It’s been over four years (February 2009) since Google and Yahoo announced support for the rel=canonical tag, and yet this single line of HTML is still causing a lot of confusion for SEOs and webmasters. Recently, Google posted 5 common mistakes with rel=canonical – it’s a good post and a welcome bit of transparency, but it doesn’t address a lot of the questions we see daily here in Q&A. So, I thought it was a good time to tackle some of your most common questions (and please forgive the following nonsense)....
What Is Rel=Canonical?
Put simply, the rel=canonical tag is a way to tell Google that one URL is equivalent to another URL, for search purposes. Typically, a URL (B) is a duplicate of URL (A), and the canonical tag points to (A). The following tag would appear on the page that generates URL (B), in the <head></head>:
<link rel=”canonical” href=”https://www.example.com/url-a.html” />
Google’s support document on rel=canonical is actually pretty good. The subject of duplicate content is complex, and I’ve addressed it previously in detail. For this post, I’m going to skip ahead and assume that you have a working knowledge of technical SEO and have attempted to use rel=canonical on your site.
Note: Rel=canonical is also referred to as “Rel-canonical” and “The Canonical Tag”. For this article, I will try to consistently refer to it as “Rel=canonical”.
(1) Should I Use Rel=Canonical for Pagination?
I’m not going to repeat all of Google’s answers, but this one is so frequently asked that it deserves more detail. Let’s say you have a series of paginated search results (1, 2, 3… n). These can be considered “thin”, from a search standpoint, so should you rel=canonical page n back to page 1?
Officially, the answer is “no” – Google does not recommend this. They recommend that you either rel=canonical to a “View All” page (if having all results on one page is viable) or that you use rel=prev/next. Rel=canonical can be used in conjunction with rel=prev/next to handle search sorts, filters, etc., but that gets complicated fast.
Pagination for SEO is a very tricky subject, and I recommend you check out these two resources:
- Conquering Pagination – A Guide to Consolidating your Content
- The Latest & Greatest On SEO Pagination
(2) Can I Use Rel=Canonical Cross-domain?
Yes – in late 2009, Google announced support for cross-domain use of rel=canonical. This is typically for syndicated content, when you’re concerned about duplication and only want one version of the content to be eligible for ranking.
(3) Should I Use Rel=Canonical Cross-Domain?
That’s a tougher question. First off, Google may choose to ignore cross-domain use of rel=canonical if the pages seem too different or it appears manipulative. The ideal use of cross-domain rel=canonical would be a situation where multiple sites owned by the same entity share content, and that content is useful to the users of each individual site. In that case, you probably wouldn’t want to use 301-redirects (it could confuse users and harm the individual brands), but you may want to avoid duplicate content issues and control which property Google displays in search results. I would not typically use rel=canonical cross-domain just to consolidate PageRank.
(4) Should I Use Rel=Canonical on Near Duplicates?
As my catastrophic canonicalization experiment and follow-up experiments showed, Google does honor rel=canonical even on very different pages, in some cases. That doesn’t mean that it’s a good idea. Generally speaking, I think it’s best to reserve rel=canonical for duplicates or very-near duplicates. For example, if a product pages spins off into five URLs for five different colors, and each color's page only differs by a sentence or two (or an image), then yes, I think it’s fine to rel=canonical to the “parent” product page.
Do not use rel=canonical as a substitute for appropriate 301-redirects and/or 404s. While it probably won’t cause large-scale catastrophes, I strongly suspect that Google will start to ignore your canonical tags, and this may impact how you control legitimate duplicates.
(5) Can I Put Rel=Canonical on the Canonical Page?
In other words, is it alright to put a rel=canonical tag on the canonical version of the URL, pointing back to itself? Practically speaking – yes, it is, but you don't have to. Early on, there were hints that both Google and Bing preferred that you not overuse rel=canonical. Over time, though, their stances seemed to soften, and I’ve seen no evidence in recent history of a properly used, self-referencing canonical causing any harm.
This is often just a practical issue – many URLs share common templates, and the code needed to display a rel=canonical tag on just the duplicates and not the canonical version of a page can get messy and increase your chance of mistakes. Personally, I believe that the search engines recognized the reality most webmasters face and adjusted their initial, conservative stance.
(6) Is It OK to Put Rel=Canonical on My Entire Site?
Should you pre-emptively rel=canonical your entire site – even if many of the pages aren’t subject to duplicate content issues? I think this gets very speculative. We have recommended this approach at SEOmoz in the past, and I think it’s generally safe. I do worry that excessive use of rel=canonical could cause search engines to devalue and even ignore those tags, but I can’t point to any clear evidence of this happening. I also worry that people often implement site-wide rel=canonical tags badly, and end up pointing them to the wrong pages.
I do think that a pre-emptive rel=canonical on your home-page is generally a good ideas, as home pages are prone to URL variations. In a perfect world, I’d say to use rel=canonical on the home-page, known duplicates, and any pages with parameters that could drive duplicate content, and leave the rest alone. However, that’s often a very difficult procedure. In some cases, site-wide rel=canonical implementation is better than no index control.
(7) Should I Use Rel=Canonical or 301 Redirects?
Please understand that while these two approaches can behave similarly, from an SEO standpoint, they are not interchangeable. Here’s the critical difference – a 301-redirect takes the visitor to the canonical URL, while a rel=canonical tag does not. Usually, only one of these approaches is the right one for your visitors. If you really want to permanently consolidate two pages and remove the duplicates, then use a 301-redirect. If you want to keep both pages available to visitors, but only have one appear in search results, then use rel=canonical.
(8) Does Rel=Canonical Pass Authority/PageRank?
This is very difficult to measure, but if you use rel=canonical appropriately, and if Google honors it, then it appears to act similarly to a 301-redirect. We suspect it passes authority/PageRank for links to the non-canonical URL, with some small amount of loss (similar to a 301).
(9) Can I Chain Rel=Canonicals (+301s, 302s, etc.)?
What happens if you rel=canonical to a URL with rel=canonical to another URL, or you rel=canonical to a URL that 301-redirects to another URL? It gets complicated. In some cases, it might work and it might even pass PageRank. Generally speaking, though, it’s a bad idea. At best, it’s sloppy. At worst, it might not function at all, or you might lose significant PageRank across the chain. Wherever possible, avoid chains and implement rel=canonical in a single hop.
(10) Are Non-Canonical Pages Indexed?
For all practical purposes – no. If Google honors a rel=canonical tag, then the non-canonical page is not eligible for ranking. It will not have a unique cached copy, and it will not appear in the public index via a “site:” search. Now, does Google maintain a record of the non-canonical URL? I assume they do. As an SEO, though, the non-canonical URL ceases to exist in any meaningful way.
(11) Can Someone Else Rel=Canonical My Pages?
I’ve seen occasional worries about someone using rel=canonical, especially cross-domain, to harm a site or steal its authority. Keep in mind that you can only grant canonical status from pages you control. So, you could rel=canonical all of your pages to someone else’s site, but why would anyone do that? To wreak any real havoc, someone would have to hack into your site. If that happens, then rel=canonical abuse is the least of your problems. The vast, vast majority of damage done by rel=canonical is self-inflicted.
(12) Can I Have My Cake and Eat It, Too?
No. Yeah, I know – you don’t want to hear it. At least a third of the questions we get about rel=canonical boil down to “I want all of these pages to rank, and they’re the same, but I don’t want to get in any trouble for duplicate content!” I don’t have any secret sauce to pour on that.
You don’t have to use rel=canonical, but. in my experience. controlling your own duplicate content is better than having Google do it for you, and eventually they’ll do it for you. In the old days, that might just mean that the wrong page got filtered out. After 25+ Panda updates, though, it could mean that your entire site suffers. You can’t have it both ways – if you have duplicate content, then remove it, control it, or improve it.
What Questions Do You Have?
If you have any general questions about the canonical tag or how to use it, feel free to leave a comment, and I’ll try to address them. Please understand that I can’t dig into your site and provide consulting-level services, but if you can ask the question in a general way that will be helpful to others, I’ll do my best to leave a reply.
