Jump to content

  • Log in with Facebook Log in with Twitter Log In with Google      Sign In   
  • Create Account

Subscribe to HRA Now!

 



Are you a Google Analytics enthusiast?

Share and download Custom Google Analytics Reports, dashboards and advanced segments--for FREE! 

 



 

 www.CustomReportSharing.com 

From the folks who brought you High Rankings!



Photo
- - - - -

How Powerful Is Link Rel="canonical"


  • Please log in to reply
5 replies to this topic

#1 qwerty

qwerty

    HR 10

  • Moderator
  • 8,628 posts
  • Location:Somerville, MA

Posted 18 March 2011 - 02:14 PM

I've got a hypothetical (well, partly hypothetical) question about the strength and proper use of link rel="canonical". I'm wondering if anyone has bothered to test this.

Let's say you're developing a site in .NET (lucky you), so your URLs aren't case-sensitive. A page called page.aspx can be reached at PAGE.aspx, Page.aspx, paGE.aspx, etc etc. But you're very consistent -- every time you link to the page, without exception, you use the URL Page.aspx.

Your page becomes rather popular, and other sites start linking to it. But your luck holds, and 100% of the links to your page, both internal and external, refer to it as Page.aspx. Life is sweet.

That is, life is sweet until the owner of the site tells you he doesn't like the way capital letters look in his URLs, and he wants you to change everything to lower-case (personally, I agree with the no-caps rule, but like I said, this is semi-hypothetical). The question is, what's going to happen if you go to your page and stick this in the <head>:
HTML
<link rel="canonical" href="http://domain.com/page.aspx" />


Before we go any further, here's my understanding of link rel="canonical" and what it's for: if a given page can be reached at more than one URL, you use it to let search engines know which of the possible URLs you prefer. Without it, the search engines would look over whatever different URLs are being used to reach the page, see that they were dealing with duplicate content, and decide for themselves which URLs to ignore, leaving just one that they would treat as canonical. This is a common situation with e-com sites, where the same content may be presented in different orders, sorting a list of products by different criteria.

So my question is about how strong a signal it is to use this code. In the hypothetical above (which by the way is not an e-com site, so it doesn't have those obvious issues mentioned above), 100% of hundreds, perhaps thousands of links from a wide variety of domains are all pointing to Page.aspx. Is one line of code telling the search engines you'd rather they count page.aspx as canonical, despite all evidence to the contrary, enough to make a difference? If not, what is? What about if all of your internal links agree with the rel="canonical" choice, but you've got a ton of links from other sites using a different URL -- who wins?


#2 Michael Martinez

Michael Martinez

    HR 10

  • Active Members
  • PipPipPipPipPipPipPipPipPipPip
  • 5,157 posts
  • Location:Georgia

Posted 18 March 2011 - 08:22 PM

I have seen evidence that suggests it may be overriding "noindex,nofollow" meta tags, but I don't have enough data to be sure of that. It seems to me that it might do what you want but there is really only one way to find out.

#3 qwerty

qwerty

    HR 10

  • Moderator
  • 8,628 posts
  • Location:Somerville, MA

Posted 19 March 2011 - 08:23 AM

I wouldn't really say that overriding large numbers of links is what I want. I just want to know.

As I recall, when Google announced the tag, they referred to it with a word like "suggestion". In other words, if you're sending out mixed signals, by linking to a document in different ways or by allowing content to be presented in slightly different ways, then you can use link rel="canonical" to clarify your intentions, and that clarification is something Google will take into account when they determine which version they're going to treat as canonical.

But if you're already being consistent, I have to wonder 1) why you'd need to clarify your intentions and 2) whether, in the case that your "clarification" goes against what had been a consistent signal (and thus actually only serves to muddy the waters) it serves as a strong enough suggestion that Google will accept it, ignoring that other consistent signal.

#4 Jill

Jill

    Recovering SEO

  • Admin
  • 33,009 posts

Posted 19 March 2011 - 09:51 AM

It would do what you want. While there's the possibility of losing a tiny portion of the link equity from the links using the capital letter version, it's not something I would worry about.

Having the canonical link element on each page is fine, imo.

#5 CeeJ

CeeJ

    HR 2

  • Members
  • PipPip
  • 10 posts

Posted 19 March 2011 - 10:20 AM

I recently viewed a video by Matt Cutts that helps shed some light on this issue.

www.google.com/support/webmasters/b...y?answer=139394

There is also no mention of rel=canonical interferring with nofollow or no index
attributes.

Edited by CeeJ, 19 March 2011 - 11:04 AM.


#6 chrishirst

chrishirst

    A not so moderate moderator.

  • Moderator
  • 7,108 posts
  • Location:Blackpool UK

Posted 20 March 2011 - 07:35 AM

You could just add some code to force all lowercase URIs




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

SPAM FREE FORUM!
 
If you are just registering to spam,
don't bother. You will be wasting your
time as your spam will never see the
light of day!