A recent thread in Webmaster World’s Google Search News forum has a discussion of how a keyphrase can lose its prime position on a serp. It’s a wild thread, with a lot of ideas and war stories bantered about. What struck me as notable is a theory put out by a couple of astute observers that attempts to relate Google’s potential utilization of user data to determine site quality. One contributor makes the point nicely:

When MC has talked about sites impacted by this phenomena he consistently references making sure your site is useful to visitors. Many of us have taken that as a very vague answer. But perhaps it is more specific then we think. How could google measure how useful a visitor might think a site is? By measuring bounce rate.

Bounce rate is used here as the percentage of users who click on an organic or paid listing, quickly determine the landing page does not meet their needs, and hit the back button to find a more suitable link. Pages with lower bounce rates could [and probably should] be considered, well, “better” — more useful for the searcher, and therefore more valuable to Google.

Let’s say your landing page tells the product story, communicates your brand’s qualities, instills trust and conveys value. As your conversion rate improves, sales are strong, customers content. But what happens if a high-demand product sells more quickly than it can be replenished? The reduced inventory levels could produce a spike in bounces: user searches for pink widget size 4, comes to site and finds no stock in desired size/color combination, backs out and moves onto the next site in the serp. In this scenario, the increase in negative user activity could be taken by Google to mean the page has lost a measure of its usefulness.

So how do you correlate your bounce rates with your keyphrase serp positions over time? Most web analytics software will provide a reasonable set of tools — various APIs and ODBCs to integrate these data sets. But in this situation, I’ll take a lower-tech approach. This is, after all, only a theory, and I’d like to track it easily. But for now I’d prefer to allocate my limited resources to more critical projects. Capturing the bounce rate data is a snap, identifying any major movement in specific landing pages is already built in to our search engine reporting. Once we see a position dive and confirm we haven’t made any detrimental changes to the page content, we can check the bounces. If user data is influencing our search engine positions, understanding our users’ goals and realities becomes all the more critical.