Glossary/Attention score

Attention score

An attention score is a single 0–100 number summarizing how well a page directs predicted visual attention: whether attention concentrates on the elements that matter — headline, offer, call to action — or scatters across decoration. It turns a heatmap into something you can benchmark and compare.

A heatmap is rich but hard to act on alone: is this blob pattern good? A score answers by reducing the map to measurable properties — how concentrated attention is versus scattered, how clear the No. 1 focal point is (hierarchy), and whether the hot zones overlap the elements the page exists to get seen.

The value is comparability. A score lets you rank a page against others in its category, compare two design variants before choosing, and verify that a redesign actually improved attention flow instead of just looking newer. It works like a performance budget for attention, the way Lighthouse works for load time.

Heatpoints computes its score from the UNISAL saliency map of a page screenshot: the predicted fixation density is analyzed for concentration, hierarchy clarity, and coverage of key regions, then aggregated to 0–100. Because the input is a screenshot, the same scoring runs on a live URL, a staging link, or a mockup.

Treat the number as a regression check, not a grade to worship: score, change one thing, re-score. A rising score with the heat landing on your CTA is signal; chasing 100 for its own sake is not.

See it on your own page

Free score /100 · no signup · results in seconds