Moz Q&A is closed.
After more than 13 years, and tens of thousands of questions, Moz Q&A closed on 12th December 2024. Whilst we’re not completely removing the content - many posts will still be possible to view - we have locked both new posts and new replies. More details here.
Rel="canonical" and rel="alternate" both necessary?
-
We are fighting some duplicate content issues across multiple domains. We have a few magento stores that have different country codes. For example: domain.com and domain.ca, domain.com is the "main" domain.
We have set up different rel="alternative codes like:
The question is, do we need to add custom rel="canonical" tags to domain.ca that points to domain.com?
For example for domain.ca/product.html to point to:
Also how far does rel="canonical" follow? For example if we have:
domain.ca/sub/product.html canonical to domain.com/sub/product.html
then,
domain.com/sub/product.html canonical to domain.com/product.html -
I'm honestly not completely clear on what the different URLs are for - I'd just add a note to keep the core difference between canonical and 301s in mind. A canonical tag only impacts Google, and eventually, search results. A 301 impacts all visitors (and moves them to the other page). A lot of people get hung up on the SEO side, but the two methods are very different for end-users.
As Tom said, if these variations have no user value, you could consolidate them altogether with 301s. I always hesitate to suggest it without in-depth knowledge of the site, though, because I've seen people run off and do something dangerous.
-
What's the purpose of the URL if there's not even any sorting or anything unique going on? If's a sorted URL (say by "size" smallest-largest for /little leage/ URL) it might be actually useful to develop some unique category content to let the page rank separately.
If the content is totally unique, I don't think you could really go wrong redirecting. To be safe, I'd probably rely on analytics to answer the question "what impact will redirection have?" For instance, is there a difference in conversion rate between the URLs. If you see a conversion bump from a more specific URL, you might want to sleuth out what's causing it.
-
Would you worry about it if the categories are somewhat useful for users to drill down the content?
For example:
/product.html
/aluminum-baseball-bats/product.html
/little-league-baseball-bats/product.htmlThey don't sell bats but it is the easiest way to describe it I guess. In this cause would you still 301 redirect the two longer urls to /product.html
-
Yes, providing that the /category1/ and /category2/ heirarchy doesn't help the user experience (e.g. product segmentation based on say, color and brand, which would be useful for users to drill down to).
I like 301s better because they are permanent, non-ambiguous, respected by all engines, and chiefly because they eliminate the possibilty of inlink dillution because the redirected URLs are never seeen.
-
Yeah, don't use rel=canonical for the same purpose as rel=alternate - the canonical tag will override the alternate/lang tag and may cause your alternate versions to rank incorrectly or not at all. It can be a bit unpredictable. If you only wanted one version to show up in search results, then rel=canonical would be ok, but rel=alternate is a softer signal to help Google rank the right page in the right situation. It's not perfect, but that's the intent.
As for multiple canonicals like what you described, that's essential like chaining 301-redirects. As much as possible, avoid it - you'll lose link equity, and Google may just not honor them in some cases. There's no hard/fast limit, and two levels may be ok in some cases, but I think it's just a recipe for trouble long-term. Fix the canonicals to be single-hop wherever possible.
-
Thanks that is what I was thinking, I just need to know more about if the bots will follow the canonical's past one level when pointing to a different domain and if so how many levels on the different sites.
-
Interesting idea, I might have to do that. Right now I have canonical elements on the .com
It is a magento store so it creates dirty duplicate content when the products are in different categories out of the box, for example magento creates the following product pages:
domain.com/store/productcategory1/product.html
domain.com/store/productcategory2/product.html
domain.com/store/product.htmlIn this case I have canonical elements pointing the categories to the main root domain.com/store/product.html
So you think it would be better to do a 301 redirect for the different product urls that are in subcategories?
-
Miles,
On your last question, I'm wondering if those two canonical tags are necessary? Are the /sub/ versions of those pages necessary for user experience? If not, I'd add a canonical element to the .com version, then redirect the /sub/product.html to /product.html. That would help you avoid splitting link authority.
-
Hey Miles,
The both are for different uses and may or may not be used in the same page depending on your situation.
If the content in the CA and COM versions is the same, then you should add a rel canonical + rel alternate, the rel alternate pointing to itself and the other version of it, and the canonical pointing to the one you consider definitive.
If the content isn't the same, then the rel canonical isn't needed (but suggested, pointing to itself in each lang/alternate), only the alternate should be in place.
