Google has published a bit of an insider’s look on how the company conducts eye-tracking studies to evaluate the effectiveness of its search results.
In addition to holding interviews, field studies and live experiments to improve the usability of its products, Google has special hardware and software that tracks test participants’ eyeballs as they scan results for the perfect link.
The official blog post doesn’t detail any groundbreaking discoveries that have been produced by this testing technique. It sounds as though it has mostly helped Google confirm the obvious: that the first few results it returns are indeed usually the most relevant, and its so-called “universal search” effort (where it mixes rich media results like images and video thumbnails among the standard text results) doesn’t distract users too much but has actually proven rather useful.
Perhaps most intriguing is the following video provided by Google that shows how quickly users glance around result pages:
The bigger the dot, the longer the person sat looking at a particular part of the page.
This heatmap-like image, which is named the “golden triangle”, also suggests that people spend a lot more time evaluating the whole results at the top of the page than the ones further down.
For more, see our previous coverage of Google’s usability lab.
What’s interesting is that you don’t see any ads in the right hand column. I’m curious how many people actually notice or read the ads…
Yeah it gives you maybe an idea of how ineffective those ads are.
I think it would be great if a firm offered this sort of user-eye testing for websites that wanted data on where thier uses look. That could be a pretty cool business.
Eye tracking hardware and software is commercially available for anyone to buy. It’s becoming fairly standard in usability testing labs.
This is done with Tobii eyetracking equipment.
We have been providing this sort of eye traking testing and heat mapping for a number of years. It is very useful in optimising the performance of web sites, intranets and so on. Check out our full User Experience offering here http://www.hyro...es/default.aspx
Big evil G’s business model is based on PPC not CPM, so how you derive the statement of how search Ads are not effective is beyond me.
Look at the effectiveness of those ads and you’ll quickly understand that big G is doing its money due to its size of exposure, and size alone.
maybe G does not want to spill the beans on its not so perfect secret sauce.
Thanks for sharing info.
Good study for who is designing the user interfaces…
Reading from left to right isn’t anything new, but this article is enlightening nonetheless. I think that over time and a natural evolution that google will find better and more efficient ways to calculate what actually makes it to the top of the searches… or so I hope.
@Shawn: I don’t think they want to tell their advertisers that people don’t scan ads. It wouldn’t be good for business. Better to publish the research with an ad-free page.
It’s pay per click, not glance, so I doubt it would matter? And may as well eliminate confounding variables. Perhaps they will move on to this with ads, but it’s far more likely they already have and just don’t release those.
I wonder how much of that time at the top is just time spend adjusting to the layout of the page… Probably fractions of a second, but still something to ponder…
Why doesn’t Google put the sponsors to the left of the results page? This might be drastic, but more people would be buying sponsor spots, most likely.
Or at least I would.
Because then that would really irritate the users and Google is nothing without its users..
Even if they moved the ads to the left, how long do you think it would take for users to overlook them?
It’s pretty much like reading a book (add to the fact that google itself looks like a book). People are used to reading from the top left, especially if the most accurate results are there.
These findings don’t surprise me really. But it’s interesting regardless.
These studies all end up showing similar patterns where the eye starts near the upper left side moves right and/or down, etc. When will we see real uses of eye tracking to determine how to make the entire page valuable? Or does this simply validate the twitter philosophy that the short attention span is all that matters. In other words, if you created something that attracted the eye to the far reaches of the page (of the galaxy???) then would users be overwhelmed and never come back?
I’m not a usability consultant, but I’ve worked with them. My experience on projects where I’ve seen tracking studies is that task success is lower on pages that place information outside the “golden triangle” because people literally can’t find the information within a reasonable timeframe and give up.
That said, consider where Facebook stuck its notification icons — in the lower right, well out of the golden triangle. So it’s not an exact science (though I wonder how well those are working).
Anyone find out what a Spanish Water Dog is?
nvm. appears to be some brand of mop
It’s a breed of dog you dope.
@michael rice, if google put sponsors to the left of the page, it might increase conversions for a while but people would learn to overlook it, the same way we night overlook banners.
There are sponsored links at the top of the page in a same as to the right, usually.
What’s interesting about the video is that it shows the obvious truth: Readers are skimmers. We’re looking for keywords to then click on the site, which just reinforces the idea that your meta tag descriptions need to be on point.
