Rel=canonical overkill on duplicate content?

[Eric_Lifescript]

Our site has many different  health centers - many of which contain duplicate content since there is topic crossover between health centers. I am using rel canonical to deal with this.

My question is this: Is there a tipping point for duplicate content where Google might begin to penalize a site even if it has the rel canonical tags in place on cloned content?

As an extreme example, a site could have 10 pieces of original content, but could then clone and organize this content in 5 different directories across the site each with a new url. This would ultimately result in the site having more "cloned" content than original content. Is this at all problematic even if the rel canonical is in place on all cloned content?

Thanks in advance for any replies.

Eric

[AlanBleiweiss]

Rel-Canonical is a signal.  While it helps mitigate problems of duplication, it is just one signal.  If enough other overriding factors exist, Google's automated ability to "figure things out" can become weak and confused.  So for example, if all of the pages are linked to from multiple points within the site, or if any of that content is linked to from outside the site, there could be negative impact.  It's not supposed to occur, however it can due to imperfect layering of multiple algorithms.

Ultimately, there may be a need to block some content more forcefully, such as via robots.txt file.  Unfortunately there's no definitive method for evaluating this specific issue in an isolated manner without serious live site testing over extended periods of time.  So my recommendation to clients is to not mess with it unless you've seen a serious drop in organic results, and only even consider testing if you can't identify other potential primary causes.