You can read more on Dr. Pete's post here: http://moz.com/blog/rel-confused-answers-to-your-rel-canonical-questions
Hope that helps!
Got a burning SEO question?
Subscribe to Moz Pro to gain full access to Q&A, answer questions, and ask your own.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
Does redirecting from a "bad" domain "infect" the new domain?
Hi all, So a complicated question that requires a little background. I bought unseenjapan.com to serve as a legitimate news site about a year ago. Social media and content growth has been good. Unfortunately, one thing I didn't realize when I bought this domain was that it used to be a porn site. I've managed to muck out some of the damage already - primarily, I got major vendors like Macafee and OpenDNS to remove the "porn" categorization, which has unblocked the site at most schools & locations w/ public wifi. The sticky bit, however, is Google. Google has the domain filtered under SafeSearch, which means we're losing - and will continue to lose - a ton of organic traffic. I'm trying to figure out how to deal with this, and appeal the decision. Unfortunately, Google's Reconsideration Request form currently doesn't work unless your site has an existing manual action against it (mine does not). I've also heard such requests, even if I did figure out how to make them, often just get ignored for months on end. Now, I have a back up plan. I've registered unseen-japan.com, and I could just move my domain over to the new domain if I can't get this issue resolved. It would allow me to be on a domain with a clean history while not having to change my brand. But if I do that, and I set up 301 redirects from the former domain, will it simply cause the new domain to be perceived as an "adult" domain by Google? I.e., will the former URL's bad reputation carry over to the new one? I haven't made a decision one way or the other yet, so any insights are appreciated.
Intermediate & Advanced SEO | | gaiaslastlaugh0 -
Best way to "Prune" bad content from large sites?
I am in process of pruning my sites for low quality/thin content. The issue is that I have multiple sites with 40k + pages and need a more efficient way of finding the low quality content than looking at each page individually. Is there an ideal way to find the pages that are worth no indexing that will speed up the process but not potentially harm any valuable pages? Current plan of action is to pull data from analytics and if the url hasn't brought any traffic in the last 12 months then it is safe to assume it is a page that is not beneficial to the site. My concern is that some of these pages might have links pointing to them and I want to make sure we don't lose that link juice. But, assuming we just no index the pages we should still have the authority pass along...and in theory, the pages that haven't brought any traffic to the site in a year probably don't have much authority to begin with. Recommendations on best way to prune content on sites with hundreds of thousands of pages efficiently? Also, is there a benefit to no indexing the pages vs deleting them? What is the preferred method, and why?
Intermediate & Advanced SEO | | atomiconline0 -
Are HTML Sitemaps Still Effective With "Noindex, Follow"?
A site we're working on has hundreds of thousands of inventory pages that are generally "orphaned" pages. To reach them, you need to do a lot of faceting on the search results page. They appear in our XML sitemaps as well, but I'd still consider these orphan pages. To assist with crawling and indexation, we'd like to create HTML sitemaps to link to these pages. Due to the nature (and categorization) of these products, this would mean we'll be creating thousands of individual HTML sitemap pages, which we're hesitant to put into the index. Would the sitemaps still be effective if we add a noindex, follow meta tag? Does this indicate lower quality content in some way, or will it make no difference in how search engines will handle the links therein?
Intermediate & Advanced SEO | | mothner0 -
Switching from HTTP to HTTPS: 301 redirect or keep both & rel canonical?
Hey Mozzers, I'll be moving several sites from HTTP to HTTPS in the coming weeks (same brand, multiple ccTLDs). We'll start on a low traffic site and test it for 2-4 weeks to see the impact before rolling out across all 8 sites. Ideally, I'd like to simply 301 redirect the HTTP version page to the HTTPS version of the page (to get that potential SEO rankings boost). However, I'm concerned about the potential drop in rankings, links and traffic. I'm thinking of alternative ways and so instead of the 301 redirect approach, I would keep both sites live and accessible, and then add rel canonical on the HTTPS pages to point towards HTTP so that Google keeps the current pages/ links/ indexed as they are today (in this case, HTTPS is more UX than for SEO). Has anyone tried the rel canonical approach, and if so, what were the results? Do you recommend it? Also, for those who have implemented HTTPS, how long did it take for Google to index those pages over the older HTTP pages?
Intermediate & Advanced SEO | | Steven_Macdonald0 -
Brackets vs Encoded URLs: The "Same" in Google's eyes, or dup content?
