Usability Study: Users wrongly trust higher Google results

[encyclo]

An interesting article about the usability of Google search results: Users wrongly trust higher Google results [upi.com]

A U.S. study suggests Internet users mistakenly have an inherent trust of Google search results that appear higher on a page.

A College of Charleston eye tracking experiment revealed college students participating in the study trusted Google's ability to rank results by their true relevance to the query. When participants selected a link from Google's result pages, their decisions were strongly biased towards links higher in position, even if content was less relevant to the search query

[JS_Harris]

The top site is usually not the best and this article doesn't cover the "pagerank" effect which is more significant to new webmasters. IE: this site isn't pagerank 6 or more so it sucks, next!

[Jbrookins]

There is no PageRank effect amongst average users. The average user doesn't have any clue what PageRank even is.

[Korrd]

I sometimes find myself riding shotgun, looking over the shoulder of college educated people as they run searches, and the consensus I see is most people view taking time to peruse SERPs as inefficient, preferring instead to immediately click something, usually the top result. Combined with their search phrasing it's amazing they ever find anything close to the information they are seeking.

[JS_Harris]

jBrookins perhaps your right depending on the topic but I assure you that pagerank plays a big role on many types of sites.

Actually its more like "pagerank" + "alexa rank" + "technorati authority" for some topics. Worse is that some text link services use a combination of those three to rank a site as well and that determines how much income the site receives from that service.

The average webmaster pays no attention to them all, which is fine if you're not monetizing the site.

[europeforvisitors]

The top site is usually not the best and this article doesn't cover the "pagerank" effect which is more significant to new webmasters. IE: this site isn't pagerank 6 or more so it sucks, next!

The study isn't about Webmasters. It--and this thread--are about user behavior.

[annej]

The challenge is to look at this data and consider what can be done about it. The first thing that comes to mind is good titles and meta descriptions. Especially titles.

Would you rather look at a page that is titled "The Mysterious Inner Workings of Widgets" or one titled "Widgets, Buy Widgets, Widgets and more Widgets"?

Beyond that we are at the mercy of the way people search. I think people are improving how they search thus the greater value of the long tail from more specific searches.

Also with more specific key words and phrases you have a better chance of being above the fold on the first page of results.

[Whitey]

With Google introducing images and maps into the first lines of results, together with visual icons like stars , perhaps we could take this discussion further into the common nature of general web usability and Google's results, and how it can interefere with "trust".

IMO a visual respresentation is far stronger than text, provided it is relevant and occupies the the upper left side of the screen.

[poynterextra.org...]

[europeforvisitors]

IMO a visual respresentation is far stronger than text, provided it is relevant and occupies the the upper left side of the screen.

Sure, and that's great if the user is looking for "Elbonia photos" or "Widgetville maps." But if the person is looking for the history of Elbonia or a Widgetco camera review, I suspect he'll go for the text results--and in many (most?) cases, text results will be the only search results on the page.

[Wlauzon]

IMO a visual respresentation is far stronger than text, provided it is relevant and occupies the the upper left side of the screen.

I used to believe that, but with the proven fact that users will pretty much totally ignore and tune out any graphic that looks like an ad, you have to be pretty selective there in size and placement.

Looking at the eye scan patterns, it appears that the best way is to have a short, relevant, and descriptive headline right up at the upper left, as that seems to be where most people start looking.

[Whitey]

It's a bit of topic, but on our web pages we seek to draw the user to action amongst text. The action is driven by icons.

Ideally we'd require no text. The bottom line is these studies show that people are time poor, trained in scanning documents and will respond to visual prompts rather than large quantities of text - the majority of which is not read.

I've no doubt Google is constantly testing the appeal of icons versus text results in different categories. Many markets which will behave differently. But of course without text qualification those results will not appear.

[trinorthlighting]

I can see where users would trust the results. Most of the results are trustworthy, but there are always a few that sneak in the top 10.

[rogerd]

To some degree, Google has brought this on itself by delivering really good results most of the time. I recall when you could search for a company name, for example, and have only an outside chance that the official company site would be on the first results page. Google fixed that, actually delivers what people are looking for most of the time. This trains people to trust the top result, or the top few results, and to not "waste time" reading what they are clicking on.

[techrealm]

Now, if there was only a way to capitalize on obvious but consistent wrong company phone calls and make some money off them... hmmm...

[gabrielk]

Here's the abstract from the original paper on which that article was written:

An eye tracking experiment revealed that college student users have substantial trust in Google's ability to rank results by their true relevance to the query. When the participants selected a link to follow from Google's result pages, their decisions were strongly biased towards links higher in position even if the abstracts themselves were less relevant. While the participants reacted to artificially reduced retrieval quality by greater scrutiny, they failed to achieve the same success rate. This demonstrated trust in Google has implications for the search engine's tremendous potential influence on culture, society, and user traffic on the Web.

And here's a link to the original publication: [jcmc.indiana.edu...]

[Whitey]

I think it's fairly academic talking about "trust". User behaviour not only follows these eye studies, but also metrics such as this one. It was for click rank user behaviour comparisons [ which is a key metric ] which was published a while ago :

Searches: 9,038,794

Total Clicks: 4,926,623 54.5%

Click Rank1: 2,075,765 23.0%
Click Rank2: 586,100 6.5%
Click Rank3: 418,643 4.6%
Click Rank4: 298,532 3.3%
Click Rank5: 242,169 2.7%
Click Rank6: 199,541 2.2%
Click Rank7: 168,080 1.9%
Click Rank8: 148,489 1.6%
Click Rank9: 140,356 1.6%
Click Rank10: 147,551 1.6%

Conversion ratios for e - commerce sites may also experience higher comparable rates. Then there are marketing interferences with the better use of compelling meta descriptions and titles.

Informational sites , may adhere more greatly to relevance with users prepared to trawl for good answers - but e-commerce is a little less about trust and more about consumers being time poor or lazy, provided the information they are served up appears relevant.

It's a different story when they enter the site where poor usability , irrelevancy and other factors may cause the user to return to the SERP's for the next result.