At the end of last week, some discussion on Hacker News started after a post claimed that Authorship created a 90% drop in traffic. Later, Matt Cutts jumped in to note that it was Penguin and not Authorship that caused the drop.
I thought it would be useful to talk about how social annotation is changing user behavior.
For a While, Social Annotations Were Ignored.
When social annotations first launched, we all assumed that it would have some type of impact on CTR, but for the most part, users ignored the new images in the search results.
Google published a paper in early 2012 showing that the majority of search users simply did not fixate on social annotations in search results.
They had some interesting conclusions.
“Our ﬁrst study yielded two unexpected results. First, in some contexts, social annotations shown on search result pages can be useless to searchers.” – Google
These studies showed that users simply did not notice the annotations.
“The more counterintuitive result from our ﬁrst study was that subjects did not notice social annotations.” – Google
However, this appeared to be the result of “inattentional blindness” due to how users process search results. In part, where the annotation appears in a search snippet has an impact on if a user fixates on the face. In addition, there is a behavioral component. Even if the user notices, will they use this information in their mental model for how they evaluate and determine which result to click?
You can see about how effective annotations were in early 2012 by looking at the data from the paper.
The data shows that the layout of the annotations plays a significant role. For example, there are a number of different ways the annotation could be shown.
The results show that users typically process in order of priority: (1) the snippet text, (2) the URL, (3), the title, and (4) the annotation text. And overall, the larger the snippet text, the less fixation there is on the annotation text.
As expected, big images work better than smaller images, as 50×50 images won over 21×21 images across the board. In short, big pictures of faces get noticed and small images are relatively ignored. And as expected, having the image above the snippet outperformed placing the image below the snippet. The two cases where this has less of an impact was on 4 line snippets and 1st results.
According to the paper, users have learned habits for how to parse search results, which don’t account for these new pieces of information. Google can, using design, help improve the inclusion of this data, making it more useful for searchers.
User Behavior is Changing
Through work by Google, and potential changed behavior from users, the way search results with annotations are parsed is changing. They’re receiving significantly more fixation from the users in early 2013, as seen in a recent paper from Google at WWW 2013.
Google is helping to shape how users use the results.
“We are at the cusp of teaching users how to parse social search results by designing where and how the annotations will appear.” -Google
And Google believes user behavior will continue to change.
“quite likely that users’ behavior will continue to evolve” -Google
Their results now show evidence that social annotations now receive strong attention from users.
“we found that users had a 60% chance of ﬁxating on the annotation when placed at the top of the snippet block”
Their research showed some really interesting data. Here is some of that data, showing the social annotation funnel. They measured how many results SHOULD have personal results, how many did or did not have personal results, how frequently those personal results came into view, and how many of those were noticed by users.
There are a few interesting stats along that funnel. They expected 53%, more than half, as results that SHOULD be personalized. Of those that they expected, 80% actually had the annotations. Of those, 66% ended up in view, which means we could assume at least 35% of them appear “below the fold”. Of the ones that appear on a user’s screen, they took notice of 60% of them.
Understanding How Users Scan Results
Years ago, eye tracking results showed what is typically called the Golden Triangle on SERPs. It shows that users typically fixate on the upper left side of results and they scan down the results in a predictable fashion. This leads to the CTR decay curve we’re all use to seeing. Effectively, eyes focus on positions 1-3 and drop off significantly as they move down the page.
This is no longer the case.
Here is how users are processing results today with social annotations.
As you can see, the first result still pulls attention, but the annotation is pulling the user far down the page. It’s position 4 or 5 (depending on how you count) and it’s receiving significantly more fixation from the user than any other result. You can also see the click icon, showing the user ultimately picked this result.
More surprising is how users react when there are also video snippets on the page.
Again, the social annotation is pulling a user’s eye down the page. In this case, the vertical distance from the top of the page to the annotation is significant. It still draws more attention than both highly visual video search results.
They also shared some data on the exact order and intensity of attention given to page elements before clicking a result. It’s really enlightening. The numbers show the order and the size of the circle shows the intensity (length of time) of the fixation.
Differences by Vertical
There was also some enlightening data about when users find social annotations valuable. Users do not treat them the same across all query types.
Chance of Fixating on Annotation by Query Type
- Navigation: 100%
- Local services: 100%
- How-to: 67%
- Shopping: 64%
- Entertainment: 56%
- Fact-finding: 40%
- News: 14%
And if you’re looking for some more clever ways to optimize your face, check out OKCupid’s My Best Face. This could also be useful in persona targeting and outreach.
It’ll run options past their users and give you personality data on those who selected your picture.
Overall though, competing in SERPs is getting more interesting as Google continues to provide users new, and potentially better, ways to understand and select data from the search results beyond simple rankings.