New Eyetracking Study: How do visitors really see your website?
Category: Design...
Posted Friday, September 10, 2004 at 10:07 AM [GMT-6]

CyberJournalist.net has a good summary of the results of the Eyetrack III study designed to measure what visitors actually look at when they come to a website.

Eyetracking is a relatively new technique of recording where on the screen a subject's eyes are focused at any given time. According to someone who admittedly makes a living selling the service, eyetracking is accurate to within 1 centimeter, which is probably close enough for the types of conclusions drawn from the studies that utilize it.

If you have any interest in website design (and all bloggers should fall into this category, by the way), you should find the results of this study quite interesting. It's not a "deep" study -- only 46 subjects were tested -- but it covers a wide range of variables.

I found the following to be the most thought-provoking of the conclusions (my observations are in parentheses):

• The eyes most often fixated first in the upper left of the page, then hovered in that area before going left to right. Only after perusing the top portion of the page for some time did their eyes explore further down the page. (This is not surprising, given that the test subjects were visiting English language pages and that the universal standard for such pages is to place the identity of the page at the top, and usually top left. Conditioning drives behavior.)


• Most people looked at text, not images, first. Dominant headlines -- not photographs -- most often draw the eye first upon entering the page. (This provides some vindication for those designers who insist that content and not eye candy are what bring people to websites. Nevertheless, I'm surprised that the eye isn't naturally drawn initially to images, even if only to quickly discard them as irrelevant.)


• Navigation placed at the top of a homepage performed best. (This is where I take small exception to the measurement standard: "eye fixations and viewing duration," with more/longer being judged better. I might draw just the opposite conclusion. If someone is fixated on your navigation scheme, it's non-intuitive and faulty. The nav tools should be easy to find, but otherwise they should be operable on an almost unconscious basis.)


• Smaller type encourages focused viewing behavior (that is, reading the words), while larger type promotes lighter scanning. (Here again, I'm not sure that this is a good thing. Your goal as a designer is to provide the visitor with the information she seeks in the shortest time possible. If she can get that information via "light scanning" vs. reading every word, then I think that's a good thing. The fact that the smaller type meant spending more time focused on the text seems to indicate that comprehension was hard to come by. I think that's a bad thing. [Writers who cherish their every word will no doubt disagree. You know who you are.])


• When people look at blurbs under headlines on news homepages, they often only look at the left one-third of the blurb. In other words, most people just look at the first couple of words -- and only read on if they are engaged by those words. On average, a headline has less than a second of a site visitor's attention. (This seems intuitive, and I know that Dawn Eden, headline writer extraordinaire, will confirm it. The fascinating aspect of this finding is that even when we're presented with only a half dozen words or so, we'll still try to get by with reading only one-third of them in order to save time. We are busy, busy people.)

There's a lot more along these lines in this study. Read the whole thing...or at least scan it. ;-)