Hello, This is the first time I've asked a question here, but I would really appreciate the advice of the community - thank you, thank you! Scenario: Internal linking is pointing to two different versions of a URL, one with brackets [] and the other version with the brackets encoded as %5B%5D Version 1: http://www.site.com/test?hello**[]=all&howdy[]=all&ciao[]=all
Intermediate & Advanced SEO | | mirabile
Version 2: http://www.site.com/test?hello%5B%5D**=all&howdy**%5B%5D**=all&ciao**%5B%5D**=all Question: Will search engines view these as duplicate content? Technically there is a difference in characters, but it's only because one version encodes the brackets, and the other does not (See: http://www.w3schools.com/tags/ref_urlencode.asp) We are asking the developer to encode ALL URLs because this seems cleaner but they are telling us that Google will see zero difference. We aren't sure if this is true, since engines can get so _hung up on even one single difference in character. _ We don't want to unnecessarily fracture the internal link structure of the site, so again - any feedback is welcome, thank you. 🙂0 -
Yoast & rel canonical for paginated Wordpress URLs
Hello, our Wordpress blog at http://www.jobs.ca/career-resources has a rel canonical issue since we added pagination to the front page and category-pages. We're using Yoast and it's incorrectly applying a rel-canonical meta tag referencing page 1 on page 2, 3, etc. This is a known misuse of the rel-canonical tag (per Google's Webmaster Blog - http://googlewebmastercentral.blogspot.ca/2013/04/5-common-mistakes-with-relcanonical.html, which says rel-canonical should be replaced with rel-prev and rel-next for page 2, 3, etc.). We don't see a way to specify anywhere in Yoast's options to correct this behaviour for page 2, 3, etc. Yoast allows you to override a page's canonical URL, otherwise it automatically uses the Wordpress permalink. My question is, does anyone know how to configure Yoast to properly replace rel-canonical tags with rel-prev and rel-next for paginated URLs, or do I need to look at another plugin or customize the behavior directly in my child theme code? This issue was brought up here as well: http://moz.com/community/q/canonical-help, but the only response did not relate to Yoast. (We're using Wordpress 3.6.1 and Yoast "Wordpress SEO" 1.4.18)
Intermediate & Advanced SEO | | aactive0 -
Can too many "noindex" pages compared to "index" pages be a problem?
Hello, I have a question for you: our website virtualsheetmusic.com includes thousands of product pages, and due to Panda penalties in the past, we have no-indexed most of the product pages hoping in a sort of recovery (not yet seen though!). So, currently we have about 4,000 "index" page compared to about 80,000 "noindex" pages. Now, we plan to add additional 100,000 new product pages from a new publisher to offer our customers more music choice, and these new pages will still be marked as "noindex, follow". At the end of the integration process, we will end up having something like 180,000 "noindex, follow" pages compared to about 4,000 "index, follow" pages. Here is my question: can this huge discrepancy between 180,000 "noindex" pages and 4,000 "index" pages be a problem? Can this kind of scenario have or cause any negative effect on our current natural SEs profile? or is this something that doesn't actually matter? Any thoughts on this issue are very welcome. Thank you! Fabrizio
Intermediate & Advanced SEO | | fablau0 -
Rel=canonical tag on original page?
Afternoon All,
Intermediate & Advanced SEO | | Jellyfish-Agency
We are using Concrete5 as our CMS system, we are due to change but for the moment we have to play with what we have got. Part of the C5 system allows us to attribute our main page into other categories, via a page alaiser add-on. But what it also does is create several url paths and duplicate pages depending on how many times we take the original page and reference it in other categories. We have tried C5 canonical/SEO add-on's but they all seem to fall short. We have tried to address this issue in the most efficient way possible by using the rel=canonical tag. The only issue is the limitations of our cms system. We add the canonical tag to the original page header and this will automatically place this tag on all the duplicate pages and in turn fix the problem of duplicate content. The only problem is the canonical tag is on the original page as well, but it is referencing itself, effectively creating a tagging circle. Does anyone foresee a problem with the canonical tag being on the original page but in turn referencing itself? What we have done is try to simplify our duplicate content issues. We have over 2500 duplicate page issues because of this aliasing add-on and want to automate the canonical tag addition, rather than go to each individual page and manually add this tag, so the original reference page can remain the original. We have implemented this tag on one page at the moment with 9 duplicate pages/url's and are monitoring, but was curious if people had experienced this before or had any thoughts?0
