Not content with using search queries and click tracking to determine your interests (along with a wealth of other factors), Google is now looking into bring information about the position and activity of your mouse cursor on screen into their calculations, even when no clicks have been performed.
In a new Google patent (filed back in Feb 05 but only just approved) with the catchy title “System and method for modulating search relevancy using pointer activity monitoring“, Taher H Haveliwala (author of the Topic Sensitive PageRank paper whilst at Stanford) outlines how this will work:
“a client assistant residing in a client computer monitors movements of a user controlled pointer in a web browser, e.g., when the pointer moves into a predefined region and when it moves out of the predefined region. A server then determines a relevancy value between an informational item associated with the predefined region and a search query according to the pointer hover period. When preparing a new search result responsive to a search query, the server re-orders identified informational items in accordance with their respective relevancy values such that more relevant items appear before less relevant ones.”
In other words, by tracking when your mouse hovers over certain areas of a page, and for how long, Google is able to retain this information and use it when choosing the order of results for search queries that are relevant to the content you hovered over.
Time is an important factor here. By factoring in the length of time that you hovered over an area on a page, Google is hoping to be able to distinguish genuine interest (slowly and deliberately moving your mouse over results whilst choosing what to click – as if running your finger under a line on a printed page) from the rapid directional mouse movements and long-periods of latency that most likely make up the majority of your mouse’s day.
This is of particular importance to search results pages where small amounts of searched-for information (e.g. a price, single feature, postcode) can be gleaned from an organic snippet, meaning that users have less need to click through to the resulting page as they have already found what they were looking for. However, the mouse hover has, it is claimed, established a link between the user and that piece of content, which can then be used to tune future searches later.
So, does non-click mouse behaviour really describe intent? The patent was filed back in Feb 2005, long before many modern eye-tracking methods had been honed. Recent evidence from Acuity (eye tracking solutions) suggests that mouse movement has little co-relation to where a users eyes are looking on the page at a specific point in time – around 10% of all people “mark” places they are looking at on a page by moving their mouse to it.
Once again this is an example of pure clickthrough data being supplemented with richer attention-led data, which if it proves nothing else shows once again that Google believes that by putting consumers interests first it can generate a win-win scenario for users, internet marketing professionals/website owners … oh, and Google.