Hi Pete,
great post as usual, but I'd like to add one more question about rel="canonical", which I'm asked a lot, also in here in the Q&A:
How to use rel="canonical" in conjuction with rel="alternate" hreflang?
The reason of this question is due to an incorrect statement Google itself did when it first launched this mark-up, which caused problems as SERPs snippets showing the Title of the Canonical URL and the URL of the Alternate via hreflang one.
So, the answer is: if you have implemented the rel="alternate" hreflang mark up in an URL, then the rel="canonical" of that same URL must be self-referantial
Example:
The pages www.domain.com/products and www.domain.com/uk/products are targeting the first USA and the latter UK, but have the same content, so that we should implement the rel="alternate" hreflang mark up.
We should do so in the case of the URL targeting searches in English from the USA:
<rink rel="alternate" hreflang="en-US" href="https://www.domain.com/products" />
<link rel="alternate" hreflang="en-UK" href="https://www.domain.com/uk/products" />
<link rel="canonical" href="http:://www.domain.com/products" />
And we should do so in the case of the URL targeting searches in English from the UK:
<link rel="alternate" hreflang="en-UK" href="https://www.domain.com/uk/products" />
<link rel="alternate" hreflang="en-US" href="https://www.domain.com/products" />
<link rel="canonical" href="http:://www.domain.com/uk/products" />
Obviously, this is the simplest situation, because what rel="canonical" to use can also depend on other factors, which you have described here and in previous posts.
That's a very good point, though, after accidentally setting things up this way I found it was quite helpful. Now on most sites with hreflang we let them share a canonical unless the meta title has changed.
If you use a canonical and have different meta titles it will always display the canonical's title, as you say. But, if you don't have different titles (which is quite common on UK/ US sites) it doesn't matter that it shows the canonical's title as it's the same thing.
So, by letting the 2 versions share a canonical you can share link equity between the 2 URLs, whilst hreflang still inserts the relevant local version into the results.
Whilst I discovered this accidentally not knowing the title would carry across (I first implemented hreflang about 3 days after it was released), if the title doesn't matter, it's really effective.
Thanks, Gianluca - I almost added a question about hreflang, but it's been such a mess that I decided to avoid it (and, honestly, I don't do that much international SEO, so I may flub it up). Plus, now we've got the new "default" option, which nobody has really had time to test very well. It's really no surprise people are confused.
Hi Gianluca,
I was doing some research on Internationalization SEO and found a very helpful link : https://sites.google.com/site/webmasterhelpforum/en/faq-internationalisation It has the FAQ on Internationalization. It does cleared many of my question hope this helps you too.
Specially check out the last 2 question of Rel Alternate Hreflang Section
Hi Sir,
I am doing seo of my office website. I use rel canonical for http to https redirection. It has some issue after this. My https://www.akclinics.com don't get ranked on serp. After doing a lots of good work. What is the main reason behind this. Kindly help me out.I am in big pressure by boss.
Hi Pete, thanks for this awesome post
I'm working with a real estate website and the property search page creating duplicate content issues
For example:
www.example.com/search.php?searchtype=BUY&PLACE-NAME (Search for Property A, BUY)
www.example.com/search.php?searchtype=RENT&PLACE-NAME (Search for Property A, RENT)
The only change in the content here is the 'price', because one search is for ‘RENT’ and other is for ‘BUY’
My Question
Did I needed to implement a canonical tag here, and if so which is my canonical URL?
What about adding a canonical tag to the main search URL?
ie, <link rel=”canonical” href=”https://www.example.com/search.php” /> (Which will show the recent listed properties)
That's a really tough one, and some cases are very site-specific. On the one hand, once you multiple these out to 100s or 1000s of pages, you do run a very real risk of Google thinking your content is thin and even of Panda-related problems. On the other hand, these are perfectly legitimate pages for end users.
A couple of options, in addition to doing nothing and hoping for the best:
(1) Rel=canonical to whichever version is most important for your business, with clear paths to the other version. That way, visitors can view both, but you've controlled the duplicates.
(2) Combine the information on one page. If you're going to rel=canonical anyway, you won't rank for both, and the user experience may actually be improved. If there's no real content difference and it's just a price, this may be the better option.
(3) Generate the buy vs. rent price dynamically, so that Google only sees one version, but users see a version based on how they navigate the site (probably with cookies/sessions). I suspect, without seeing your site, that that could be confusing.
I'd probably make a case for (2), but it really depends a lot on your business model and site structure.
Going with (2) for now, option 3 is also good but I guess it will not work with our case.
Thank you Pete :)
Thanks Dr. Pete. This is really informative! Rel-Canonical is a simple definition but the implementation can be tricky depending on the uniqueness of the scenario.
Recently I came across a situation where one of my client's had his listing placed in one of the niche specific .org site that had pagination. It was a good, authority link for his site but unfortunately, I found it was of no value just because of the way they had implemented the canonical tag.
I checked the source of the page number where the listing was present and found that they had rel=canonical all the pages 1,2,3,4...n to Page 1.
So Google never indexed those inner pages and only the one's listed on Page 1 got all the juice.
In my opinion, it's important to verify Rel-canonical not only On page, but also while auditing your site's back link profile.
Definitely a good point - if you have a back-link on a page with rel=canonical pointing to another page (without the link), then you basically don't have a link, for SEO purposes.
I agree with You, SEO girl, but i wana add Rel=Canonical on my new page but little bit confused also is it appropriate or not, what do you think, here is the concerning page of Umrah Packages
https://umrahexperts.co.uk/umrah-packages.html
It may be important to add that WordPress puts the "Rel=Canonical on the Canonical Page" on every page per default.
Good point - many CRMs have their own default behavior, so you have to be careful before you start modifying it. Do it wrong, and you could end up pointing your whole site to one page, or removing rel=canonicals that had a purpose and were helping.
I wonder which canonical Google would honor (if at all) in the event that there are two canonical tags unintentionally placed on a page. I've seen people screw things up by doubling up on plugins in their CMS (for example, having both "nofollow, noindex" and "follow, index" on the same page), and then Google goes with the more restrictive of the two. My guess would be they would wind up ignoring both canonicals?
Google's recent post suggests that, if there's more than one rel=canonical tag on a page, they'll ignore them both. I swear I've seen situations where they picked one, though. In either case, it's open to their interpretation and generally a bad idea. Mixing a rel=canonical and a NOINDEX can make a mess, but my gut feeling is that NOINDEX still usually wins.
I've done this too on a site where we had regular syndicated content and were required to canonicalise it back to the source site. We added one for that in the code, then accidentally had Yoast add a second.
As soon as that happened you could find both URLs in the index, so would seem to confirm that Google do ignore multiple canonicals. Once fixed the duplicate pages on our site dropped out of the index again.
Great post, thanks.
I'm curious about your take on pre-emptively adding the rel="canonical" tag on your entire site - specifically for WordPress users who have the WordPress SEO by Yoast installed. By default, this enables rel="canonical" tags on every page. Could there be any potential for negative effects for brand new sites? Your thoughts?
Does the yoast plugin add canonical tags to every page by default? I thought otherwise. Wordpress itself adds canonical tags to every page by default. If you're going to use the Yoast plugin for WP, you need to comment out the canonical line in your includes file.
At least for now, in 2013, I generally think it's safe. When you have a structured add-on like Joost's plug-in, it's designed to handle the problems and having it is better than not having it in the vast majority of cases. You just have to tread carefully when you go to make changes.
I think it's possible that, down the road, Google could start to view site-wide canonical as a slightly negative signal, but I think it's like 301s. In other words, if you 301 100s of pages to your home-page or start to do odd things, it might look manipulative, and Google could raise an eyebrow. For most people, though, 301s are useful and effective and 100% white-hat. I see rel=canonical the same way. Don't abuse it, but liberal, legitimate use is ok.
Our site used templates created in Wordpress and we didn't notice that WP included a link rel='canonical' tag leading to a nonexistent page on every page, and as a result most pages of our site were not indexed, at all. Even after fixing it, Google and other search engines seem to remember that there is a canonical command there and nothing is indexing. This command is extraordinarily dangerous.
https://unheardbeethoven.org
Thank god! Somewhere i can go and refresh my memory every time i need this..