This is hardly new… Google might call the heat map the golden triangle, but Jakob Nielsen has known about the F-shaped pattern in eye scanning patterns for a very long time…
http://www.usei...ng_pattern.html
I agree that this isn’t anything new. While studying Direct Marketing back in 1999 at NYU I learned about the Johnson Box – the first few sentences at the top of a Direct Marketing Letter…despite this…it’sstill a really powerful image or “heat map”…insight + something cool = people listen.
This is great.
I think if Google started putting sponsorships down the left side, then users would be re-trained to automatically first look down the RIGHT side. People know how Google works, although this study doesn’t specify the user level of the participants, does it? Is there actually a first time Google user?
The same goes with banner ads — user eyes train to look away from the “flash” and placement of the ads after becoming mid level users. They increasingly become less effective the more experienced web user.
Still Google Adsense generates revenue.
However it seems more important to spend your money on getting high in Search results (SERP).
This comes down to lots of content and blogging (as blogs get higher results)
As some has said, eye-tracking is nothing new. Yahoo! does (or did) the same thing with their Home Page – but nobody cares.
Discoverability (sp?) is one thing. *What* you click (user intend) is another issue.
Eye-tacking is just one of the many tricks in usability study….
Just because a site is shown high in the results does NOT mean it is the most relevant.
Nor does the fact that people are focusing mainly on the first results vindicating Google’s selections for relevance.
The way Google currently handles its index, most of the MOST relevant content for competitive queries never sees the first page of Google’s search results any way.
Eye-tracking is nothing new. Academia has been doing this at universities for years; I know many a grad student in cognitive psychology who as done research and theses on said subject.
google _does_ have ads in the golden triangle. ad layouts are randomized and there are occasional top-ads. i think i’ve seen left-ads but i’m not certain.
Eye lingering doesn’t indicate a successful ‘find’. It equally indicates an ambiguous, irrelevant and frustrating inclusion in the results.
So, eye tracking doesn’t evaluate effectiveness of search results. It just evaluates people noticing what you want them to see.
Well,another type “interesthing technologhy”this tools has proven how effective is “research in marketing industry” I think this tools is great for google to assure there client at adwords that “Bidding Bigger is Better in marketing spending”because the higher it is ,the better it will.hope thats not to much words to say
The innovation just keep on going. Google should rebrand itself to some type of “technology inventor” they just keep on coming up with cutting edge results.
so whats new! These studies have been around since 2001
Why did Google publish these results?
The result of this testing shows we do no need a search engine then.
We can collect all queries possible (even we can give people a hint what to search) and then display few items that we think are the best. BOOM! the search engine job is over.
We all know the best search results can not possibly be in the top two, if this is the case then probably we need a better bookmarking technology rather than a search engine. A search engine should distance itself from human behavior otherwise it becomes limited to what humans are capable to do when it comes to analyzing peta bytes of data.
Google tracks everything,how a user searches and how user views the search results.
Even though we have all seen these results before it is very cool to actually see the eyes actually tracking on Googles page.
Woopty fuckin’ doo. People view the top left of the page more than anywhere else. Who would have guessed?
It’s crazy how much data they have and analysis they do. In this case, much of their effort is actually for the good….to improve the user experience.
(oh yeah, and to figure out where to place those ads so they can make more money……)
This article is old wine in new bottle. The UI / UX community has studied and known this for ages. As far as eye-tracking hardware and software is concerned, even my school’s UI lab had it and it is a fairly common commericially available setup.
Im tellin ya dude, one day Google is going to rule the world!
RT
http://www.anon...ity-tools.us.tc
Uhhh, this is nothing new… I work as a usability engineer at Microsoft and we have been using this since 2001…
i noticed the logged in google user in the screenshot was Anne Aula…
http://anne.aul...curriculumvitae
Eye tracking studies was noticed long back, I feel its not new discovery
Uhhh, this is nothing new… I work as a usability engineer at Microsoft and we have been using this since 2001….
Interesting. I think Google got something up their sleeve, maybe a secret coming out of a project they’ve been rendering…Google is like Japan, they’re going to secretly take over hahah jk
but interesting post though
OMG! Google is really doing great in innovating their search engine. Unlike Yahoo, their search engine is really the best. Yahoo is a little choosy in crawling a site for any matter. They limit their robots on sites that they like which is I don’t get the point.