Scroll Depth

Scroll depth measures how far down a page a user scrolls before leaving. It is a proxy for engagement and content effectiveness. It tells you whether users are actually reading your content or bouncing before they reach the important parts.

Category: Cro (also: Analytics, UX)

Author: Atomic Glue Research Team

## Definition

Scroll depth is a behavioral metric that tracks the percentage of a page a user has scrolled through. It is typically measured at thresholds: 25%, 50%, 75%, and 100% of page height. Scroll depth is a leading indicator of engagement. It tells you whether users are consuming your content or abandoning it. Scroll depth is not a conversion metric. It is a diagnostic metric. High scroll depth without conversion means the content is engaging but the CTA or value proposition is failing. Low scroll depth means the content is not compelling enough to keep users reading.

## What Scroll Depth Actually Measures

Scroll depth measures scrolling behavior, not reading behavior. A user can scroll to 100% of the page without reading a word. This is the 'scrolling past content' problem. Scroll depth is a proxy for engagement, not a direct measure. Combine it with other metrics: time on page, click heatmaps, and attention maps. A user who scrolls to 75% in 10 seconds is not reading. A user who scrolls to 75% in 90 seconds is. The combination matters more than the raw number.

## Using Scroll Depth for CRO

Scroll depth reveals: (1) Whether users are reaching your CTA (if the CTA is at 90% scroll depth and most users stop at 50%, move the CTA up). (2) Which content sections are losing users (if 70% of users reach the 50% mark but only 30% reach 75%, the content between 50-75% is the problem). (3) Whether long-form content is justified (if scroll depth to 100% is high, long-form content is working. If not, truncate it).

## The Infinite Scroll Problem

Infinite scroll pages (social feeds, content galleries) make scroll depth harder to measure. A user can scroll through 300% of 'page height' on an infinite scroll page. The metric shifts from 'percentage of page' to 'pixels scrolled' or 'sections loaded'. For infinite scroll, use time-based engagement metrics (time on page, number of items viewed) rather than percentage-based scroll depth.

## Note

A common mistake is placing the CTA at the bottom of a long page and wondering why it does not convert. If scroll depth data shows 70% of users never reach the CTA, the CTA is not at the bottom of the page. It is at the bottom of the ocean. Move it up. Repeat the CTA at multiple scroll depths.

## In code

<!-- Scroll depth tracking with Google Analytics 4 -->
<script>
  // Track scroll depth thresholds
  let scrollDepthsSent = new Set();

  function trackScrollDepth() {
    const scrollHeight = document.documentElement.scrollHeight - window.innerHeight;
    const scrollPosition = window.scrollY;
    const scrolled = (scrollPosition / scrollHeight) * 100;

    const thresholds = [25, 50, 75, 90, 100];

    thresholds.forEach(threshold => {
      if (scrolled >= threshold && !scrollDepthsSent.has(threshold)) {
        scrollDepthsSent.add(threshold);
        gtag('event', 'scroll_depth', {
          'event_category': 'engagement',
          'event_label': `${threshold}%`,
          'value': threshold
        });
      }
    });
  }

  // Use passive scroll listener for performance
  window.addEventListener('scroll', trackScrollDepth, { passive: true });
</script>

## Common questions

Q: What is a good scroll depth?

A: It depends on page length and content type. For a standard blog post, 50-75% average scroll depth is typical. For a landing page, 75%+ is the target. Compare against your own historical data, not benchmarks.

Q: How do I improve scroll depth?

A: Improve the content above the fold to give users a reason to scroll. Use progressive disclosure: reveal compelling content at each scroll level. Use visual cues (arrows, partial images, text cutoffs) that signal more content below.

Q: Does scroll depth correlate with conversion?

A: Generally, yes. Users who scroll further are more engaged and more likely to convert. But correlation is not causation. Some users scroll the entire page without converting. Others convert immediately without scrolling. Use scroll depth as a diagnostic, not a target.

## Key takeaways

• Scroll depth is a proxy for engagement, not a direct measure of reading
• Combine scroll depth with time on page and click data for context
• If users do not reach the CTA, move the CTA up. Repeat it at multiple scroll depths
• Infinite scroll pages require different measurement approaches (pixels scrolled, not percentages)
• Use scroll depth as a diagnostic tool, not a conversion target

## Related entries

• [Above the Fold](atomicglue.co/glossary/above-the-fold)
• [Hero Section](atomicglue.co/glossary/hero-section)
• [Micro-Conversion vs Macro-Conversion](atomicglue.co/glossary/micro-conversion-vs-macro-conversion)
• [Friction](atomicglue.co/glossary/friction)

Last updated July 2026. Permalink: atomicglue.co/glossary/scroll-depth