You didn't mention syndication! I consider this very very important for publishing sites, particularly when trying to show up as the source, and rank in Google News results (both news search and universal). I believe it is still in Google's best practice suggestions to use cross-domain canonicals when others are publishing your content. Is it?
Of course, getting publishers to canonical to your page is pretty difficult. Most sites will at least feign "impossible technical implementation" even if they do understand what you are asking for. That means that suggesting "syndication-source" as an alternate doesn't make headway either. On the bright side this usually makes negotiating for a link back to the original article on your site a tad easier, which is probably more valuable anyway.
Interested in further input on canonicals for syndication requirements and any tips on how to actually get it implemented. Barring them your content often hurts you as much as them, as the syndication offers impression value, but still you don't want to be outranked for your own content.
Cross-domain rel=canonical can get tricky, so I didn't dig too deep. You're right, though - if you control the syndicated content, it is still a good bet, and stronger than just a link. Unfortunately, as you said, getting buy-in can be tough. I know major newspapers that have used cross-domain canonical tags with quite a bit of success.
Unfortunately, syndication-source not only was a weak signal, but it was apparently (very quietly) deprecated by Google last year:
https://productforums.google.com/forum/?fromgroups#!topic/news/Bj2lzoy45Rw%5B1-25-false%5D
Syndication is a tough topic, and your next best bet is just having a solid linking structure back to the source and making sure the source is crawled and indexed before the duplicates.
Great informative post. Rel="canonical" is really only a band aid and not a cure.
I think there is one legitimate question you may have overlooked. What is the impact of many to one canonicals. That is, many pages (considering 2 scenarios that the pointing pages are near duplicates and when they are not duplicates) pointing to 1 page. But, a good ready reckoner by Dr Pete again :)
If they're "true" duplicates, I think it's fine. For example, let's say that you have a link that generates a unique ID for every visitor (some kind of registration form, for example) - using rel=canonical to point that to the main form is fine, and probably a good bet. You could also just META NOINDEX that form, as it probably has no search value. Frankly, you may just want to change your approach altogether and not use a URL parameter. In cases where these pages have already been created and are causing a problem, I've found rel=canonical to be a good solution.
If you're trying to use rel=canonical like a 301-redirect, to point a bunch of your pages to one page (like the home-page), I think that's generally a bad idea. A sensible combination of 301s and 404s is probably better in most of those cases.
For near duplicates, I understand that it maybe best to handle it through 301s and 404s, but take the case where you have a listing page with a bunch of sort options. Obviously, someone would want to just have view of this page to be indexed and to rank. Now, since the dataset is same for the listing page, just the ordering would differ, it would be sensible to use canonicals to handle this. Am I right in this observation?
Hi saibose, if you have the same dataset in one page, but sorting, ordering or filtering change from 1 Url to another, i recommend using Google url parameters in WMT. (assuming you're aiming Google rankings above all SE). The only requirement would be you have a normalized url for parameters (exple: domain.com/cataegory/param1=x¶m2=Y). There is a useful Google vid about the issue on youtube.
thanks
Surely this is more of a duplicate content issue than a rel canonical issue. If there are 2 sites with near-dupes but canonical-ing to the correct source then surely that's not really an issue? Unless I'm missing something?
This is a great post Dr. Pete! You have completely covered almost all the possibilities or confusions in dealing with "Rel=canonical".
I work in a franchise system where the all franchisee web properties are identical or near-identical. Would a rel=canonical strategy help us to get all local or franchisee pages indexed separately?
I have a new site going live in a few weeks time and I am working on 301 redirects. The current site is set up with canonicals.
The canonical tags on the older site has the following tag <link rel='canonical' href='https://example.com/' />
For some reasons the tag isn't <link rel="canonical" href="https://example.com/" />
Will this error in the tag have caused a split in authority between www and non www versions. The www version does hasn't been indexed in Google. Is it safe to say the canonical has worked.
Is it worth 301 redirecting www and non www versions to the new site urls just encase authority has been split? Please advise.
Just a follow-up - my post was shared on Google+, which led to a few comments by John Mueller. He's suggesting that a page with rel=canonical point to another page is indexed, and can show up in "site:" operator searches. It's unclear how Google is defining "indexed" - again, I'm sure it's crawled and stored, but it isn't part of the public index, cache, etc. My own experiences also show that, once Google honors rel=canonical, the non-canonical URL is dropped from the public index and doesn't appear under a "site:" command. So, I'm finding it difficult to reconcile John's comments with my own observations. For transparency's sake, I thought I should share the additional information.
John's comments:
"FWIW we can and do index pages that have a rel=canonical pointing to other URLs. It's not a server-side redirect where we never see content, it's not a noindex robots meta tag."
"Those pages can show up in site:-queries and normal rankings."
https://plus.google.com/u/0/+RickBucich/posts/Le21sQa3KwN
Is there any tool to run over the site to make sure the rel canonical is on the right pages? that would be very handy. I often find I am scratchings ones head if I should add or don't add a rel canonical
Unfortunately, not that I know of. It's very, very difficult to interpret whether a canonical is "right" for any given page. Our campaign management tools here at SEOmoz track rel=canonicals and can help you spot problems, but we don't have a good way to tell, except in extreme cases, if the canonicals are doing what you meant them to do. Screaming Frog (desktop spider software) also tracks canonicals, but it's the same situation - they'll help you spot-check and inventory.
Yes, Screaming Frog has a column that identifies the canonical urls pointing to each page of the scraped site.
The *technical* solution here would be combining a tool just like Screaming Frog to get the list of all the URLs with their respective canonicals and compare it one to each other (you can use PHP for this) to see if their contents match. If they do, there you have a high change of good rel=canonical, if the content match is below, let's say, 50% then there may be some wrong rel=canonical in your site. I don't know if I'm being clear on what I'm trying to explain here.
Super helpful, Pete, thanks.
So here's another question: For PPC landing pages I don't want indexed, as well as a noindex meta tag should I also use rel=canonical to the home page of the site? What about for A/B testing?
oh, that was two. =P
I was wondering the same thing about A/B testing. Should you add a canonical tag to the experimental pages so they point to the original/indexed page?
For PPC landing pages, unless they crossover into useful pages and might attract links, I'd probably just META NOINDEX them. They're really separate animals that generally should stay out of sight. It is a bit situational, though.
Some A/B testing tools manage duplicates or don't create separate URLs (which is ideal), but if you do spin out new URLs, then rel=canonical is definitely a good solution. That way, human visitors can see the test URLs, but Google will know to ignore them.
Thanks for clearing that up Dr. Pete. Great post! Learned a few good things that we'll start implementing.
Thanks Pete. I think that makes sense, it's the 'might attract links' thing which is key. If the PPC URLs start attracting links or shares I'll have an argument to do something other than NOINDEX them.
=)
Hello I am not tech savvy I need HELP!! I have duplicate content for my pages of my products noted in my webmaster tools area and I do not know how to add the rel=canonical to my template for my store. i asked my web designer and host i use and they dont have a clue what i am asking them to do. I need to know if anyone can help me here. I dont have the funds to pay someone to do it. My site is hurting alot i need help!! my website is posh birthday store com hope someone can help!
Stacy
Thanks for the informative post. I have run into a duplicate content issue and I cannot seem to wrap my head around the best way to fix it. Maybe someone on here can help. I cannot expose the url to the site per my client's request...but...
I am working on a drupal 7 site (yes I wish it was wordpress but it is not and switching it is not an option)
This organization has a few different products. They have a top-level product page where they are dynamically pulling snippets from the respective product pages to create the content. Because of this, it is causing a ton of duplicate content sitewide.
For example: product 1, product 2 and product 3 all have individual pages. There is a "Products" page that lists product 1, product 2, and product 3 snippet descriptions pulled from the top paragraph of each of the product pages respectively.
I would rather the individual product pages get the bulk of authority than the top-level "Products" page.
