Canonical and Alternate Advice
-
At the moment for most of our sites, we have both a desktop and mobile version of our sites. They both show the same content and use the same URL structure as each other. The server determines whether if you're visiting from either device and displays the relevant version of the site.
We are in a predicament of how to properly use the canonical and alternate rel tags. Currently we have a canonical on mobile and alternate on desktop, both of which have the same URL because both mobile and desktop use the same as explained in the first paragraph.
Would the way of us doing it at the moment be correct?
-
That would normally be the case but not tonight.
LOL, I am picking up a lot of the UK Q&A I will be at BrightonSEO and search love London if any of you guys will be in the area I'd love to grab a pint?
sincerely,
Thomas
-
The reason we answered 'quickly' by the way is because we are in the UK - you were still in bed lol!
-
There is only ONE URL that is the point.
If they share the same URL then you only have one page of code so ONE canonical
Regards
Nigel
-
Sorry Nigel
was not trying to make this more complicated was just trying to make sure that we were all on the same page.
FYI if you need a method of adding the rel canonical to your website quickly you can use Google tag manager or if you want to add to the header
https://support.stackpath.com/hc/en-us/articles/360001445283-EdgeRules-Adding-a-Canonical-Header
-
So a self referencing canonical on both mobile and desktop versions of the site, regardless if they chuck out two version with the same content?
-
Hi JH
I'm sure Thomas means well with his multiple complicated posts but all of this is totally unnecessary.
Both sites are serving the same URL
You can't put a rel=alternative because there is nothing to point to.
Just put a self-referencing canonical. I said that 2 hours ago!
That is all.
Regards Nigel
-
Use a self-referencing canonical
https://blog.seoprofiler.com/google-recommend-self-referencing-canonical-tags/
Please let me know if you want me to remove the image below?
you can use this one if needed http://bseo.io/c1vMSv
-
I've been told to pass on a URL, thanks for your help Thomas!
-
Hey man I understand is a big deal
could you do me a huge favor and run your site through screaming frog SEO spider send me a couple of pages with the domains whited out so I can tell you 100% what to do in this situation because I am basing this on what you have told me and honestly I would like to look at what a tool can show me and that will tell me what I need to do.
Or you can tell me if the mobile version of the site hit's Google's index yes or no?
respectfully,
Tom
-
So both mobile and desktop require a self referencing canonical(in both headers)?
Sorry for the questions, just need to make sure! It's a very touchy subject!
-
The single self-referencing URL will work.
-
What URLs are you using with the “alternate” tag on?
You said
”1. We have multiple brand sites, that have a similar setup. They all have mobile and desktop versions of the sites running on the same URL, both of which show the same content.2. The server determines whether if you're on a desktop or mobile devices using the header information, and points the user to the site relevant files for the given device.”
thats Dynamic serving same URL
Dynamic serving is a setup where the server responds with different HTML (and CSS) on the same URL depending on which user agent requests the page (mobile, tablet, or desktop).
that would NOT give you the mobile or m.example.com & www.example.com different URLs
**But If you do have a different m.example.com & www.example.com URLs you should use this code or XML site maps **
for different URLs use this:
Annotations in the HTML
On the desktop page (http://www.example.com/page-1), add the following annotation:
<linkrel="alternate"media="only screen="" and="" (max-width:="" 640px)"<="" span="">href="http://m.example.com/page-1"></linkrel="alternate"media="only>
On the mobile page (http://m.example.com/page-1), the required annotation should be:
<linkrel="canonical"href="http: www.example.com="" page-1"=""></linkrel="canonical"href="http:>
This rel="canonical" tag on the mobile URL pointing to the desktop page is required.
Or
Annotations in sitemaps
We support including the rel="alternate"annotation for the desktop pages in sitemaps like this:
<urlsetxmlns="http: www.sitemaps.org="" schemas="" sitemap="" 0.9"<="" span="">xmlns:xhtml="http://www.w3.org/1999/xhtml">
<loc>http://www.example.com/page-1/</loc>
<xhtml:linkrel="alternate"media="only screen="" and="" (max-width:="" 640px)"<="" span="">href="http://m.example.com/page-1"/></xhtml:linkrel="alternate"media="only></urlsetxmlns="http:>You should have the same URL on mobile and desktop
You should have the same rel canonical tag on your URLs unless and this is a big unless you're talking about using Google AMP?
If the URL you want to be indexed is the same URL point everything to that URL if that makes it easier to understand.
respectfully,
Tom
-
Just to confirm, are we suppose to have a canonical on desktop and mobile or just desktop?
This would mean removing the alternate?
Want to confirm everything before iterating this across to others.
We are not using AMP, just a standard site setup.
-
Unless you are using AMP?
Then you would add
Linking pages with
In order to solve this problem, we add information about the AMP page to the non-AMP page and vice versa, in the form of tags in the .
