Google Use Data From Google Toolbar To Rank Related Websites?
Posted by nullbit on September 23, 2005, 11:49 pm
This is actually something I noticed a long time ago. I'd assumed it would already be widely known but a quick google suggested otherwise, so here goes:
In working on a client's site I frequently needed a second site -- that was totally irrelevant in terms of theme, but, relevant to to the work I was doing -- for reference, and would often switch between these two sites.
Several weeks later, I noticed the irrelevant site listed in a Google search for related:www.client-site.com (note though, that the reverse was not true). Given the total irrelevance of these two sites I can only conclude that Google uses data from Google Toolbar users' visiting habits as part of their algorithm for deciding related sites -- actually I was using a Firefox PageRank extension, but since this sends the same data about URLs you visit back to Google, it's incidental.
It may seem strange that my browsing alone could have such influence, but consider the fact that the client's site had very low traffic and the majority of it was me, and that I have dynamic IP. That meant from Google's perspective it would have appeared that the majority of unique visitors to the client site, had also visited the irrelevant site within a close time frame.
What can be gained from this knowledge from an SEO perspective? Probably very little at face value, as I don't think many users actually use related: queries. But, if this is only the tip of the iceberg, and Google use similar data for the main ranking algorithm, it could mean a lot more.
Of course, my support for this theory is extremely anecdotal. I'd invite anyone with the inclination to investigate further and provide a more controlled study to do so.
Add to del.icio.us | Post Comment
Discuss
No comments
Submit Comment