With the way drupal is set up (I am using the meta tag module and the global redirect module), If I put a canonical link in the product 1 page to cascade to the top level "Products" page and then do the same thing on product 2 page, the "Products" page will have more than 1 canonical link which will cause search engines to ignore the canonical all together.
Would it be better to put a canonical on all the product pages respectively to go back to the top-level "Products" page? If I do that, the "Products" page will have more authority than the individual product 1 page, product 2 page and so on.
There is much more content on each individual product on their respective pages. The top-level "Products" page only contains that 150 character snippet.
I hope this makes sense and someone can help. My head is spinning trying to figure this one out!
Getting a lot of specific questions about sites/problems in this post, which we can't really do justice to in blog comments on an older post. I'd highly recommend taking those to Moz Q&A, where the entire community may be able to help.
Dr.Peter, in the way you have explained things are really very easy to get for everyone. Appreciable ! Thanks .
Good post!
We have various redirects on our site that end up landing on the same page. Google has indexed some of these redirects instead of the landing page which we think is hurting some of our SEO (since we have inbound links only to the landing page).
If you do a Google search for 'carlos gonzalez razzball', you will see the top result has the URL (razzball.com/player/7287/) while the 4th result has the correct URL format (www.razzball.com/player/7287/Carlos+Gonzalez).
We recently added the canonical URL to the page but it doesn't appear to have rectified the situation.
Is this something that Google 'figures out' eventually? Did we implement the canonical URL correctly for that page?
Thanks in advance!
Scott
This appears to be a 302-redirect. If your goal is to consolidate the URLs, that generally won't help. Be wary of your internal links, too - sometimes people 301/canonical to their "preferred" URL, but then all of the internal and inbound (external) links go to another URL. Mixed signals can be a problem.
Thanks! We do have mixed signals in that regard to enable an easy shortcut for our writers in Wordpress. Example:
Today's post (https://razzball.com/sounds-like-drafters-cry/) links to Prince Fielder's page using a shortcut: https://razzball.com/player/Prince+Fielder/. This then 302 redirects to https://razzball.com/player/4613/Prince+Fielder/. I don't even know where that other variation (razzball.com/player/7287/) would exist anywhere.
All links outside of our site use the proper URL.
So should I switch this to a 301 redirect?
Hi Peter -
Per your advice, I switched to 301 redirects and Google did not update the URLs on the last crawl. If you do a search for "troy tulowitzki projections razzball", you'll see an example (URL is razzball.com/player/3531/). If you view the source of the cached page, you'll see the canonical URL suggestion on the page. When clicked, there's a 301 redirect to the URL noted in the canonical URL field.
Is this something that may correct in time or is there something else that's awry?
Scott
Hello...Google is indexing non existing sub-domains for my site. All are garbage sub-domains and they are destroying my main indexed pages. Will this work if place this tag in the header section of each web page?
tiumenx.optionsokanagan.com/
gudai.optionsokanagan.com/staff.php
ko.optionsokanagan.com/admissions.php
s35.optionsokanagan.com/staff.ph
pru.optionsokanagan.com/staff.php
https://id.optionsokanagan.com/
**** all garbage *********
Thanks
James
Thanks for the blog!
I think I have a unique case and hope you will able to advice me.
I have a website which sells website templates, I have links from a homepage to these templates.
Example of my homepage url: www.example.com
All templates Links are go to www.example-templates.com this domain has a permanent redirect to the first domain www.example-templates.com
In every website template the texts are nearly the same on all templates, sometimes it is slightly different, like headings... All texts mainly are lorem ipsum text. Everything else besides text is different, layouts, colors, images, design and etc...
Should I use canonicals urls? If yes which address should be canonical?
Examples of my urls:
www.example-templates.com/template-1
www.example-templates.com/template-2
www.example-templates.com/template-3
Thanks!
Alex
I am about to launch a rebrand of a website and I have unintentional rel=canonicals throughout the site. However, they are within a comment (i.e. <!--). How problematic is that going to be when we launch tomorrow?
Hello Mr.Peter
can you please tell me one thing actually i had posted same article contents on various article submission sites.And now is afraid of google's duplicate content issue....can you suggest me some ways that now i can take to correct my mistake...
What about for a syndicated site like Buzzfeed. How do i add the rel=connical tag for that?
I have a link from my web site to my Blogger blogspot as a sub domain.
On my blog I have cycle tours that I have done previously. I want to take them and create specific pages on my own web site, but I don't want to delete the blog entry. Neither do I want to be penalised for duplicate content.
I want the web page to be the main copy, but I cannot see a way of getting rid of the rel="canonical" on the blog entry.
How do I do this?
Thanks
Stephen
Hi Pete,
I have to share some content with some affilliates. This content is already in my site and it is original content. My affilliates will copy it and paste it at their websites. How can I avoid Google will consider this content as a duplicated content? Could I ask to my affilliates the use of the canonical cross domain pointing to my site?
Thanks in advance
I think I understand the canonical however I was wondering if someone could give me a few pointers to a certain situation I have.
So I have a client who offers a service across a wide range of countries.
He has a separate page for each country and there are certain elements that we cant change.
So there are giant chunks of text which are duplicated throughout all of the pages that we cant change.
Would implementing a canonical on one of the pages associate the templated content to that page and negate the internal duplication issue with other pages?
Hello,
can anybody confirm/explain to me following thoughts, please?
- canonical & utm parameters:
It is recommended to use rel=canonical in case of using links with utm parameters, to avoid duplicate pages. OK. Is it just me, that I cannot understand WHERE to put it? As it is being used in the head section of non-canonical page, and in these cases non-canonical pages don't really exist, in fact, just generated URL with parameters..
- canonical pointing to itself
Used for each page or by definition for some page that doesn't have any duplicate, it changes totally nothing, yes? First, the page would be indexed the same without being canonical to itself. Second, it says nothing of any other URLs (eventually duplicate to this one).
Thank you a lot in advance, I try to solve this out for really long time :-)
Since one template drives both the canonical and non-canonical URLs, you would put the canonical tag on that template. You could modify it to only show up when the non-canonical version is created, but that's not really necessary.
Keep in mind that, while it's one "page" to you, Google sees every URLs as a unique page. So, that one template is really driving dozens or hundreds of pages in their eyes.
Hello Peter, I got a little bit of confused about rel=canonical but after reading your post, I know now it's definition and when to use it.
Thank you for this info!
Does adding google analytics querystring or any querystring to a URL for a 301 redirect hurt SEO?
For example, does it hurt SEO to have www.oldsite.com redirect via a 301 redirect to www.new.com/?utm_source=oldsite.com&utm_medium=redirect&utm_campaign=campaign1
or should www.oldsite.com redirect via a 301 redirect to www.new.com without the querystring?
or if I use the querystring URL to redirect via the 301 redirect would I just put a
<link rel=”canonical” href=”https://www.example.com/url-a.html” /> on the homepage?
I would like to keep track of where sites are coming from via google analytics but I don't want to hurt SEO.
Thanks again!
Pete, I have a different problem. My portfolio lists all the apps with content like a blog post. When you click on the title, you will be taken to a separate URL for that app. I want the separate URL for that app to be indexed rather than having list of all the apps in one page. I want to make list as canonical for all the individual pages. Is there a way I can acheive this? Thanks in advance.
Sorry, I'm confused. Having the separate URLs indexed and making the list canonical are essentially in opposition to each other. Either the separate URLs are stand-alone pages in the index or the list is canonical, but you can't have it both ways. What are you trying to achieve, from a search and user perspective?
Hello Dr. Pete
Excellent article as always. I've read many articles about the canonical tag (on moz, google, and elsewhere), but I still have a question that I have not seen addressed anywhere else. I am trying to publish identical pages across different domains. I am OK with the duplicates not ranking, however I don't want any of the sites involved to be harmed. Here is an example:
www.my-site.com/myproduct
www.their-site.com/myproduct
www.another-site.com/myproduct
All those pages would be the same.
I want to have other sites (which are not owned or operated by myself) to have my content, and point to my site as the canonical source.
