It is NOT about bounce rates.

Yes you're right, it's not bounce rate. I never said it was. Bounce is a bulk metric....What I am talking about is a segmented notion of bounce rate. There are multiple ways to bounce off site, they can go anywhere....but the problem is...if they came in via a google search, and then returned to that search, then I feel that that is the problem...and those metrics you can not measure. Your right, it's NOT Bouncerate...it's the part of bounce rate where they go back to the google search and click on someone else....that can not be measured via bounce rate.

Plenty of people use Google to do shopping research, so they will naturally return to Google after checking out one site to visit another.  That is in no way a quality signal.

People seeking different opinions on topics will do the same thing.

People seeking as much information about a topic will do the same thing.

People seeking a specific recipe will do the same thing.

Detecting whether people are bouncing off sites doesn't tell a search engine whether the quality of the site is good or bad.  They have to be able to interpret the reason for the bouncing.

Plenty of high bounce-rate sites are doing just fine (including my own).  Plenty of sites that have been focusing on reducing bounce rates have complained of getting nowhere.

It's time for people to stop discussing bounce as a part of the Panda update.  We can leave it on the table for future consideration if something new develops but it's been beaten to death for now.

The day after Panda was released, there was an interview with Amit Singhal (a google engineer) called "Google’s Singhal On Why Google Had Spam Issues"

to quote that article:

"Singhal did say that the company added numerous “signals,” or factors it would incorporate into its algorithm for ranking sites. Among those signals are “how users interact with” a site. Google has said previously that, among other things, it often measures whether users click the “back” button quickly after visiting a search result, which might indicate a lack of satisfaction with the site."

So you can ignor those words if you'd like....that's your choice.

So the Wall Street Journal reporter points out that Google has previously discussed how it measures user clicks without attributing that signal to Amit Singhal's discussion of what went into the Panda algorithm.

On the basis of this attachment to the discussion, you're assuming that Google must be looking at bounce rates.

And the sky is also made of white cake.

Wow, that article was a lot of words to express one simple piece of the Panda pie.

My reviews of sites hurt in the Panda Update would corroborate that Jim's certainly onto something, but there are tons of other factors that contribute to pages having been nuked.

The main emphasis with Panda was on quality and the ease of people finding exactly what they were looking for when they started at Google. So if they're coming right back to Google to keep searching, well, it makes sense that this would be factored into whether they found what they wanted or not.

SEOTheory..Nope...again...my thought is that it's not bounce rate....its bounce rate when they return to the google serp and then they click on another site in the SERP's and then they dont come back to your site. If I were google, this is a major signal that I'd use to measure quality of the pages in the search results...if you don't think that's being measured and used, that's fine...I believe it is....and I believe its one of the biggest factors in Panda...and my plans for clients who were effected is to work on having searches think that my clients site is the final destination for what the searcher is seeking... I feel confident that by at least going on this premise, that the advice that I'm giving to clients is in line with what Google would like to see for results in the SERPS.