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 between mirrored domains
-
Hi all & happy new near!
I'm new to SEO and could do with a spot of advice:
I have a site that has several domains that mirror it (not good, I know...) So www.site.com, www.site.edu.sg, www.othersite.com all serve up the same content. I was planning to use rel="canonical" to avoid the duplication but I have a concern:
Currently several of these mirrors rank - one, the .com ranks #1 on local google search for some useful keywords. the .edu.sg also shows up as #9 for a dirrerent page. In some cases I have multiple mirrors showing up on a specific serp.
I would LIKE to rel canonical everything to the local edu.sg domain since this is most representative of the fact that the site is for a school in Singapore but...
-The .com is listed in DMOZ (this used to be important) and none of the volunteers there ever respoded to requests to update it to the .edu.sg
-The .com ranks higher than the com.sg page for non-local search so I am guessing google has some kind of algorithm to mark down obviosly local domains in other geographic locationsAny opinions on this? Should I rel canonical the .com to the .edu.sg or vice versa?
I appreciate any advice or opinion before I pull the trigger and end up shooting myself in the foot!
Best regards from Singapore!
-
Wow - that's a huge impact. It's hard for me to believe this one change would have such an impact, but hopefully these new numbers stick.
-
Hi Dr. Peter,
Just thought I would share the initial results of your suggestion because they appear to be so dramatic!
I put it a rel=alternate tag on every page indicating the .edu.sg version of the URL for "en-sg" and the com version of the url for "en".
From MozAnalytics: Before
Rank 1-3: 4 keywords
Rank 4-10: 21 keywords
Rank 11-20: 3 keywords
Rank 21-50: 2 keywords
Rank 51+: 216 keywordsFrom MozAnalytics: After
Rank 1-3: 50 keywords!!!
Rank 4-10: 19 keywords
Rank 11-20: 3 keywords
Rank 21-50: 13 keywords
Rank 51+: 167 keywordsThat's pretty crazy in under a week - Not sure if its 'real' since I never personally went to check rankings on these keywords, but wow! If the MozAnalytics information is correct I am blown away! Unfortunately its for a niche site in a tiny market so its not going to lead to fame and riches, but its still an amazing result!
Thanks so much for the great advice
Alex
-
Happy to help - hope it does the trick.
-
Fantastic response Dr. Peter! Thanks for taking the time to explain and to suggest the rel="alternate" tag which I have now implemented in a little experiment.
I very much enjoy your posts on the moz blog and thank you for all the indirect help they have given us in the past.
Best regards,
Alex
-
I'm not sure there's a one-sized-fits-all answer. If the .com is more geared to an audience outside of Singapore, and the .edu.sg site is more geared to the local audience you could set a region and/or language with rel="alternate" hreflang:
https://support.google.com/webmasters/answer/189077?hl=en
This is a bit more subtle canonicalization signal that Google can use to sort out sites with language and/or regional copies (the regional aspect may be relevant even if both sites are in english).
The next question would be: where does your traffic come from? If you want to consolidate, but most of your traffic is coming from outside of Singapore, then I'd probably stick with the .com - it still has a "generic" status. The .edu.sg may rank more strongly in Singapore but fall off everywhere else.
I wouldn't worry much about the DMOZ link, especially if you have a solid link profile. DMOZ links have gotten buried over the years and typically don't carry nearly the value most people think. At some point, they could be a solid boost to a new site with a small link profile, but I think even those days are well behind us.
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
-
We switched the domain from www.blog.domain.com to domain.com/blog.
We switched the domain from www.blog.domain.com to domain.com/blog. This was done with the purpose of gaining backlinks to our main website as well along with to our blog. This set us very low in organic traffic and not to mention, lost the backlinks. For anything, they are being redirected to 301 code. Kindly suggest changes to bring back all the traffic.
