Nielsen reports that 42% of users click the #1 SERP result Options

[Black_Knight]

Jakob Nielsen has some interesting observations regarding a recent study of how users select links listings in SERPs (Search Engine Results Pages)
http://www.useit.com/alertbox/defaults.html

42% of users studied selected the topmost search listing from the results. This was far and away the most-clicked single link position in the study. However, it is still interesting to note that the majority (58 percent) of all users studied did not select the top search result.

8% selected the second link in the SERPs, and the percentage went down in the lower positions. That shows that only one fifth as many clicks were taken by the second listing in the SERPs as for the top listing.

What's interesting to me is that when they secretly employed a script to reverse the positions of the top two listings in SERPs, the percentages were not unaffected. Now the link in first place still recieved more clicks than any other single position, but it had dropped significantly from 42% to just 34%, a drop of 8% of total clicks. The second listing (the one the SERPs ahd really rated top before the script reversed them) now got 12% of all clicks (a 50% increase on the 8% it had got before the reversal).

I think this is showing that there are of course a percentage of users who simply click the top result, but that these are not a majority. From this study it would seem that a third of people were clicking the topmost result even when it was not the one the engine had truly placed top. The other two-thirds of users were obviously applying judgement on which listing to select, and not all were even staying within the top three results.

Take a look at the article, and indeed the study, for yourselves and have a think about the issues it raises. Were the findings of the study what you expected? What further information would you have liked to see researched in such a study?

[sonjay]

These results aren't surprising. It's why everyone wants to be #1 in Google.

What I'd really like to see is, are these users clicking the top organic result, or the top-top result, even when the top 3 are paid Google ads. I.e., do users distinguish between paid results vs organic, and to the extent that they do, do they have a preference for one over the other?

I'd also like to see some qualitative research into what causes some people to click #2, #8, or what-have-you rather than blindly clicking the top link. Do those people fit into some particular demographic groups? Is it more a function of the titles and descriptions that appears in the SERPs? Is something else at work?

[cre8pc]

The first thought that came to my mind was how I make my own choices. I don't automatically choose the first search result, often because I don't like the title or description. I'm looking for how their content relates to my need at the moment, not whether or not Google (or any search engine) determines its relevancy.

Am I unusual?

The study didn't create user personas, so user behavior and ways in which we categorize people (browsers, transactors, accidental tourist, etc.) aren't taken into account.

Categorizing habits isn't factored in either. Therefore, a methodical person may choose differently than someone who glosses over data and chooses on impulse rather than specific data.

From the study:

Second, there is a
“quality bias”: the users’ clicking decision is not only influenced
by the relevance of the clicked link, but also by the
overall quality of the other abstracts in the ranking. This
shows that clicks have to be interpreted relative to the order
of presentation and relative to the other abstracts.

This reminds me of what Mike Grehan had written about months ago where he had written about connections and relationships between sites being a factor as much as link popularity and on page optimization. He was thinking in terms of algorithms but I had churned for weeks in my head about how this related to search end user behavior as well.

Search results can show or validate whether or not someone made the correct query. This can be determined by the results themselves and how or if they

a - relate to the query
b - satisfy the query
c - meet subtle requirements of the query
d - meet the needs of the person who made the query (for example, if they like sentences that make sense vs. descriptions that pull key words and toss them together like a salad

If the SERPS don't meet a - d, a change in the query likely follows by the end user. Whether or not the first search result was the "best" choice isn't or may not be, the decision maker.

From the study:

While the linear ordering of the results suggest reading
from top to bottom, it is not clear whether users actually
behave this way. Figure 2 depicts the instance of first arrival
to each abstract in the ranking. The arrival time is measured
by fixations; i.e., at what fixation did a searcher first view
the nth-ranked abstract. The graph indicates that on average
users tend to read the results from top to bottom.  

In addition, the graph shows interesting patterns. First, individuals
tend to view the first and second-ranked results right
away, within the second or third fixation, and there is a big
gap before viewing the third-ranked abstract. Second, the
page break also manifests itself in this graph, as the instance
of arrival to results seven through ten is much higher than
the other six. It appears that users first scan the viewable
results quite thoroughly before resorting to scrolling.

This was left out of Jakob's roundup, but could be useful to SEO/SEM's who are fighting for their client's number one spot.

