Initial takeway findings from eye tracking

Peeper by CraigPJ on http://www.sxc.huWe’ve been carrying out eye trackingWikipedia Logo testing on the new Channel 4 pages (homepage, TV Listings page, Programmes page, and show page) to see where people look and to inform decisions on where we position elements on the page. We haven’t got the final results yet, which come in the form of heat maps and scan path overlays (the path where the eyes look is overlaid on the webpage), but here are some of the initial things I’ve observed.

People are naturally drawn to look at faces

Faces are a huge draw for the eyes. Images of people are the first things that people generally look at, and in particular their faces. The takeaway is that if you want people to notice something use an image that features a persons face. This was strongly emphasised when adverts with faces were shown on a page. In this case the eyes tended to look more often and quicker on these adverts.

People will look at elements on a page but may not see them

People will look at things (as proved by the eye tracking software), at least their eyes will look at them, but they don’t seem to conscious of the fact that they’ve looked at it. When asked to complete a task users will have seen the area but will still need to take time to hunt for it.

Advertising kills the right hand column of a webpage

The right hand column isn’t as invisible as we thought. People do look there, but unless the content is interesting to them they don’t take any more notice of it. People will notice an animated advert, but will then block it out. Content in the right hand column that is a similar style or colour as an advert will be ignored. However….

Advertising recall seeme to be age biased

“Older” users (40+) tend to look at adverts more than younger users, who have advert blindness and ignore the right hand column. The recall of adverts is greatly increased if the advert has a familiar face in it (like the face of a celebrity). This is also reflected on the time users in this age group will spend on a page. Younger users will very quickly scan and click, rather than looking at the page content in detail, which older users seem to do.

Most people look down at their keyboard to type

Our new predictive search mechanism has been very well received, but most people look down to type, then look back up and see the predictive drop down and then click on it. Some will look down to type and click return, without ever seeing the predictive search drop down.

A different look and feel is ok for different content

When comparing an entertainment page (which is in the new look and feel, for example Shameless) and a factual page (in the older look and feel, for example Grand Designs) users are happy with the difference. They say that the Shameless page is nicely focused on clips, pictures and catch-up; whereas the Grand Designs page is far more information rich, with links to similar shows.

I’ve made no comment on these inital findings nor have I tried to qualify them – these are merely my initial observations. Overall the new Channel 4 site was extremely well received and people completed all tasks that were put to them quickly and easily – indeed, changes that we made after our last round of user testing made a significant improvement to task completion on some pages.


I’ve posted some of our results and a little more detail about the methodology that we used in Mixing eye tracking and qualitative user testing.

7 thoughts on “Initial takeway findings from eye tracking

Leave a Reply