Everything I have read seems to indicate that this would be OK, but I want to make sure I am not misinterpreting something. It is a lot like syndication, but instead of articles it is product description on non-commerce sites.
Any input?
Thanks,
-Noe G
Dear Peter,
When I was working on a website, I found the same content in the website's index page
For eg:
www.abc.com is the main website
www.abc.com/index.html is also having the same content which is already present in home page.
I just want to know whether the search engines rankings may effect in this case.
Please reply me..
It's definitely worth having a canonical tag on that page pointing to the root (www.abc.com) version. Duplicates of the home-page are common and, while sometimes harmless, are well worth preventing.
Will the Search Engines see and abide by a change in my Canonical link? I have more than 1 page of similar content. I decide later I want a different page to be the canonical page, after I have set up the original one previously and pointed canonical links to it. Will the Engines go along with my changing my mind?
I know I am very late here but hoping for a reply of my question.
When I was checking source of my website, I found that there is a rel=canonical tag which has URL of same page. I am confused that is it okay to use, or I have to remove the rel=canonical tag.
Hi! I have a question about rel=canonical
Does the order of attributes in a rel=canonical tag matter?
I mean...it's better to use
<link rel="canonical" href="https://www.mysite.com/" />
or
<href="https://www.105.net/" link rel="canonical" />
thanks
Laura
Dr. Pete,
Fantastic article! I do have one question - do you have any thoughts or tips on how to manage the canonicals for extremely dynamic sites?
As the punchline to that old joke says: "Very carefully." :) It's really tough to speak generally to that, as it's so platform dependent. I think dynamic usage can be incredibly effective, especially for things like tags, product variations, etc. It can also be very dangerous, though, if done wrong. The devil is definitely in the details on that one.
How i know in my site rel=”canonical” is working Proper?
Dr. Pete,
Awesome article on the rel=canonical topic. I've been reading pretty much everything you've written (including comments) to better understand the subject.
Back in Sept 2012 you had made a comment here
https://moz.com/community/q/removing-dynamic-noindex-url-s-from-index about removing dynamic content from the index. You mentioned using the sitemap as an option. We have unfortunately been using robot.txt (because of bad SEO advice) to block Google from indexing sorts for our category pages. Unfortunately Google has stil indexed 10,000+ parameter string URL duplicates. We have since unblocked Gbot from those pages by removing it in the robots.txt file. We have now implemented rel=canonical tags for all of various sort/filters back to the canonical URL.
The issue now is we have removed the sorts/filter links from our site in fear of Google continuing to index the parameters and create additional duplicates. I know in your comment back in 2012 you said that it could be dangerous to open those links back up, but besides the WMT parameter option, is there another way? Also, would opening the sort/filters back up make it easier for Google to reindex and honor the rel=canonical?
I really appreciate any insight you could give. It's hard to get good advice and after reading a good chunk of what you've written it's clear you know what you're talking about. Thanks!
If you've got rel=canonical now on the sort/filter pages as your solution, but you've removed the links, you may find the removal process takes an incredibly long-time. I think it's pretty safe, in this case, to re-open the links just to help clear out the bad URLs. I'm not a huge fan of combining tactics, just because I think it's hard to tell what works, but Google has suggested that it's ok to use something like rel=canonical and WMT parameter handling.
I'll be honest - on any give complex site, no one is an authority but the people who monitor that site. Every best practice approach when it comes to duplicate content is just a starting point. With clients in the past, I've always had to monitor carefully and adjust along the way. My gut feeling is that, if you're not running into any huge ranking issues or major risks, re-open those links and let Google crawl the canonical tags. Monitor, and look for a gradual decline in the numbers. If that decline doesn't happen or the numbers go up (over days or weeks - don't panic over one day), then adjust course.
Wow...thank you for the quick reply! Can't even tell you how much I appreciate it.
In your opinion, it would be best to turn off the parameter handling since the rel=canonicals are in place? Then I would know if the rel=canonicals works vs. WMT's parameter handling. That makes sense to me.
We have seen a ranking decline. It was in large part (i believe) due to Panda after the "real time" update went live in March 2013.
After going through the site there were some major indexing issues, mainly the using robots.txt to block duplicates and thin pages. I've later come to realize this was a big mistake and we should have implemented rel=canonicals (sorting/filtering) and noindex, follow tags (thin category/product pages, but useful for customers). Now the junk is in the index and we're trying to purge.
I appreciate the help, I really do. It's been a fiasco sorting through this. I figured with a site that is 2,000 real pages and 10,000 duplicates, this can't be good in a Panda world!
It's tough - I think that it's easier to interpret what happens if you just use rel=canonical and don't add WMT parameter handling to it. On the other hand, using both together should be ok and may be more effective. So, it's the risk of not being able to measure what happened vs. the risk of delaying a fix.
The biggest issue with Robots.txt in these cases is that it's just really bad at removing content that's already been indexed. Blocking a page/folder/etc. in Robots.txt before it gets indexed can work great. Removing a ton of URLs with Robots.txt almost never works very well, in my experience.
I didn't want WMT parameter handling to interfere, but if it speeds up the process that's probably best. I'll switch it to sorting, NO URL's and make sure all of the links are live to sorting/filtering functions on our site.
I'll give it a few weeks and see how the count is doing. Hopefully it'd going down!I'd imagine this probably takes 60-90 days to get most of the junk from the index. Trying to be patient!
Thanks again! Appreciate the input, this is one thing noone really talks about. What you do after you've implemented the rel=canonicals! :)
Hey Dr. Pete,
I just wanted to post a follow up. It's been about 25 days since my last post. Your recommendation has been working. We have been dropping duplicates from the index and it seems to be picking up speed as time goes on. We were at 17,000 duplicates when we set up canonicals and WMT's url handling. To date we are down to 4,000.
In your opinion...do you think sites with large duplication issues from sorting/filtering can trigger Panda penalties?
Thanks again for your help!
Greg
I absolutely think that sites with massive duplicates, even "honest" ones like that, can occasionally get hit by Panda. Usually, it's a combination of those technical duplicates and other problems, but it depends a lot on the strength of the site, size of the index, scope of the problem relative to the overall index, etc.
This article answers many questions, but I still have a question about local. We are a locksmith and have many locations. We have done a 'mini-site' for each location, reflecting local locations. However, the products are the same for all locations. We were going to do a rel=canonical for each of the products, but don't know how to get local indexing? Help?
Does google index the canonical URLs content of paginated posts or it only index the main URL?
For example the content at www.example.com/article-seo/1... /2 gets indexed or not when the paginated urls are set to cannonical ?
how does google see it ? if it does not index the content of paginated posts then it will effect the ranking of article as it will only index the content on main url which is www.example.com/article-seo/
Hi there! Just FYI, you're much more likely to get an answer to this question in our Q&A forum. Have a great day! :)
Google isn't crystal clear on this, and people who have tested it seem to get mixed results. The theory is that the paginated pages are eligible for ranking if they're relevant *but* that Google won't treat them as duplicates. That sounds good, in theory, but large-scale use of rel=previous/next doesn't seem to show great results (nothing terrible, but nothing spectacular).
Hello
I have changed my site from http to https and applied 301 redirect. However, 9 months on and I am still getting duplicate content warnings for the http and https variants. Can I use canonical in this instance, if so how? Thanks.
[Link removed by editor.]
Hi there! This would be a good question for our Q&A forum. You're much more likely to get an answer there. :)
Hi,
I used to have a website for web design and SEO service, Now I decided to split these activity into 2 different website. One web site for Web design and one for SEO. What if I use the same content from Web Design on SEO website? or what if I make a same exact copy of a website to the second one and one website focus for web design and another for SEO? do you recommend this?
Thanks in advance for your answer
So lets say I have a site with multiple domains for multiple languages, would it be correct to put this in the header for each domain start page?
From the swedish domain itself is listed as canonical, and the others as alternates. The same is done for all the other languages, itself as canonical and the others as alternates.
