I have some duplicated pages that show the exact same content on different URLS on my website, and I use canonical links to point them all to one of the pages.

So if Page A, B, C, D and E have the same content, the canonical links on pages B,C,D, and E all point to page A, and page A does not have a canonical link.

I see some warnings in Google Webmaster Tools -> Optimization -> HTML Improvements, regarding duplicated descriptions, titles, and more even though I have used the canonical links.

Have I misunderstood how canonical links work? Isn't this their correct usage?

2 Answers 2


You are probably missing a canonical link or Google did not have a chance to see them all. Many errors shown in Google Webmaster Tools are accumulated over time and only get removed after Google had time to see and re-index the page.

From you description it sounds like you are missing the canonical link on one page. TO be safe, all identical pages should have the canonical link and it should point to the same URL.

Keep in mind that canonical is still considered a suggestion by Google and if for some reason there is substantial differences between those pages, Google may still consider them distinct.


Personally I'd opt to use:

  • Page A <link rel="canonical" href="http://xxx.com/pageA" />
  • Page B <meta name="robots" content="noindex, follow">
  • Page C <meta name="robots" content="noindex, follow">
  • Page D <meta name="robots" content="noindex, follow">
  • Page E <meta name="robots" content="noindex, follow">

Edit: Reasoning for this method for people asking about back link flow when using the noindex, follow tag.

Pages that are no index can still achieve a page rank and still pass backlinks by using the noindex, follow you are informing Google to not index your page but follow the links on the page this means that the juice gained on the noindex page is dripped down to the rest of the site. These back links 'can' even show up in Google Web Master tools.

Extremely popular addons such as All in One SEO for WordPress use such method on Tag pages that are in turn duplicate but may obtain backlinks that don't want to be wasted. There are hundreds of reports if not thousands confirmed cases of this working as well as official wording from John Mueller at Google Webmaster answering hangout questions (Skip to 33:30 and watch for a couple of mins).

Also in terms of the canonical tag its ok to use it on the correct page and not the correct pages... This tag is only to tell Google 'This is the master' if its found on the master then its ok, the benefits of this is in your CMS if that page magically appears on a page it shouldn't for example no-www, or a page=id?42 then it automatically has the "canonical" tag... Again many of the top SEO plugins for various CMS use this to avoid duplicate content - to imply it must be used on all duplicates but not the master is false. Also the likes of big sites such as BBC and so on all use canonical on the master pages as well as the duplicate pages.

Alternatively you could use this since both methods are valid:

  • Page A <link rel="canonical" href="http://xxx.com/pageA" />
  • Page B <link rel="canonical" href="http://xxx.com/pageA" />
  • Page C <link rel="canonical" href="http://xxx.com/pageA" />
  • Page D <link rel="canonical" href="http://xxx.com/pageA" />
  • Page E <link rel="canonical" href="http://xxx.com/pageA" />

Ensure that you use the canonical on the master page. As long as you have something like this it doesn't matter what GMT reports.

  • Hi DKOATED, pages that are noindex, follow are still crawled and processed - they are just hidden from the search - In the event noindex pages receive PR from backlinks this is then feed down to the rest of the site. I've updated my answer with references and have many more if you would like ;) Mar 22, 2013 at 10:22
  • Also if you are going to use terms like 'simply non-sense' then please cite references that contradicts the answer, Simply telling someone its nonsense doesn't help people reading this page and can wrongly give the impression the answer is incorrect when it may be correct. Mar 22, 2013 at 10:32
  • "to imply it must be used on all duplicates but not the master is false": wrong. you MUST use it on all duplicates, but MAY use it on the master (has no effect).
    – David K.
    Mar 22, 2013 at 10:46
  • and still, simply setting duplicates on noindex may refrain google from penalized said pages. Setting canonicals instead or preferably canonicals WITH noindex on duplicate pages is better.
    – David K.
    Mar 22, 2013 at 10:47
  • Yes, your not reading though the lines.... By using it on the master then assume that the same page is accessible via no-www, or a pageID rather than the canonical then because the page renders as it normally would it includes the canonical this is why its better to include canonical on the master page incase that page is accessible though another way that your not aware off... Your going of situations where you know the duplicate pages. Mar 22, 2013 at 10:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.