scroll depth

Scroll depth turns a vague question (“is anyone actually reading this?”) into a number. Analytics tools track the viewport position as a visitor moves down a page and record the deepest point reached. Google Analytics 4 ships this in enhanced measurement: toggle Scrolls on for a data stream and GA4 logs a scroll event the first time someone hits 90% vertical depth, surfaced through the percent_scrolled parameter and the Percent scrolled dimension.

That default only tells you whether people reached the bottom. To see where they drop off earlier, you add finer checkpoints (25%, 50%, 75%, 100%) through Google Tag Manager, or use a tool like Plausible that records every percentage point from 1 to 100.

Reading the number

Scroll depth is context-dependent, not a universal target. A long guide that most readers abandon at 30% has a hook or formatting problem. A landing page whose CTA sits at 60% depth, where only 40% of visitors get that far, is leaving conversions on the table. Pair scroll depth with engagement time: high scroll plus low time usually means skimming, while low scroll plus high time can mean readers stall near the top.

Watch for false signals. Short pages can hit 90% almost instantly and inflate the metric, and sticky footers or lazy-loaded sections can move where the “bottom” actually lands.