I've been reading up on:

<link rel="canonical" href="(original url)">

And one thing that I'm trying to figure out is the lowest percent of duplication must any two pages need to be in order for one to require the above tag. Any ideas?

  • Very good question. That issue sorta came up today already. It would be good to know what the opinions on this would be. Duplicate content is a semantics thing anymore. Google is not comparing the whole body of text, but how many and (I assume) scale of semantic links there are between pages. Google can find duplicate paragraphs, sentences, quotes, etc., but what is the tolerance level for duplication these days? Who knows. One up-vote. – closetnoc Apr 17 '15 at 3:36
  • rel=canonical is never "required" for two pages. If Google thinks two pages are duplicate it will just choose one to index. Canonical allows you to choose the one that is indexed instead. Canonical is also important when there are tens of variations on the URL (tracking parameters for example) so that Google can assign link juice and no spread it between all the versions. – Stephen Ostermiller Dec 13 '15 at 11:29

In this Google page: https://support.google.com/webmasters/answer/66359?hl=en

Duplicate content generally refers to substantive blocks of content within or across domains that either completely match other content or are appreciably similar. Mostly, this is not deceptive in origin. Examples of non-malicious duplicate content could include:

  • Discussion forums that can generate both regular and stripped-down pages targeted at mobile devices
  • Store items shown or linked via multiple distinct URLs
  • Printer-only versions of web pages

If your site contains multiple pages with largely identical content, there are a number of ways you can indicate your preferred URL to Google. (This is called "canonicalization".)

In this page: http://searchengineland.com/googles-matt-cutts-duplicate-content-wont-hurt-you-unless-it-is-spammy-167459

Duplicate content is a huge topic in the search engine optimization (SEO) space; heck, we even have a category devoted to the topic. But should we worry about it? Google’s head of search spam, Matt Cutts, said he wouldn’t stress about it — that is, unless it is spammy duplicate content.

In a video posted today, Matt Cutts answers, “How does required duplicate content (terms and conditions, etc.) affect search?”

Matt Cutts said twice that you should not stress about it, in the worse non-spammy case, Google may just ignore the duplicate content. Matt said in the video, “I wouldn’t stress about this unless the content that you have duplicated is spammy or keyword stuffing.”

Google has said time and time again, duplicate content issues are rarely a penalty. It is more about Google knowing which page they should rank and which page they should not. Google doesn’t want to show the same content to searchers for the same query; they do like to diversify the results to their searchers.

From this page: http://moz.com/learn/seo/duplicate-content

The Three Biggest Issues with Duplicate Content

  • Search engines don't know which version(s) to include/exclude from their indices
  • Search engines don't know whether to direct the link metrics (trust, authority, anchor text, link juice, etc.) to one page, or keep it separated between multiple versions
  • Search engines don't know which version(s) to rank for query results

Some back-ground:

Duplicate content is no longer measured by looking an entire page and has not since about 2008. Semantic links are used. In this case, fragments of content between pages are matched using semantic links which normally helps tie content together by topic, expertise, author, citations, and so on. If a high number of semantic links are discovered, then the pages are marked as duplicate content. No-one knows what this tolerance would be. For example, how many semantic links per X amount of content would be considered duplicate? Since content is often quoted, I would assume that the tolerance would be somewhat high. The earmark of duplicate content is that:

  • It is spammy in nature.
  • It adds no value.
  • It is designed to be deceptive.

I would say that any page that falls into one or more of these categories and contains a high number of semantic links would be worrisome at least.


While it would not be highly likely that a duplicate content penalty would be applied if some pages are a sub-set of another page's content, it might best to use the canonical tag to point to the page that is best for SERP performance. It may not always be the page with the complete content, but rather a more succinct version of the content. That would be up to you. Also consider which of the similar or duplicate pages have authority and the highest value back link profile? Either way, the canonical tag is used to help the search engine know which specific page of the rather similar pages is best for search. Assuming that you do use the canonical tag, consider that the pages linking to the original may lose some of their authority and link value. This is not clear. It is also not clear if the value of the duplicate pages is transferred to the original page using a canonical tag. While MOZ equates the canonical tag to a 301 redirect, how Google handles the canonical tag remains a mystery.

In that respect, no-one can really say how much is too much, however, I would venture to assume that a canonical tag is your best friend.

| improve this answer | |
  • I'm gonna need a week. I want to see what adsense thinks. Shortly after I began used canonical tags on the mobile version of my site to point to the desktop equivalent versions, My RPM went to $0. – Mike -- No longer here Apr 17 '15 at 4:46
  • BTW- Did you see my correction on an answer I gave to you about canonical tags pointing from mobile to desktop and back again?? Also, it is an interesting question if Adsense exists on a page that has a canonical tag that points to another page, how much value does that ad have?? – closetnoc Apr 17 '15 at 5:05
  • I'm just not sure canonical will work here because my mobile site doesn't contain the exact same text content as the desktop site nor does it contain the same layout. only thing the same is the photo. – Mike -- No longer here Apr 18 '15 at 0:04
  • Oh. Well then, you may be okay. 8-) Yippie! – closetnoc Apr 18 '15 at 0:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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