(14)
Nick Roshon has an excellent post on the difference between how Google & Bing handle the canonical tag, something non of us really noticed before until he pointed it out.
The key differences are two fold:
(1) Mass use of the canonical tag and
(2) Canonical tag pointing to itself.
How does each engine differ?
(1) Google doesn't care if you use the canonical tag on hundreds or thousands of pages, where Bing says they will trust the tag less if you do.
(2) Google doesn't care if you link a canonical tag to the same page (i.e. a redirect back to the redirect), but Bing does and says don't do it.
Nick quotes Bing's Duane who wrote, "the rel=canonical is that it was never intended to appear across large numbers of pages" and "Its best to leave them [canonical tags] blank rather than point them at themselves."
Nick then quotes Google's Cutts and Ohye who say, "it doesn’t hurt to have this on every page of your site," and "it's absolutely okay to have a self-referential rel="canonical". "
These are very interesting distinctions between Bing.
Forum discussion at Sphinn.
Comments:
John Doherty
10/11/2011 02:09 pm
Barry - Interesting article here and Nick definitely has some good thoughts. His comments from Cutts and others are telling as well. I think Bing is really giving us a picture into that they are not as sophisticated as Google when it comes to identifying these self-referential canonicals. Why should a safeguard like this against duplicate content be something that would harm rankings? Of course, maybe this also shows us that Bing does not consider duplicate content to be as big of an issue. I wrote some thoughts here yesterday: http://bit.ly/n85aNj. I'd be interested to hear your and Nick's thoughts.
Michael Martinez
10/11/2011 02:49 pm
Actually, I like Bing's point of view on this. I think people are blowing it out of proportion. Google tolerates a lot more crap than Bing does. It's always been like that.
Nick Roshon
10/11/2011 03:37 pm
Barry, thanks for sharing my post - it means a lot. It'd be great to get more clarification on this issue, particularly from Bing. For now, I plan on still recommending site-wide rel=canonical tag for large e-commerce sites that generate lots of duplicate content through action tracking parameters & dynamically generated content, as I think that's the best option we as SEOs have when a 301 redirect isn't possible.
Marc Perez
10/11/2011 05:02 pm
Have you ever tried to write a rule that only work on certain pages but not others?
Michael Martinez
10/11/2011 05:06 pm
Yes, I have written such rules. I would be surprised if anyone could correctly canonicalize to another page through a rule anyway. But for sites that are generating search-superfluous URL parameters it should be sufficient to just strip the parameters for a self-referential canonical URL.
Matthew Hanagan
10/11/2011 05:22 pm
I agree Michael. It's very odd to have a 'virtual redirect' on a page to itself. Although I have done it on my current site, it breeds laziness in design. My reaction to Bing is 'too late now'.
Cicada Mania
10/11/2011 08:02 pm
Google sends me 75% of my search traffic, so I'm gonna do what Matt says. Actually, I'm going to do what Wordpress does for my blog. I'm not going to hack the code.
Tony Adam
10/11/2011 09:13 pm
The logic doesn't make sense on Bing's part to have a blanket statement like that. I love Duane and all, but, he should know after working as an SEO at Microsoft, tons of sites have thousands, if not millions of URLs that might have duplicate content issues that the canonical tag resolves. After working at Myspace and Yahoo, and advising a few companies on this, it's ludicrous to think it won't be used widespread like that...very short sighted on Bing's part.
Brandon Fritz
10/11/2011 09:28 pm
Great Bing. Clap, clap, clap. /sarcasm>
SearchRocket
10/11/2011 10:59 pm
Well, Bing's interpretation doesn't agree with the draft RFC, available at http://tools.ietf.org/html/draft-johnston-addressing-link-relations-01. Of course, the draft RFC was written by a Googler, but someone's got to write the standards. I'd quite like to know Duane's source for the intent of rel=canonical.
Mark Warner
10/12/2011 10:32 am
Well I have slapped it all my pages Maybe it wasn't what the tag was designed for, but I like the idea of giving each page its "own name" set in stone. I can see Bing's argument , but leaving the canonical tag blank doesn't feel right to me and seems worse than pointing to self
SEO Catalysts
10/12/2011 02:01 pm
Yes in some area bing look rude in compare to Google...but both are coming with new features on every day -Search Engine Optimization
Adam Sherk
10/12/2011 08:42 pm
Speaking from the news publisher POV, adding rel=canonical tags to every page is needed to mitigate issue with things like appending tracking codes to URLs. Because in most cases their CMS considers site.com/page-name and site.com/page-name?trackingcode=X as the same page. So the only way to add rel=canonical tags to the duplicates is to add it to the main page.
Joe Chasse
08/20/2013 06:44 pm
I agree with Adam. If you have tracking strings appended to a page URL, only by adding a rel=canonical tag to every page will it mitigate potential duplicate content issues.