Welcome to WebmasterWorld Guest from 18.104.22.168
New EyeTracking Study verifies the importance of page position and rank in both Organic and PPC search results for visibility and click through in Google.
Nothing really shocking here, but the images which identify the important "triangle" are kind of neat.
"Eye scan and click through behavior changes dramatically as users moved “below the fold” to the section of results that required scrolling down. At the top of the page, the amount of eye movement declined rapidly through the top 4 or 5 results, and then at the bottom of the screen, tends to become more consistent through to the end of the page."
>dropping to 85% at the bottom of the “above the fold” listings
This differs from my experience. "Bottom of the “above the fold” would be SERP positions # 6-8 on most screens. At least in my subject area dropping from #1 to #6 means about 3x less clicks for that search term, compared to just 15% decrease according to this study.
What is your experience?
And this has been around and known for a long time.
If you scour some of Jacob Nielsens usability reports, in particular his research on "banner blindness", I think in late '96, or take a look at the "Macintosh Human Interface Guidelines" from Apple, you will find similar results and suggestions.
Maybe this time it gets picked up by the general media, make it into a webmaster fad, and every page gets chopped down to non-scrolling page. :) THAT would be funny.
Of course, we still have menus on the left side of view screens, and that's an even bigger usability issue, but who cares, right? It's not in the papers...
It seems that they are making a judgement on the quality of the search by what appears in the first few spots. So if you are not in the top few spots your CTR is affected not by the fold, or your general position but by the quality of the descriptions above you.
It would be a great follow up to this study to see how many unique searches are performed for a give topic by each subject, how many listings are followed before each adjustment to the search, and the average position of the listings that were selected.
It's no advantage to be in the top 5 if most of the clickers are just window shopping.
This is just another VooDoo Science survey to push higher bids from the PPC advertisers.
Now if they incorporate, the number of qualified customers and actual conversions to sales, this report would be worth something.
Summary: This study discusses the contributions of eye-tracking data to traditional usability test measures for first-time usage of websites. Participants viewed the homepages of three different websites. Results showed that eye-movement data supplemented what users verbally reported in their reactions to a site. In particular, the eye-tracking data revealed which aspects of the website received more visual attention and in what order they were viewed.
[edited by: tedster at 2:39 am (utc) on July 3, 2008]
[edit reason] update the link [/edit]
From what I can see, this scenario is horribly manufactured...
They're told to look at a search page specified by the survey company. So they're not actually searching for information, they're just doing what they're told.
If you were told to look at a page about something you're not interested in ("tibetan mountain goats" perhaps), would you look at the page in the same way as you would one that you are interested in ("google allegra" perhaps)? Would you scour the tibetan mountain goats page looking to see which is relevant? No, because nothis is relevant, you're not interested in tibetan mountain goats!
Beware of statistics based on manufactured scenarios.
So if users look at web pages in an "F" pattern
No users don't look at web pages in an "F" pattern. They look at google results pages in an "F" pattern. Change the page layout and users will look at it in a different way.
If each google page had a picture of a reclining Pamela Anderson in the bottom right hand corner users would look in an "-@@o" pattern.
Yep. One of the implications of this is that if you can get a page anywhere into the top 5, the key is to have a page title people will want to click on. #3 can be as good as #1 if the top 2 positions are occupied by pages with an unappealing page title.
The large circular graphic in the one page, and specifically at the face of that kid on the other site. Using graphics could also be a way to draw attention to the items you want people to notice.
The face thing really shows me something. The page after also had an image in the same spot but it had a lower attention factor probably because it was a less interesting image, there were no faces in it.
Not particularly. It's been my observation that for most SERPs that people don't go beyond page one. ALTHOUGH, the exceptions might be relevant for people reading here. For someone doing a simple informational search, the first site they find that answers the question tends to be adequate. However, if the person is searching looking to buy an expensive item, they are far more likely to dig down in the SERPs for the best price/value.
The eye-track studies I would like to see:
* if tracking/clicking differs when users are opening "tabs" instead of "windows".
* if tracking/clicking differs when apparent "spam" results list higher or are mixed in with "wanted" results.
* if tracking/clicking differs when a search term can mean (and return) conflicting results i.e. Washington (US state, US capitol, several persons, etc.).
* all of the above differentiated between novice, casual, and professional searchers.
Let's get real and practical and useful. This "release" appears to be a wonderful example of a PR exercise.
The total lack of attention to the right-side paid ads is pretty interesting, though not surprising. Obviously #1 organic ranking is 6 x more advantageous than #1 paid ranking! So get off your lazy butts paying for Clicks and work on getting in the SERPS again :-).
[edited by: MikeNoLastName at 8:25 pm (utc) on Mar. 2, 2005]
Of course I wasn't aiming for that spot but if I can't have the top 5 (not for these keywords, it would be insanely hard) its a good place to be.
People have scrollwheels these days any they aren't afraid to use them.
Page TWO however is an entirely different matter - most novice users won't go that far.
[edited by: amznVibe at 8:25 pm (utc) on Mar. 2, 2005]
There's a reason why the term "above the fold" is used. It has been well-known for almost a century that there are particular areas of any publication (like a folded newspaper ... with the area "above the fold" on the top issue being what people see when they walk past a stack of them on the street) that people look at more often or for a longer duration than other areas. We've known about the top-left corner for many, many years. The "F" shape seems incidental: there is a second "virtual" page being formed by the PPC ads on the right, and experienced surfers know it and respond accordingly, either by ignoring that area or by giving it a cursory glance.
I agree that using Google as a control is a poor choice, as there is nothing visually different about any result, whether for toys or for information, so there is no incentive/motivation to look anywhere other than where the people in the study looked.
This (admittedly unfinished) study would do well to evaluate factors like site familiarity, surf purpose and user comfort in their final results.
The other studies seem to bear this out: different sites, different goals and different levels of experience all skew the data differently. They do seem to imply, however, that (a) a big, round graphic grabs the eye first, then (b) the eye spends more time on text, even in an ad where the picture might be looked at for a very brief period. (Sorry, Flash/heavy-graphic users! :)
I seem to remember (from a long time ago) that right-handed people scan from left-top, where left-handed people scan from right-top. If this is borne out in the data, all of you who run Leftoriums should run out and bid-up your PPC ads ... but don't go over position #3, or you'll end up as a banner ... not where your customers are likely to look first.