<link rel="alternate" hreflang="fi-fi" href="https://www.company.fi" />
<link rel="canonical" hreflang="sv-SE" href="https://www.company.se" />
<link rel="alternate" hreflang="de-de" href="https://www.company.de" />
<link rel="alternate" hreflang="en-gb" href="https://www.company.co.uk" />
I'm using tumblr and wordpress for my business and I'd like to avoid duplicate content. Unfortunately, there's no way for me to set a canonical link per tumblr post so I was thinking of setting the canonical link (sitewide on the tumblr) to my wordpress homepage - good idea or bad?
Hello Sir,
I need your help with one of my client’s website. In that website from the last 1 week it is showing in the WP-admin that “Your homepage cannot be indexed by search engines” this warning is given by Yoast SEO plugin and when I checked the issue on Onpage ORG then it is saying that the home page link is Canonicalized.
What they meant is that the PR of the link https://example.com/ is passing to https://example.com (I am not mentioning my site URL here)
I have also checked the canonical tag in the website source and have found it as -
<link rel="canonical" href="https://www.example.com" /> but as suggested by Onpage ORG the code should be <link rel="canonical" href=https://www.example.com/” /> (the href is having a missing “/” now)
So do you have information about how I can edit the canonical tag of Home Page? Also what is the reason behind the issue as I haven’t done any technical changes to the website?
I hope you can give a suitable reply to me. I am looking forward to it.
Thanks!
Hi there! This is the sort of question that is much more likely to get a response in our Q&A forum. I recommend asking there. :)
I work out of a different office than our web developers, and they told me they are not sure it matters where red="canonical' is placed -- which example is correct? Or does it not matter?
Example A) <link rel=”canonical” href=”https://www.example.com/url-a.html” />
Example B) <link href=”https://www.example.com/url-a.html” rel=”canonical” />
(A) is generally considered the correct usage, but, practically speaking, I think either attribute order will function.
I am new to canonical tags, but have noticed that some people set up the rel="canonical"> at the back of the url versus before href. Google's examples are not done this way, does it matter?
<link href="https://www.example.com/xyz" rel="canonical">
Hi All! II have question regarding Is Rel=Canonical links.
Does Re=canonical duplicate appears in Google Analytics or not all visits are automatically redirect to Canonical page in Analytics results as well ?
Thanks.
Mac
What about the variations on "canonical", e.g., "canonical-human" and "canonical-orgnization"? Wouldn't "canonical-human" (proposed) be basically the same as "me" (accepted)?
Hi. In the past few years I was publishing content in some forums and now I will use this content on my blog (with a lot of new content). I used canonical tag in the pages with duplicated content but now I have a problem. When I share an article with a canonical, it shares the page from the canonical tag. Is there any way to solve this problem? I mean, I want to maintain the canonical but share my own page.
Thanks for your attention.
Best regards
What would be your advice for restructuring a site, for example, mysite.com/product123.html to mysite.com/products/123 - I don't want to lose seo value of the existing URL, so 301 or canonical link to new page and measure performance?
the menus/top level links would begin to use the new links i.e. /products/123
thanks in advance
sam
Hey, nice article. But just want to know & confirm a few things.
1- How frequently can i use Canonical URL's? (i run a blog on which i post articles related to Health & Fitness, Diet, Meditation, Fashion, Beauty, Dating & Relationship. I am the Sole Author. It is not possible for me to write more than 1 article a day.)
2- Is it OK if post 1 Original article/day + 1 duplicate article with CANONICAL URL/day?
3- If i post as per (2. above) then i will be having 50% OG content and 50% Duplicate content with CANONICAL's. So will it affect my OG posts while ranking in Search Engines?
4- Will my blog be banned or something by Google?
5- What is the acceptable OG to Canonical Ratio?
plz answer them all...thanks
Sorry, it's tough to say without knowing more about the site and articles. What would the duplicates be? Would they be duplicated on your own site or other sites? I'm not aware of a limit to canonical tags (it won't be "duplicate" if you use rel=canonical), but I can imagine that there might be a point where it could look like a quality issue.
This is likely and ignorant question but I haven't found an answer yet and hoping you can help. I'm running into issues with the Google Crawler over "redirects" in my sitemap. I believe the issue is caused by "mydomain.com" and "www.mydomain.com/index.htm" being considered separate entities. Would inserting rel canonical into the "index.html" page and pointing it at the "mydomain.com" url resolve this issue? Thanks for any help, obviously I'm new to SEO
Unfortunately, it depends on the situation. A canonical tag could help prevent some problems, but it probably won't solve the error in Google Webmaster Tools. You may be including a version of the URL that redirects to another version. Generally, you'd only want the canonical version (the target of the redirect, most likely) in your XML sitemap. I could be a warning you can safely ignore, or it could be a sign of something more serious. Honestly, it's hard to tell without more details.
I have a client that is a financial advisor, an independent branch of a large financial services company, and he has access to and posts on his website page content generated by the large firm for branches like his. The content is a duplicate of what is on the large firms site, as well as other branches are using that same content to fill the pages of their websites too. If I use Rel=conononicle will it point his site in search back to the originator of the content. What is the best way to resolve this bedsides rewriting pages and pages of text?
My web site is www.my-bicycling-adventure.com
I have a link from it to my blogspot which is my-bicycling-adventure.blogspot.com
And the link from my web site is blog.my-bicycling-adventure.com
On my blog I have cycle tours that I have done previously. I want to take them and create specific pages on my own web site, but I don't want to delete the blog entry. Neither do I want to be penalised for duplicate content.
I want the web page to be the main copy, but I cannot see a way of getting rid of the rel="canonical" on the blog entry.
How do I do this?
Thanks
Stephen
Is there any equivalent to rel=canonical, not for the whole page, but just for parts of the page which might appear across a given website?
For example, using a "php include" to deliver the same product description in various locations across the site?
Advice much appreciated folks :o)
Short answer is, unfortunately, no. Google doesn't have a way to control indexation or canonicalization of just part of a page. There are tricks (like JavaScript-based content), but those are tough to implement and the results can be unpredictable.
Great post! I've been very confused about this for months because my site has 3 pages that are 93% the same (this happens for 3 different destinations) but I can't merge them because people use different words to refer to the same concept although they are not synonymous.
I have posts for all these different words they use and from here they are taken to the page that contains that specific word they like to use. To help this issue we have differentiated those words in a way that the products offered are a bit different (7%).
I'm a little stuck here and don't know if I should use rel=canonical and choose one of the 3 pages or if I should create content for all the pages in order to reduce the similar content (but I don't think it would make sense to talk so much about a small difference in the product). My last solution would be to do a 301 redirect from the 3 pages to a new page and mention all the keywords they like to use here. Thank you.
Hi Dr. Pete! I have a somewhat unusual circumstance. I am working on a redesign of a site that will involve taking existing content and creating a new version of it in a separate section of the website aimed at prospective customers. The existing content will remain as-is for current customers. The existing content has been in existence for quite awhile and therefore has value to it.
I'd like to be able to use rel=canonical on the current customer version of content and direct all the link juice and Google search results to the newer, prospective customer version of that content. The only challenge is that I'd still like existing customers to be able to find their version of the content via internal site search. We use Google Search Appliance on our site.
Is there a way to make our internal site search ignore rel=canonical?
Unfortunately, I'm not very well versed on Google Search Appliance. My general experience is that it's hard to modify its default behavior. If it honors canonicals, you may have a rough time reversing that. So, basically one version is for customers (behind a paywall?) and one version is for prospects? You could NOINDEX the customer version - it'll cut off some internal link equity, but then only the public version would be eligible for ranking.
Let me throw another curve ball
We run a bunch of employment job boards our main "Generalist" site (webjobz.com) and a bunch of 14 "niche" boards (ie expatengineer.net)
Problem is that all the content of the "niche" sites is also contained on the "generalist" site
So from reading all the posts here, if I have a job on a niche site ie
https://www.expatengineer.net/jobs/locations/Kuala+Lumpur/Kuala+Lumpur/ALL/job_5325424?
the same job will also be on the "generalist" site as follows
https://www.webjobz.com/jobs/locations/Kuala+Lumpur/Kuala+Lumpur/ALL/job_5325424?
