Eye-tracking Study Shows How Users Scan Web Pages

apatrick writes "An article in UsabilityNews.com describes an experiment where Internet users' eyes were tracked while they searched for information on WWW pages from three well-known newspapers. The findings indicated that people learn very quickly where ads are usually placed on web pages, and then they no longer look there. The results also show that users look to the left hand side for navigation menus, and they scan from the middle of the page outward. Such results may be useful for developers wanting to make their pages more usable, or to attract the users' attention."

Comments

Redo

[Apreche]

I would love to see the study redone comparing users of say IE and Firebird. I think that users like myself who don't see any ads on websites could potentially have different reading patters. I often find myself looking in the top left or center of pages first to look for new headlines and to verify which site I am at. Of course that's when I'm conciously thinking about it and is not empirical data.

I'm also worried that studies like this may be used to put advertising in different more annoying places in more annoying ways making it harder to block and ignore.

And if you are wondering how to remove all ads in firebird check this out

http://www.texturizer.net/firebird/adblock.html

Menu Placement

[waldoj]

The information regarding menu placement is very interesting. As a developer, I've long been torn over the side on which the menu should go. UI testing on some of my client's sites has shown that people are more likely to look on the left-hand side,but I've also seen credible studies that keeping the menu on the right-hand side (near the scroll bar) is preferable, because it puts the menu near where the mouse will already be.

Now that a proper study has been done on the topic, I imagine that I should start moving menus over to the left-hand side of the page. It might be less efficient, but even crappy standards are still standards.

-Waldo Jaquith

Re:Menu Placement

[SandSpider]

On the other hand, if there's no menu on the left, it'll be easy enough for them to find it on the right hand side. Whereas if you put the menu on the left, then that's always going to pop over your content on a mobile device like a phone. So you should consider whether the menu or the content should hit first on a mobile device for any given application, especially if it's a site that is likely to be visited by mobile devices.

=Brian

Re:And this surprises you how??

[Anonymous Coward]

Well, indeed most webpages have menus at the left. But if there's a lot of text on the webpage, I prefer the menu on the right - that way, when reading left-to-right text, the end-of-line "flyback" is a bit quicker (for me, anyway) because I don't have to avoid the left menu.

Re:Applied Research

[idiotfromia]

I haven't seen any of those large boxes at slashdot in a long time since I have added the Adblock extension to Firebird.

Speaking of the scroll bar

[jtheory]

I was on a useability mailing list for awhile, and one of the curious tidbits that came out in research was that scrollbars themselves would have been better suited to the LEFT side of the screen.

Think about it -- most text is left justified; titles and headings start at the left (we read left-to-right, after all)... there's just more data over there if you're skimming. Technically, it would make more sense to have the navigation over there too... but again we're stuck with a standard that can't be changed now without a lot of pain.

Regarding your actual post... It think you have no idea how many users do not have scroll wheels, or don't use them. Sure, most people you know don't use their scrollbars, but believe me, the unwashed masses are using them all the time.

A company I worked for did this research.

[Anonymous Freak]

WebCriteria (now defunct, bought by another company,) did research on this along with Reed College back in 1999-2000. We had a much larger sample than three daily newspaper websites, and used all sorts of fancy tech to do it. (Cameras linked to computers doing eye-tracking, combined with screen capturing, to match exactly what they were looking at. They were allowed to surf whatever websites they wanted, but we tried to keep them viewing 'commercial' websites.)

We used this data to write a program that would analyze websites using this data. (So a website that had lots of side menus would be ranked lower, because people tended not to look at side menus that were too long.) Great idea, had lots of big-name customers (Intel, IBM, Microsoft, GM...) but in the end, trying to add extra services killed the company.