Add the following to the non-AMP page:
<link rel="amphtml" href="https://www.example.com/url/to/amp/document.html">
And this to the AMP page:
<link rel="canonical" href="https://www.example.com/url/to/full/document.html">
are you using AMP pages?
https://support.google.com/webmasters/answer/139066?hl=en
https://www.ampproject.org/docs/fundamentals/discovery
I hope that helps you if not please let me know.
Respectfully,
Tom
-
Cool, that's what I thought when I heard your description I just wanted to be very thorough because sometimes you get very little information and I appreciate you letting me know that.
dynamic serving URLs are identical to each other so you should have a self-referencing canonical tag because the URL does not change the real canonical tag just decides what should be in the index and the same URL.
You're Rel canonical should be something like this example below
Example URL https://www.example.com/example-url/
because the end URL is the same and URL that you want to be indexed in Google you want to be certain that you have a self-referencing URL to prevent query strings and other things like that and you do not need to point a URL to an identical URL you just need a self-referencing canonical if that makes sense.
See: https://yoast.com/rel-canonical/
I hope that is of help,
Tom
-
Hi,
I can't give off too much information as it's not my call, but I can answer your questions without mentioning the brands.
1. We have multiple brand sites, that have a similar setup. They all have mobile and desktop versions of the sites running on the same URL, both of which show the same content.
2. The server determines whether if you're on a desktop or mobile devices using the header information, and points the user to the site relevant files for the given device.
3. Our sites would quite clearly fit in the dynamic serving category.
We have 301 redirects on none www to www and http to https.
-
This is the correct solution!
-
The URLs are identical it is just the content that is served that may be slightly different.
Since you can only specify one canonical for each URL it makes no difference. Just self-reference and that is it.
If you had to different URLs then it would be an issue where you woudl need a rel=alternative so there is nothing to worry about.
Regards
Nigel
-
You guys are fast I was going to answer this and had to do some other things but let me weigh in on couple things.
as you said
“We are in a predicament of how to properly use the canonical and alternate rel tags**. Currently we have a canonical on mobile and alternate on desktop, both of which have the same URL because both mobile and desktop use the same as explained in the first paragraph.”**
so what you’re saying is that you have a dynamic site so you don’t need to add “alternate"media” tags to the site.
https://developers.google.com/search/mobile-sites/mobile-seo/dynamic-serving
As it is not immediately apparent in this setup that the site alters the HTML for mobile user agents (the mobile content is "hidden" when crawled with a desktop user agent), it’s recommend that the server send a hint to request that Googlebot for smartphones also crawl the page, and thus discover the mobile content. This hint is implemented using the Vary HTTP header.
**you don’t need this **
Annotations in the HTML
On the desktop page (
http://www.example.com/page-1
), add the following annotation:<code dir="ltr"><linkrel="alternate"media="only screen="" and="" (max-width:="" 640px)"<br="">href="http://m.example.com/page-1"></linkrel="alternate"media="only></code>
On the mobile page (
http://m.example.com/page-1
), the required annotation should be:<code dir="ltr"><linkrel="canonical"href="http: www.example.com="" page-1"=""></linkrel="canonical"href="http:></code>
This
rel="canonical"
tag on the mobile URL pointing to the desktop page is required.Annotations in sitemaps
We support including the
rel="alternate"
annotation for the desktop pages in sitemaps like this:<code dir="ltr"><urlsetxmlns="http: www.sitemaps.org="" schemas="" sitemap="" 0.9"<br="">xmlns:xhtml="http://www.w3.org/1999/xhtml"> <url><loc>http://www.example.com/page-1/</loc> <xhtml:linkrel="alternate"media="only screen="" and="" (max-width:="" 640px)"<br="">href="http://m.example.com/page-1"/></xhtml:linkrel="alternate"media="only></url></urlsetxmlns="http:></code>
The required
rel="canonical"
tag on the mobile URL should still be added to the mobile page's HTML.**to be sure **
Are you willing to share your domain with us? Or one domain?
-
We're talking about multiple websites that all have the identical site structure or at least mobile and desktop site structure?
-
Your server is making the change for you?
-
Would you be kind enough to install this plug-in on chrome in order for you to show a couple examples of the canonical and the URL?
- https://chrome.google.com/webstore/detail/portents-seo-page-review/babgchcegnkbiojmdpnoilficladccfm?hl=en-US
- https://chrome.google.com/webstore/detail/link-redirect-trace/nnpljppamoaalgkieeciijbcccohlpoh?hl=en
In addition, would you be kind enough to run your site through the two tools here ( 100% free and very easy to use)
If you would not mind doing this and sending screenshots it would mean a lot to us and getting your canonical's straightened out.
screenshots https://snag.gy/ then upload to http://imgur.com/
everything is on the same server I'm assuming?