Which should I use on the niche sites
<link rel=”alternate” href=”https://www.webjobz.com/jobs/locations/Kuala+Lumpur/Kuala+Lumpur/ALL/job_5325424?” hreflang=”en-au” />
or
<rel="canonical" href="https://www.webjobz.com/jobs/locations/Kuala+Lumpur/Kuala+Lumpur/ALL/job_5325424?" />
They are ALL English language site and mostly international sites ...just serving different industry sectors.
Hi,
Our Website is a complete dynamic website.
All the results we gather are from a backend functions.
The other pages are the extension of first page i..e
Foe example
Home page
https://www.abc.com
The other pages are
https://www.abc.com?shfsfkla;hflslf
So it i put rel=canonical Tag in the first page header, the search engine will not index it or devalue it?
What is the solution to this problem
Regards,
We have a website where users post stories or articles. In this case there is a 100% chance users must be posting the same content to other websites also. What is the best way to protect the website from duplicate content pages? How to use rel=canonical. We use rel=nofollow for any external links. If any in the website so that juice is not passed to external urls.
Hello Pete and thank you for the informative post, but I still have questions regarding rel=canonical and have yet to find my answer.
I have a small eCommerce site with about 35 products. Each product belongs to one of 5 categories (brands). I have implemented links from the main menu to the category pages. These category pages also serve as shop pages from which a visitor may "Add to Cart" or click on a product to view more information. On some products, there are few if any differences, other than the weight.
I have read about using rel=canonical to associate one product to another, but what about associating several products to the category/brand, even those with varying descriptions and metas?
To use an example in the Rolite brand/category, we have "Pre Polish," "Metal Polish," "Protectant" and "Premium Polish & Sealant."
Premium Polish & Sealant is not utilizing rel=canonical because it is a one of a kind product/description. Other products have multiple sizes, but duplicate descriptions and meta. For these, does it make more sense to use rel=canonical to the category page or to one of the duplicates?
Does it even matter with such a small site?
Thanks in advance.
Hi dr Pete,
I was watching John Muller's video on Webmaster Central hangout and in this video https://www.youtube.com/watch?v=XzzY2T2Ph8w&list=U... he has some really interesting points:
1. at the minute 5:56 he is saying that when using rel="canonical" is important that the pages to be equivalent. Furthermore he says that a page about blue shoe is not equivalent with a page about the same red shoe and that he doesn't recommend to place a rel="canonical" across these types of pages.
Until this video, I thought they were equivalent and I believe that was your thought too (unless I misinterpreted the (4) point in your article (near duplicates)
2. he is saying: "don't worry about duplicate content within your site" as a possible penalty. It can only harm you:
- from a crawling point of view
- or if the content is spammy, low quality, not relevant for the user
- or if you use it excessively
So, in the light of this new information, I am hoping that you will help me with your opinion (that I believe is very valuable).
We have a client that is selling bus tickets for different cities. There are pages created for every route (different points of departure - same point of arrival), such as, let's say:
- example.com/bus-tickets-amsterdam-paris
- example.com/bus-tickets-frankfurt-paris
- example.com/ bus-tickets-warsaw-paris
etc
For every arrival point we have around 40-50 departure points, therefore different 40-50 pages, with different title-tags, meta-descriptions,etc, but with thin content as this point. The only thing that is different, is the tickets prices.
In order to make those pages more useful for the users, we would want to add city guides for each city, but of course, it would be the same guide (original text content) for each set of 40-50 pages. So we will have duplicate content, but it won't be thin anymore, it won't be spammy nor low quality and is also relevant and useful for the user that is travelling to a new city.
From your experience, do you think this could be interpreted otherwise and harm the website overall?
The other choice is to keep the most searched route as original and mark up the others with rel="canonical" , but they won't have the right prices displayed and they would be forced to perform a new search.(eg: if they searched for bus tickets frankfurt paris and the frankfurt-paris page has canonical to amsterdam-paris, Google will be serving the amsterdam-paris page, but the user will see the amsterdam-paris prices and not the ones that interests him: frankfurt-paris).
Thank you
Re: (2), I think Google has routinely taken a "let us handle it" approach to duplicate content, and yet I've seen many cases, even before Panda, where such content did considerable harm. Post-Panda, it seems incredibly disingenuous for Google to make that claim. It strikes me as nothing but a CYA tactic, to be brutally honest. They're punishing people for their inability to handle duplicates, IMO. The simple fact is that duplicates cost them a fortune and, perhaps understandably, they'd rather we clean out our own thin content than have to do it for us.
In my experience, these kinds of geo-targeted, thin-content pages are exactly the type that have been clobbered in some of the Panda updates. Of course, that's a broad generalization. If those pages have unique content, represent legitimate routes (and your unique service), are a reasonable proportion of your indexed pages, and if you have the authority/links to back them up, they could be perfectly fine. The devil is in the details.
Hi Dr. Pete,
thank you very much for your feedback. It is very valuable and it helped me taking the decision.We will take the safe approach for now:) .
I have a static website in which client created city specific pages of each state , but all of city page's contents are same only city name changes, in this website 9 States Pages and in each state 40 city wise pages, all of Page's contents are same so my question is that should i use rel canonical in each 40 pages ?? if yes then i have to use rel canonical in total 360 pages , i am bit confused that this much of tags in each of those pages can raise a flag on Google eyes ???
Please let me know best option and why ??
1) rel canonical tag in all 360 pages
2) noindex , nofollow in all 360 pages
3) disallow all 360 pages from robots.txt file
Thanks in Advance
R
Kunal
This would be a good question for our Q&A section at https://moz.com/community/q. Hope to see you over there!
Hi Great post, but to clear up a question. I have a number of reviews for my website and when each review moves onto another page it creates a new url www.example.com/revpage1, www.example.com/revpage2 and the connonical tag is on rev page 1 do i then need to place a non canonical tag on page 2 page 3 page 4 etc?
Pete
We moved a well ranked .co.nz site to .com.au to achieve traction in Australia which worked (at the expense of our NZ rankings). I had thought of doing subdomains but at that time, profile in Australia was the drive and I was worried a .com wouldn't deliver the rankings in Australia. This was so successful that Australian entity has now been sold. The new owner will also handle Papua New Guinea. The old site went with the new owner although the parties agree to still work together.
So the plan is to create a new .com site with subpages for middle East and NZ that allow the client to expand in other regions. We don't really want to have to recreate all content,
Can you help point me in the right direction. Should I
1) Move everything to the .com and have subdomains with alternate tag on the subdomain identical pages. Will this then cost me my ranking in Australia? This might not make the new owners happy, as they think they have the .com.au site and we need good outcomes for both parties. Assume the .com site would have these:
<link rel=”alternate” href=”https://example.com/en-ie” hreflang=”en-au” />
<link rel=”alternate” href=”https://example.com/en-nz” hreflang=”en-nz” />
<link rel=”alternate” href=”https://example.com/en-au” hreflang=”en-pg” />
<link rel=”alternate” href=”https://example.com/en-pg” hreflang=”en-pg” />
<link rel=”alternate” href=”https://example.com/” hreflang=”x-default” />
2) Create a new .com site, leaving the .com.au site, and just linking to that site off a world map on contact us page. Presume I'd need to put relative canonical on the Aussie site but wouldn't that also affect their rankings in Australia?
Whichever way, the new countries we need to rank in are NZ & Dubai, without sacrificing Australia!
Rel=alternate/hreflang will help you avoid duplicates and help the appropriate pages match in their relevant regions, but it doesn't really boost your ranking ability. How to most strongly target a country is a much more complex question. Country-specific TLDs do still have power, but splitting a site into multiple domains has trade-offs.
I'm a bit confused because it sounds like you no longer have the Australian site. If you duplicate the site you sold, then you'll be competing with that site, too. If you still have an existing .com.au site and move it to .com, you may benefit the other countries slightly, but you will almost definitely lose traction in Australia. There's no risk-free proposition here, I'm afraid.
PS I meant .ae where I duplicated the .pg suffixes!
