(Entry was posted by Rick Molenaar on this post in the Dumb SEO Questions community on Facebook, 07/10/2013).

Should rel canonical be added to every page?

I had an arguement with a colleague who wants to add a rel="canonical" to any page, pointing to the page itself (except for some duplicate pages who will point to the `main` page).

This question begins at 02:05:00 into the clip. Did this video clip play correctly? Watch this question on YouTube commencing at 02:05:00

Share on Twitter

Share on Facebook

YOUR ANSWERS

Selected answers from the Dumb SEO Questions Facebook & G+ community.

Federico Sasso: Officially Google (differently from Bing) invites to add it to the non-canonical versions (without saying not to do it in the canonical one, at least last time I checked; I suspect they changed the wording since the first appearance of the support pages).

In my experience adding it to any page (canonical and non-canonical) is a best practice if you are working with IIS - or any other web server with a case-insensitive file system: this way you prevent badly cased links to cause duplicate content issues.

Working with IIS, I do normally use it on every page (it also simplify coding and maintenance) and never felt to be somehow penalized.
Rick Molenaar: Great comment +Federico Sasso! Thanks!
Ahmed Khalifa: Very interesting comment +Federico Sasso. I've had this conversation with another SEO before and it's one of those issues where there are no clear cut answer. I've done a lot of research and I remember reading it somewhere that Google does not mind it, but Bing does not encourage it. For me personally, I only want to use it when it's needed and it's for its original purpose: control duplication, etc. Therefore, I try to avoid self-referencing canonical tags, but I don't believe that it causes a serious issue, and I certainly haven't seen evidence of that either.

It's confusing, I know. But that's what makes SEO a challenge :-)
Masatake Wasa: +Federico Sasso ;makes really good points.

They are useful in other instances too: if you have +1 ;/ Google+ share buttons on your page, for example, and you do not specify the href attribute for the buttons, then Google will use the canonical URI. Third-party plug-ins for social sharing and other such may also use the canonical URI (or that which is specified in og:url). In any case, I really cannot see how self-referencing canonicalization could harm a site.
Luca Scaggion: I fully agree with +Federico Sasso. I have a big client who uses IIS and a custom CMS that primarily used to duplicate contents. Rel "canonical" successfully fixed the problem.
W.E. Jonk: Since it wasn't mentioned the rel canonical can help against scraping. Although it is weak, if the whole HTML is being scraped the rel canonical will be scraped too and it does work cross domain.
Rick Molenaar: Thanks guys! I appreciate your help!
Justin Y: I think it's pretty much a standard process nowadays to include canonical tags across the board. Everybody in this thread is on the same page and +Masatake Wasa made a great point about the 3rd parties.