How does it compare with mouse tracking?

Mouse tracking is often used to check what parts of a website or product were seen as it is assumed to correlate with eye position.

Contrary to popular belief, mouse position only has a 22% correlation with gaze and there is no correlation between mouse and eye movements at all!

Our experiments show that the mouse cursor was within the field of fixation only 3.5% of the time.

The problem is that most people just use the mouse to interact with a website, for example by clicking on buttons or scrolling the page. It does not give any indication of whether non-interactive content such as advertisements or information such as product pricing has even been seen.

You might think just adding more participants will resolve this situation. However, after about 12 participants the informational content of mouse tracking saturates and will no longer improve with more participants. While the same is true with eye tracking, which already saturates at 6 participants, it provides about 8x more information compared to mouse tracking.