For example, if they don't get that number one spot, the world will not end :wink:

I have a nagging feeling that SEO's will take the report to mean they have to find more ways to get the number one spot, whereas in my mind, the lessons are more interesting for the search engines themselves, in that they must do more than factor in keywords and on page data, or simply place purchased spots at the top. They need to consider the human doing the searching and their needs, search requirements and Internet use behavior.

I consider this a first-pass study and more could be added and tested to get into the behavorial aspects and responses to SERPS. Age is a consideration for example. I'd even go so far as to say that how well someone reads (fast/slow, comprehension, dyslexia, etc.) would be interesting test subjects.

[Brad]

It seems to dovetail in with my observations on some of my directories: the top few listings on the first page of each category get the lion's share of clicks.

[projectphp]

I consider this a first-pass study and more could be added and tested to get into the behavorial aspects and responses to SERPS

I agree! Some questions this instantly raised for me:

- the 8% of people that changed when the results are switched, these people need researching., i,e, what differentiates them, demographically, in terms of expertise and level of internet competency etc etc. Some possible hypothesises: more established users trusted themselves more than the SE.

- what about result two encouraged users to not click it as much as result one when they were swapped? What can we learn from this? For example, where better titles important, or better snippets?

- What effect does this have on conversions? Do the people that click result one blindly become customers as often as those that do not? Do these users have different needs both at the SERP and on a site?

[randfish]

I wish I had the funding to conduct some serious testing of my own in these arenas. There's so much data in the search world that we don't have access to; it's unbearably frustrating at times trying to make sense of things and worse, trying to make clients understand such a new field (with no reliable data).

Maybe I should start up SEOmoz testing laboratories...

[swainzy]

The first thought that came to my mind was how I make my own choices. I don't automatically choose the first search result, often because I don't like the title or description. I'm looking for how their content relates to my need at the moment, not whether or not Google (or any search engine) determines its relevancy.  
 
Am I unusual?

Not as far as I'm concerned.
I look at the url to tell me if the site has relevancy to me.

For example: If I am looking for information on workers comp, I click on
say - workerscomp.gov as opposed to aspecificinsurancebusiness.com. Depends on what I am type of info I seek, of course. I read the descriptions before clicking and I almost never click on sponsored links.

The article states:

This study goes far to address why users tend to click on the top hit. There are two plausible explanations:

-Search engines are so good at judging relevancy that they almost always place the best hit on top.  
-Users click the top hit not because it's any better, but simply because it's first. This might be due to sheer laziness (after all, you start from the top) or because users assume the search engine places the best hit on top, whether that's actually true or not.

I would say that this type of searching behavior is in correlation to how much experience people have using a search engine. Seasoned searchers have figured out which type of link will most likely be relevant for them.

D

[bragadocchio]

A recent article published on the Public Library of Science in their Medicine section a couple of weeks back points to some of the failures in research within medical studies, and in the sciences overall.

I think that it can be applied to a much wider range than just medicine:

Why Most Published Research Findings Are False

It's a great paper, and I think that it pinpoints that we really need to think critically when reading a study, even if it is peer reviewed, and published by folks who are experts in their fields.

It bothered me that Dr. Nielsen's essay focused on one aspect of the study, and ignored the actual focus of the paper to draw some conclusions that I don't necessarily agree with. For instance, in this section of his article, he states:

What I'd really like to see is, are these users clicking the top organic result, or the top-top result, even when the top 3 are paid Google ads. I.e., do users distinguish between paid results vs organic, and to the extent that they do, do they have a preference for one over the other?

I don't know if they filtered out the paid results the first time through, but they took them out of the second set of results.

As for that second time, there's nothing in there that says that people used the same queries to find their results. I'm not sure that I would. I don't see how they could have effectively performed this part of the study based upon the methods used.

I think I would agree with Kim that this is a first pass study, and that there are many other things that could, and possibly should be tested. It is an interesting idea - studying how people may react to search results.

One statement that I liked in the study was the notion that search results aren't always chosen based upon some absolute relevance, but rather a comparative relevance based upon some of the other results. That seems to go against Dr. Nielsen's thesis in his article.

[subnet_rx]

It would seem to me that some of the smaller SE's would want to sell or publish this information for PR.