Is reported duplication on the pages or their canonical pages?

[Safelincs]

There are several sections getting flagged for duplication on one of our sites:

http://mysite.com/section-1/?something=X&confirmed=true
http://mysite.com/section-2/?something=X&confirmed=true
http://mysite.com/section-3/?something=X&confirmed=true

Each of the above are showing as having duplicates of the other sections. Indeed, these pages are exactly the same (it's just an SMS confirmation page you enter your code in), however, they all have canonical links back to the section (without the query string), i.e. section-1, section-2 and section-3 respectively.

These three sections have unique content and aren't flagged up for duplications themselves, so my questions are:

Are the pages with the query strings the duplicates, and if so why are the canonical links being ignored?

or

Are the canonical pages without the query strings the duplicates, and if so why don't they appear as URLs in their own right in the duplicate content report?

I am guessing it's the former, but I can't figure out why it would ignore the canonical links. Any ideas?

Thanks

[Safelincs]

This is good news sugar-coating bad news Thanks!

[Adam.Whittles]

Hi,

The URLs that are reported by the crawl as being duplicates are the duplicate pages. Unfortunately the way the crawl from SEOMoz works, it does not factor the rel=canonical tag when reporting duplicates. In other words, even with the tag implemented, it will still report these pages as duplicates. Don't worry though, as long as the tag is implemented, the search engines should treat the canonical like a 301 redirect and not penalise you for duplicate content.

So to answer your question:

Are the pages with the query strings the duplicates? - Yes.

Hope that helps,

Adam

[Marcus_Miller]

Hey,

It's kind of tricky to answer this without seeing at least two of the category pages but I am guessing that the duplication is in the category pages themselves and if they are simply very thin pages with little to differentiate category A from category B then there is your problem.

Rather than look at the web tool, if you export the spreadsheet this is a lot easier to understand and for each page there is a duplication column which has a comma separated list of the pages that are being flagged as possible duplicates so this should answer your question.

What to do though?

I may be telling you how to suck eggs but this is always a good read when it comes to thin content problems and solutions:
http://www.seomoz.org/blog/fat-pandas-and-thin-content

If it was me, and these pages are thin, but that is the way they are supposed to be, and they are not really search landing pages then there is a good argument to noindex them and remove the possibility of them causing you any problems. If you do this, next time the campaign tool crawls your site they will be ignored and will not show up as a possible duplicate.

Obviously, from a Panda perspective, if these pages are listed as thin, they could be damaging other pages on the site so it is certainly an issue worth addressing.

Hope this helps!
Marcus