How People Read Online: The Eyetracking Evidence

2nd Edition

Content is the core of building positive relationships with users. However, we know that people don’t read digital content completely — they’re much more likely to lightly scan your words than read it in order. So how can you create effective content for users who aren’t actually reading?

This report provides detailed analysis of how people seek out and consume your content. Learn how people process digital content, so you can leverage that understanding in your writing. The report includes 62 recommendations for designing your content to meet the needs of your users.

The findings in this 412-page report are the culmination of three large-scale eyetracking studies spanning 13 years, involving over 500 participants and more than 750 hours of testing session time.


  • How to direct and attract attention with content design
    • Making predictable, scannable pages to improve readability
    • Designing for different page types
    • Optimizing specific elements people look at and read
    • Writing and arranging content to lead people through your site
  • How and why people scan digital content
  • Factors that influence scanning: 
    • Motivation
    • Type of task
    • Level of focus
    • Personal characteristics
  • Gaze patterns users commonly exhibit on destination pages (like articles) and how to accommodate these behaviors:
    • F-pattern
    • Layer cake pattern
    • List bypassing pattern
    • Section bypassing pattern
    • Spotted pattern
    • Commitment pattern
    • Zigzag pattern
    • Lawn mower pattern
  • How people process search engine results pages (SERPs)
    • How much time and effort users spend considering results
    • Which results receive the most attention, and why
    • How SERP features (like the knowledge panel and featured snippet) influence information-seeking behavior
    • The implications of good abandonment (when users find their answers on the SERP, without clicking any links) for content creators
  • Gaze patterns commonly seen on search engine results pages and other routing pages like category listing pages:
    • Sequential pattern
    • Pinball pattern
    • Love-at-first-sight pattern
    • Layer cake pattern
  • Common scanning behaviors and why they occur:
    • Appraisal
    • Skipping
    • Backtracking
    • Regression
  • Encouraging reading through content formatting techniques
    • Designing headings and subheadings correctly
    • Highlighting key phrases to direct attention
    • Using information-bearing words in headings, links, and body copy
    • Attracting attention with lists and bullets
    • Using table layouts for easy data consumption
  • Writing styles that improve comprehension, as well as plain language guidelines
  • How to use content to build trust and credibility
  • Tips for writing complex, technical, or scientific content

Illustrations and Videos

The research findings are supplemented with 344 illustrations, which include the following:

  • Gaze plots (dots to indicate points where one person looks)
  • Heatmaps (color-coded visualizations of where many people look)
  • Screenshot examples
  • Data tables and charts

Additionally, the report includes a folder containing 14 supplemental gaze replays: video clips showing user gaze in real-time as they perform tasks in the eyetracking studies.


Eyetracking systems use specialized cameras to monitor a person's eye movements and pinpoint exactly where that person is looking, moment by moment. Based on many of these observations, we've identified regular patterns that describe how people move their eyes around different types of pages, in different situations.

The report also contains findings from the Search Meta-Analysis Project. In this project, we analyzed 471 different search queries from four usability-testing studies conducted between 2016 and 2018.

What’s New in the Second Edition?

  • New examples and analysis: Of the 344 illustrations in the report, 90% are new (310 new illustrations). We only kept 34 of the best examples from the first edition, which were too good (and still too accurate) to remove.
  • New gaze patterns: In a recent large-scale eyetracking study, we discovered the:
    • Pinball pattern on search engine results pages
    • Lawn mower pattern on tables and description pages
  • Chinese eyetracking examples: Our most recent round of eyetracking research included a phase conducted in Beijing, China. While some of our readers may be able to read Chinese as well as English, we provide extra description of the screenshots for those who cannot.
  • Expanded analysis of how people read SERPs: Search engine results pages (SERPs) play a huge role in how people find and consume content online. 
  • Video clips of eyetracking sessions: The report includes 14 gaze replay clips, which are discussed in the report. Read our analysis, and then play the video clip to see the user’s gaze in action.