Technical SEO | | arun.negi0 -
Rel=Canonical For Landing Pages
We have PPC landing pages that are also ranking in organic search. We've decided to create new landing pages that have been improved to rank better in natural search. The PPC team however wants to use their original landing pages so we are unable to 301 these pages to the new pages being created. We need to block the old PPC pages from search. Any idea if we can use rel=canonical? The difference between old PPC page and new landing page is much more content to support keyword targeting and provide value to users. Google says it's OK to use rel=canonical if pages are similar but not sure if this applies to us. The old PPC pages have 1 paragraph of content followed by featured products for sale. The new pages have 4-5 paragraphs of content and many more products for sale. The other option would be to add meta noindex to the old PPC landing pages. Curious as to what you guys think. Thanks.
Technical SEO | | SoulSurfer80 -
Spammers created bad links to old hacked domain, now redirected to our new domain. Advice?
My client had an old site hacked (let's call it "myolddomain.com") and the hackers created many links in other hacked sites with links such as http://myolddomain.com/styless.asp?jordan-12-taxi-kids-cheap-T8927.html The old myolddomain.com site was redirected to a different new site since then, but we still see over a thousand spam links showing up in the new site's Search Console 404 crawl errors report. Also, using the links: operator in google search, we see many results of spam links. Should we be worried about these bad links pointing to our old site and redirecting to 404s on the new site? What is the best recommendation to clean them up? Ignore? 410s? Other? I'm seeing conflicting advice out there. The old site is hosted by the client's previous web developer who doesn't want to clean anything up on their end without an ongoing hosting contract. So beyond turning redirects on or off, the client doesn't want to pay for any additional hosting. So we don't have much control over anything related to "myolddomain.com". 😞 Thanks in advance for any assistance!
Technical SEO | | usDragons0 -
Domains
My questions is what to do with old domains we own from a past business. Is it advantages to direct them to the new domain/company or is that going to cause a problem for the new company. They are not in the same industry.
Technical SEO | | KeylimeSocial0 -
Use webmaster tools "change of address" when doing rel=canonical
We are doing a "soft migration" of a website. (Actually it is a merger of two websites). We are doing cross site rel=canonical tags instead of 301's for the first 60-90 days. These have been done on a page by page basis for an entire site. Google states that a "change of address" should be done in webmaster tools for a site migration with 301's. Should this also be done when we are doing this soft move?
Technical SEO | | EugeneF0 -
Beating a keyword Domain
Has anyone here managed to beat a keyword/exact match domain to top spot? I am currently second and wondering if it is worth the time and effort to knock it off the top spot. How hard is it to get these very annoyingly favoured domains off 1st? Any help and advice much appreciated.
Technical SEO | | pauledwards0 -
How to structure rel=canonical for a e commerce site
Hello, So I have searched the Q & A , Google, the zen cart forum and at this point I am looking for some one to give a concrete answer on what I should do. There is a lot of different opinions on " rel=canonical" and how to apply it , since there are many other variable in place. I have a zen cart site. I am using the latest 1.3.9 version. The default setting ( seem to me) uses the rel=canonical to point back to the specific link product or category respectively. Most of the time I have two scenarios. 1. Main category ---> Sub category----> Product 2. Main Category----> Product I'll give an example http://www.perfectindesign.com/awards ---main category http://www.perfectindesign.com/awards/acrylic-awards sub category http://www.perfectindesign.com/awards/acrylic-awards/slanted-award product (this example has three sub categories with maybe 12 products in one 4 in the second and 5 in the third) From looking at the source code for each url it the rel=canonical just points back to its own url. I want to avoid competing against my self, for the example above keyword "acrylic awards" so should the use of the re=canonical be changes site wide to have products point back to sub categories when they exist and have products point back to main categories when no sub categories exist? I am very new to seo, specifically eCommerce seo. If you have experience and have done this to a site you manage for a client or your own please advise how to proceed. Also if I'm missing some thing that will give me a better understanding of the bigger seo picture that would be great. Thanks, Yevgeny
Technical SEO | | Yevgeny0 -
What is the best method to block a sub-domain, e.g. staging.domain.com/ from getting indexed?
Now that Google considers subdomains as part of the TLD I'm a little leery of testing robots.txt with something like: staging.domain.com
Technical SEO | | fthead9
User-agent: *
Disallow: / in fear it might get the www.domain.com blocked as well. Has anyone had any success using robots.txt to block sub-domains? I know I could add a meta robots tag to the staging.domain.com pages but that would require a lot more work.0