Questions about rel=canonical
February 12, 2009 by ian
I'm already getting questions about the new rel=canonical standard. And I have several of my own. I'm going to put them into a post here. If you have your own, leave 'em in the comment thread and I'll move them into the post:
- Shouldn't we just add link rel=canonical to every single page on our sites? Wikia, which Google cites in their example, appears to have done that.
- How much link authority will the canonical tag pass? 100%? 90%? Less?
- How will each search engine support the new standard? What differences will there be?
- What if rel=canonical conflicts with the canonicalization settings in Google Webmaster Tools?
I will post answers as I get them.
Join the Conversation:
Jarrod Medrano
says:
February 12, 2009 15:17
Wouldn't it be easier to avoid having duplicate content in the first place?
Ian
says:
February 12, 2009 15:20
@jarrod Oh, you poor, innocent soul.
Daniel allen says:
February 14, 2009 21:26
Q. What are the possible uses/implications on deep linking?
diario says:
February 16, 2009 04:52
Hi Ian, the problem is that anyone outside can link to your webpage with a new attribute, so canonical is wonderful. Imagine: http://www.conversationmarketing.com/?param=hello
Charlie says:
February 16, 2009 08:58
Question 4 is perhaps answered by Matt Cutts in the .ppt on his blog post on the issue (http://www.mattcutts.com/blog/canonical-link-tag/):
"Break ties in Google by submitting your preferred url in a sitemaps file"
Jim Robinson says:
March 2, 2009 15:01
To me, #2 is the big question. If it passes link value, it's infinitely more useful, imo.
Mike says:
April 27, 2009 04:35
Sometimes you need to have different paths to a page. In e-commerce for example - route via category or route by manufacturer. I'm going to add this code to my manufacturer path. Is this the solution??
Jeff says:
April 28, 2009 09:22
So, if I have a home page URL that I want to use as the indexed URL and also a URL with tracking code from a banner, do I place the link tag on the index page? Because, obviously, the tracking code URL doesn't really exist as a page - it's a virtual URL.
With that said, how exactly would that work? Google finds any pages on my domain that seem similar to my homepage and give it the link juice?
Thanks!
Jeff
Ian
says:
April 28, 2009 10:39
@Jeff Use the primary URL in the canonical tag. So if your site is www.mysite.com, you want the tag to be rel="canonical" href="http://www.mysite.com".
PeterW says:
June 24, 2009 02:06
For eCommerce I need to show different content in short form, in detail form, I have different groupings for marketing purposes and all are valid and distinct. All the detail items are database driven including full population of Title, description, keywords, everything related to the product, images, price, yah-de-yah-da.....
It should be possible to add a link with a wildcard to explicitly tell the search engines that all are values are valid and distinct, e.g.
Is such a wildcard possible in a link?