 2.2 is not a near miss, it is unreadable.
Re-measure after the change
A new ratio moves the pair up the ranking.
Try it on the contrast pair / sheet 3
The board below runs in your browser. It opens with a worked sample so the result is visible before you paste anything; switch scenarios, edit the text, or reset to the sample.
running the sample…
first-load · sample values are illustrative, not verified facts
Everything you paste stays in this browser tab. Nothing is uploaded, there is no account, and no request leaves the page to generate the result; clearing site data removes anything saved locally.
What breaks, and what to do
Ratio not measured
The page cannot compute it; measure with a browser inspector or a checker.
Text over an image
The ratio varies per pixel; measure the worst case.
Disabled controls
Exempt from the threshold in most standards, but still worth ranking.
Go deeper — Contrast Pair
The free board above is the whole loop. The paid tier keeps the arranged output so you do not rebuild it.
- Printable sheet with the failing pairs boxed for a design review — estimate: 16 minutes saved per audit, basis: 5 pairs re-tabulated at ~3.2 minutes each
- Saved local history across audits — estimate: 11 minutes saved per audit, basis: re-entering 5 pairs at ~2.2 minutes each
$4 one-time