Skip to content
InsideUX Design

UX glossary · engl. Eye Tracking

Eye Tracking

Eye tracking measures where people look on a screen, for how long and in what order. It shows you what gets noticed and what gets missed.

Imagine a product page where the most important note, “Shipping within Germany only,” sits in a box to the right of the text. Yet requests from Austria keep coming in. Did people read the note and ignore it? Or did they never look at it?

That’s exactly the kind of question eye tracking can answer.

Explained simply

Eye tracking measures where people look on a screen, for how long and in what order. A special device records eye movements while someone uses a website or app.

Think of it this way: if you look over someone’s shoulder while they read a newspaper, you only see which page is open. Eye tracking is as if you could also see a laser pointer showing exactly which line they’re reading and which ones they skip.

How eye tracking works

Most devices use infrared light, which the eye can’t see. It creates reflections on the pupil and cornea. A camera captures these reflections, and software calculates the point of gaze from them. Before each session, the device is calibrated for each person by having them look at a few points on the screen.

The analysis focuses mainly on fixations, short moments when the gaze rests on one spot, and saccades, the quick jumps in between. These produce gaze plots, which show the gaze path of one person, and heatmaps, which combine many people.

Eye tracking became well known in UX through the studies of the Nielsen Norman Group. In 2006, based on recordings of 232 people, Jakob Nielsen described the so-called F-pattern: many people first read a line across the top, then a shorter line further down, and then scan down the left edge. Later analyses show that this pattern mostly appears with unstructured walls of text. Good subheadings and short paragraphs guide the eye better. Learn more in How to Structure Content the Right Way.

Costs and limitations

Eye tracking takes effort. You need hardware, software, a quiet setup and experience in analysis. For meaningful heatmaps, the Nielsen Norman Group recommends around 30 people per heatmap, far more than for a classic usability test.

There are methodological limits, too. Eye tracking shows where someone looks, but not whether they understood something or why they missed it. Thinking aloud interferes with the measurement, because people linger on one spot while they talk. Glasses or fidgeting can make calibration harder. For most questions, a simple test with observation is perfectly enough.

An example

Here’s an example: for the product page above, the gaze data shows that hardly anyone looks at the box on the right. It looks like an ad and gets skipped. The team moves the shipping note directly below the price, formats it as regular text and starts it with the word “Shipping.” Requests from abroad go down.

In practice

  • Use it deliberately. Eye tracking pays off mainly for specific questions about perception, such as whether a note gets seen or how people scan a homepage.
  • Pair it with questions. A short conversation after the task reveals why someone missed something or looked at it for a long time.
  • Apply the known patterns. Even without your own lab, you can learn from the findings: put important things at the top and left, and start headings and list items with meaningful words.

Related terms

Heatmap, Usability Testing, Cognitive Load, Information Architecture

Sources

More terms

Full glossary