As part of Google’s aim is to provide more relevant and useful results, on Friday Google announced on their blog that they are now personalising results based upon 180 days of a users search activity. This has been the case for some time for Google users signed into their accounts, who have ‘Web History’ enabled, but this has now been rolled out to user’s results whether a user is signed in or not and has not specifically opted out of ‘Personalized Results’.
Using a cookie placed on users’ machines to track their search behavior Google uses web history to artificially inflate a sites ranking depending on a users previous preference to use that site in previous searches. You can opt out of personalised results, but it is unlikely the majority of users will.
Although this change will go relatively unnoticed to the average Google user, this has an impact on SEO. Traditional ‘ranking reports’ will be less representative of what users will see in the SERPs and the importance of regular, new, interesting and relevant onsite content and distribution of this online will be more key to help increase a sites ranking.
Ranking reports ensuring that there is no previous web history on the machine to skew results will show a snapshot of the SERPs at a basic level of a where sites rank in relation to one another prior to a user having any web history.
Providing more content on a site and attracting visitors to it regularly could help the site to rank higher in a Google users personalised SERP. If for example a hotel company provides regular and interesting content which a user regularly visits, when the user comes to search for ‘hotels’, they are more than likely to see the site they regularly visit for information ranking higher than it naturally ranks in a SERP where there is no previous web history of having visited that site.
Personalised search results is another one of Google’s ways to provide relevant content to the user and is in addition to other recent changes made by Google to improve relevancy in their results, such as Social Search, Twitter inclusion and the use of search query patterns & click through data to order results.
Gregory Lyons on 
Jeremy Head on 
Digital Marketing on 
Ali Ahmadyar on 
Dinesh Thakur on 
Dave Oliver on 
Art Rogue Studio on 
Rob Pearson on 
December 8th, 2009 at 10:12 pm
There seems to be a split opinion on how much of a huge deal this will be from some of the posts I’ve read – with the majority probably saying it is big!
One interesting point is the effect on spam. Having a level of behavioural tracking should in theory mean that spam sites will find it more difficult to rank (perhaps pushed down based on click throughs and bounce rate data?)
I read an interesting post on this which noted some suggested types of personalisations via Geo-targetting, technical, time-based and behaviour – and possible others around social search, (general) user history and demographics.
It’s the latter I find particularly interesting. While most are probably aware of these possible metrics as technical, seach, or social – it’s more important than ever to ensure the basics are done – that we understand a client’s objectives and target market (including demographics). It could be something as small as considering this data during keword research to collaborating with as much of the client’s other marketing channels as possible.
As a wise colleague of mine says – “Don’t isolate the marketing channels!”