Of the three below how would you categorize your site?
- https://developers.google.com/search/mobile-sites/mobile-seo/separate-urls
- https://developers.google.com/search/mobile-sites/mobile-seo/dynamic-serving
- https://developers.google.com/search/mobile-sites/mobile-seo/responsive-design
Respectfully,
Tom
-
-
Would this mean we need canonical only on desktop or mobile site?
-
You are right - you could only use teh rel=alternate if there was an m. version or similar
Regards
Nigel
-
The self referencing canonical advice was solid and I 100% agree with it. The rel=alternate advice, I felt would cause problems (IMO). But as we all know, fiddly issues like this are highly subjective
-
Then there is no problem simply putting a self-referencing canonical. There is in effect no mobile version as there is a single URL so no need for a rel=alternate.
It's an even easier solution. Well, there isn't a problem in the first place.
rel=alternate is only necessary if you have two different URLs! The fact they are the same takes away the problem.
Regards
Nigel
-
Your problem is that you have two different sites loading on the same URL. If you are returning both the mobile and desktop / laptop site on the same URL, you would be expected to be using responsive design. In-fact, you may have re-invented another different way to implement responsive design which is probably, slightly less fluid yet slightly more efficient :')
Since your mobile and desktop pages both reside on exactly the same URL, I'd test the page(s) with this tool (the mobile friendly tool) and this tool (the page-speed insights tool). If Google correctly views your site as mobile friendly, and if within PageSpeed insights Google is correctly differentiating between the mobile and desktop site versions (check the mobile and desktop tabs) then both URLs should canonical to themselves (self referencing canonical) and no alternate tag should be used or deployed. Google will misread the alternate tag, which points to itself - as an error. That tag is to be used when your separate mobile site (page) exists on a separate URL, like an 'm.' subdomain or something like that
Imagine you are Googlebot. You are crawling in desktop mode, load the desktop URL version and find that the page says, it (itself) is also the mobile page. You'd get really confused
Check to see whether your implementation is even supported by Google using the tools I linked you to. If it is, then just use self referencing canonical tags and do not deploy alternate tags (which would make no sense, since both versions of the site are on the same URL). When people build responsive sites (same source code on the same URL, but it's adaptive CSS which re-organises the contents of the page based upon viewport widths) - they don't use alternate tags, only canonicals
Since your situation is more similar to responsive design (from a crawling perspective) than it is to separate mobile site design, drop the alt
-
The problem with this is, where you say "corresponding mobile URL" - there isn't one as OP has stated that, two different source codes (pages) can be rendered on the same URL depending upon the user's screen size / user-agent (however they are detecting mobile, and serving different pages)
-
Hi JH
This is very straightforward.
Use the following annotations:
- On the desktop page, add a rel=”alternate” tag pointing to the corresponding mobile URL. This helps Googlebot discover the location of your site’s mobile pages.
- On the mobile page, add a link rel=”canonical” tag pointing to the corresponding desktop URL.
It is that simple and doing this will not create duplicate content
More here: https://developers.google.com/search/mobile-sites/mobile-seo/separate-urls
Regards Nigel
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
-
This one is complicated... canonicals, href lang tags and no index
Bear with me, this is complicated (I REALLY hope one of you comes along and says, no it isn't!) Scenario A client has multiple english pages, as they have a unique product offering in AUS, US, UK, NZ and also have a global site in english. Obviously there is a lot of duplicate content and they have the relevant href lang tags set-up to help Google untangle what should be ranked where. They also have rel-canonical on each page. I've set-up search console for each of the folder structures, i.e. en-us, en-gb, en-au and so on. They have an optimised page for one of their primary keywords, which ranks nowhere for this exact keyword, but this page DOES rank for 40 similar keywords. For the exact keyword, they rank 52nd, and frustratingly, it's the homepage that ranks. We know the correct page is ranking and is indexed because search console tells us so and we see the exact page appear in SERPs for the other 40 keywords. When I look at the en-us site in Search Console, it tells me that the home page is not being indexed, because a rel canonical tag is prioritising an alternative page (probably the global site) - however, the en-us homepage is showing up in rankings for a lot of their important keywords. The site has been live for 6 months and the optimised page for about 3 months. Questions 1. If search console is saying the homepage is not ranking, how is it showing up in SERPs?
Intermediate & Advanced SEO | | Algorhythm_jT
2. Why is the homepage ranking for this important keyword, when there is virtually no mention of the keyword versus the page that is almost perfect according to Moz's on-page grader?
