F-pattern layout
On poorly formatted, text-heavy left-to-right pages, readers fall into an F-shaped scan: full first line, a shorter second pass, then a skim down the left edge.
See it
What it is
On text-heavy pages people don't read, they scan. Eye-tracking work from the Nielsen Norman Group found the gaze often traces an F: a full sweep across the first line or two, a shorter sweep further down, then a vertical skim along the left edge hunting for something worth stopping at. Blog posts, docs, search results, and long product copy all get this treatment.
Design with it by front-loading value: put the point in the first few words of every heading and paragraph, keep the left edge loaded with subheads and bullets, and never bury the important noun at the end of a line. In right-to-left languages the F flips to a mirror image, so the left rail becomes the right rail.
Gotcha: the F is a symptom, not a goal. NN/g's own reading is that people fall into F-scanning when text is undifferentiated, and good formatting (subheads, bold lead-ins, short paragraphs, images) breaks the pattern and gets more of the page actually read. Don't shape content into an F. Shape it so the F isn't necessary.
Ask AI for it
Restructure this text-heavy page for F-pattern scanning. Put the single most important sentence in the top line of the content column. Front-load every heading and paragraph so the key noun is in the first three words. Add scannable subheads every 2 to 3 paragraphs, bold lead-ins on list items, and keep the left edge of the column flush so the vertical skim has a clean rail. Cap the measure at 65 to 75 characters and keep paragraphs under 4 lines.