3. Do you need href lang tags AND rel canonical on a page?
4. How long before a new page that is optimised for a keyword take to replace (and hopefully surpass) the homepage?
5. If the US is the most important market, should we guide Google to that fact using rel-canonical? Really appreciate your feedback, hivemind. Thanks0 -
Screaming frog Advice
Hi I am trying to crawl my site and it keeps crashing. My sys admins keeps upgrading the virtual box it sits on and it now currently has 8GB of memory, but still crashes. It gets to around 200k pages crawl and dies. Any tips on how I can crawl my whole site, can u use screaming frog to crawl part of a site. Thanks in advance for any tips. Andy
Intermediate & Advanced SEO | | Andy-Halliday0 -
Block in robots.txt instead of using canonical?
When I use a canonical tag for pages that are variations of the same page, it basically means that I don't want Google to index this page. But at the same time, spiders will go ahead and crawl the page. Isn't this a waste of my crawl budget? Wouldn't it be better to just disallow the page in robots.txt and let Google focus on crawling the pages that I do want indexed? In other words, why should I ever use rel=canonical as opposed to simply disallowing in robots.txt?
Intermediate & Advanced SEO | | YairSpolter0 -
Unpaid Followed Links & Canonical Links from Syndicated Content
I have a user of our syndicated content linking to our detailed source content. The content is being used across a set of related sites and driving good quality traffic. The issue is how they link and what it looks like. We have tens of thousands of new links showing up from more than a dozen domains, hundreds of sub-domains, but all coming from the same IP. The growth rate is exponential. The implementation was supposed to have canonical tags so Google could properly interpret the owner and not have duplicate syndicated content potentially outranking the source. The canonical are links are missing and the links to us are followed. While the links are not paid for, it looks bad to me. I have asked the vendor to no-follow the links and implement the agreed upon canonical tag. We have no warnings from Google, but I want to head that off and do the right thing. Is this the right approach? What would do and what would you you do while waiting on the site owner to make the fixes to reduce the possibility of penguin/google concerns? Blair
Intermediate & Advanced SEO | | BlairKuhnen0 -
Rel Canonical attribute order
So the position of the attribute effect the rel canonical tags' ability to function? is the way I see it across multiple documents and websites. Having a discussion with someone in the office and there is a website with it set up as: Will that cause any problems? The website is inquestion still has both pages indexed within Google using the SITE:domain.com/product as well as SITE:domain.com/category/product
Intermediate & Advanced SEO | | jasondexter0 -
High level rel=canonical conceptual question
Hi community. Your advice and perspective is greatly appreciated. We are doing a site replatform and I fear that serious SEO fundamentals were overlooked and I am not getting straight answers to a simple question: How are we communicating to search engines the single URL we want indexed? Backstory: Current site has major duplicate content issues. Rel-canonical is not used. There are currently 2 versions of every category and product detail page. Both are indexed in certain instances. A 60 page audit has recommends rel=canonical at least 10 times for the similar situations an ecommerce site has with dupe urls/content. New site: We are rolling out 2 URLS AGAIN!!! URL A is an internal URL generated by the systerm. We have developed this fancy dynamic sitemap generator which looks/maps to URL A and creates a SEO optimized URL that I call URL B. URL B is then inserted into the site map and the sitemap is communicated externally to google. URL B does an internal 301 redirect back to URL A...so in an essence, the URL a customer sees is not the same as what we want google to see. I still think there is potential for duplicate indexing. What do you think? Is rel=canonical the answer? In my research on this site, past projects and google I think the correct solution is this on each customer facing category and pdp: The head section (With the optimized Meta Title and Meta Description) needs to have the rel-canonical pointing to URL B
Intermediate & Advanced SEO | | mm916157
example of the meta area of URL A: What do you think? I am open to all ideas and I can provide more details if needed.0 -
Is it ok to add rel=CANONICAL into the desktop version on top of the rel="alternate" Tag (Mobile vs Desktop version)
Hi mozzers, We launched a mobile site a couples months ago following the parallel mobile structure with a URL:m.example.com The week later my moz crawl detected thousands of dups which I resolved by implementing canonical tags on the mobile version and rel=alternate onto the desktop version. The problem here is that I still also got Dups from that got generated by the CMS. ?device=mobile ?device=desktop One of the options to resolve those is to add canonicals on the desktop versions as well on top of the rel=alternate tag we just implemented. So my question here: is it dangerous to add rel=canonical and rel=alternate tags on the desktop version of the site or not? will it disrupt the rel=canonical on mobile? Thanks
Intermediate & Advanced SEO | | Ideas-Money-Art0 -
Will the Canonical tag fix this issue?
I recently joined promoz and I've been busy working through the issues raised brought to light during the crawls of our Magento site, www.unitedbmwonline.com. One of many issues were the 10,000+ Duplicate Page Titles which I believe are the result of not using Canonical tags when setting up the store. This is now corrected and hopefully I'll see a significant drop in this value after this next crawl. Am I correct in this assumption? Cheers, Steve
Intermediate & Advanced SEO | | SteveMaguire0