I have a website with state-level geolocation. Content is roughly the same, but prices and available deals may vary from state to state, so we have a structure like:
And so on and so forth. All pages in the aforementioned example have a rel=canonical to www.mysite.com/deals; search engines seem to obey that and I don't have state-localized URLs in their index, so our initial goal has been met.
The only question we have is whether we should 302 or 301 a user that access directly the canonical URL, in that case www.mysite.com/deals, to the state-specific URL.
Right now we use 302 redirects because it seems as the right thing to do. Should we keep it that way or change to 301 redirects? We're worried about duplicate content, cached browser redirections etc.
Any suggestions are appreciated! :)
Regarding blog tags, can someone verify if we can add the canonical tag for individual tag pages with the canonical url pointing to a page containing all the tags?
That can really get tricky, because now you're into what are effectively internal search pages that may have pagination, filters, etc. I think it's best to control tags, especially user-generated tags. Otherwise, you can end up with 100s of pages of thin content. On the other hand, some tags can be useful category-type pages for ranking purposes and visitors.
With low, controlled numbers (a handful of tags), I'd generally say leave them alone. If you allow user-generated tags, I think META NOINDEX,FOLLOW may be a better choice. Rel=canonical to a master search page may not be useful and could be ignored, especially at large scale. It's pretty situational, though.
As I understand, Google recommends the use of absolute URLs in the Canonical tag. I've seen a case where relative canonical URLs were essentially telling the search engines that 4 to 5 different URLs were the canonical, because of the extensive use of sub-domains in this case.
Any additional thoughts on absolute vs. relative URLs in canonicals?
Google stated in that recent post that relative URLs are ok, but they also warned that relative URLs are much more prone to problems. I've definitely seen issues where relative URLs were used incorrectly or referenced badly in rel=canonical tags. In theory, they should work, but in practice, I find absolute URLs to be a lot safer.
I second that. I had a problem with my +1 button implementation that only happened because I was using relative URLs in the Canonical tag. Switching to absolute URLs fixed it right away.
Great post Pete, we often forget to take over “rel=canonical” issue, maybe it's a little hard to understand but it very useful because it tells search robots which page is the singular "authoritative" version which should count in web results.
It would be great if you cound confirm my thoughts on a smaller issue with canonical.A webshop has a lot of products in a category "ironing". Let's say they have products for 40 pages. When using canonical for sorting, meaning price ascending and price descending, page 25 with price descending won't have any simmilar products on it as page 25 with ascending.Still, using canonical, showing to the pages without any sorting, is the way to go, because all the pages are allready indexed as without the sorting.Am I right?
Thanks
Yeah, this gets tricky, and you may want to put rel=prev/next into play (although that only solves half your problem). While I completely agree about the mis-match, rel=canonical is still probably helpful here. The other option would be to use a form/menu for sorts that isn't crawlable or to block the sort parameter in Google Webmaster Tools (just tell them to ignore it). Ideally, it might be better to just tell Google to ignore the sort altogether.
Do you know if rel=canonical pass full link-juice to canonical page or some small amount of juice is lost, like with 301 redirect?
Thanks for sharing your knowledge with us Pete!
We don't have strong evidence, but I believe so, yes. Otherwise, people could abuse them too easily. It's interesting to note that Matt clarified the 301-redirect loss of link-juice on a video recently by explaining that it's similar to the loss of any link. As part of the iterative way that PageRank is calculated, there has to be a loss factor, or it would continue to accrue forever. So, part of the "loss" is just to keep a 301 from being stronger than a regular link. I'd have to think this is true for canonicals as well.
Hi Dr. Pete,
Finally, you brought this complete guide on rel=canonical, i missed your share when it published anyway. I got this and seeing there is everything included now, what i need to know about REL=canonical.
thumbs up!
Thanks for answer!
Maybe this was asked already but I was wondering if anyone knows of a tool online to test if your canonical tags are working properly. I have a site where I can't add the code to the <head> section and am wondering if it still works.
If you mean "working properly" in the sense of being processed and honored by Google, I'm afraid there isn't. When you say you can't add it to the <head>, what are you doing instead? From my tests and Google's comments, I'm 99% sure that a rel=canonical in the <body> won't work.
I didn't know all of the capabilities of rel=canonical. Really eye opening post.
Its really good post! now someone tell me how to resolve the pagination issue of my site i.e. (www.333acre.com) please check the search for any property then pagination comes. please help for resolving this issue.
thanks
Dr Pete
As a newbie in the SEO world I have been absorbing as much as I can from these Blogs, I would like to ask advice around the subject of rel=canonical though.
I am currently having problems on webmaster tools with duplicate title tags, we need to show almost identical pages for vehicle lease offers as per the following URLSs as an example,
https://www.carlease.uk.com/car-leasing/nissan/qashqai/hatchback-special-editions/1.6-[117]-360-5dr
and
https://www.carlease.uk.com/car-leasing/nissan/qashqai/hatchback-special-editions/1.6-[117]-360-5dr?show_vat=1
could we use a canonical tag to prevent these HTML improvement messages from Webmaster Tools?
Yes, I think rel=canonical is a good bet here. You could also tell Google in Webmaster Tools that the "show_val=1" parameter should be ignored for search purposes. Both should work, although I think rel=canonical is generally better (since Bing also honors it). If you do that, you might want to link to the VAT version from the non-VAT version. The other option would just be to show both prices on one page, but I'm not sure if there are legal issues with that (I'm in the US, so I'm only vaguely familiar with the VAT).
Thanks Dr Pete appreciate the advice
What are your thoughts on implementing rel=canonical on a meta robots noindex page to a separate indexed page elsewhere?
I think it's generally a bad idea to mix signals. The META NOINDEX will probably overrule the rel=canonical or keep it from being processed. Choose the more appropriate signal for the situation.
The "Can I have my cake and eat it" section made me giggle. Added bonus to an already excellent post :)
Hello Pete!
Its an incredible guide about canonical tag and how we can implement it on various situation.
My question is, If one of my website is based on multiple language and every multiple language website contains their own language domain extension and content such as
www.preditmydomainname.com,
www.preditmydomainname.com.au,
www.preditmydomainname.co.uk,
www.preditmydomainname.co.jp,
www.preditmydomainname.nl
If my main domain name is www.preditmydomainname.com, and if i use canonical tag on ".nl" and "co.jp" extension domain name, then is it right in case of duplicate content?
Require Advise
Thanks
Pierre Far recently suggested to implement the hreflang mark up also in the case of sites in different languages and in "cross domain" situation, and that seems to be your case.But if you use the rel="alternate" hreflang, then I suggest you to check out my comment to this post.
Agreed. See my comment on a similar question above and Gianluca's comment.
Hi PeteGreat post as per always with your stuff. This all probably fits under sub-regular-seo / aka website admin as a category ... this said it does server as a reminder to those not using canonicalisation that is can be really quite userful!
Nice post. It's always good to remember the situations where rel=canonical must be used or not.
Hello Pete,
Thanks for Sharing great post about rel=canonical. Here I have question regarding same.
Suppose I have website www.example.com and another is www.example.eu and the content and all other information are the same in both website.
So Can I use <rel="canonical" href="https://www.example.com" /> in www.example.eu
It's a complicated subject, that's a situation where rel=alternate/hreflang is a more appropriate tool. See Google's resource:
https://support.google.com/webmasters/bin/answer.py?hl=en&answer=189077
It's not just for languages - the tag can be used to specific the same language across different regions (like English in US/UK/Australia). It helps Google rank the appropriate version in the appropriate region.
Hi Dr.Pete
This was amazing post and has clarified many of my questions and confusions.
But I do have same question which gfiorelli1 asked
Example:
If a company has a ccTLD example.com, example.com.au and example.co.uk
Do we need to use: rel=”alternate” hreflang and rel=canonical?
i.e:
<link rel=”alternate” href=”https://example.com.au” hreflang=”en-au” />
<rel="canonical" href="http:://www.domain.com.au" />
<link rel=”alternate” href=”https://example.co.uk” hreflang=”en-uk” />
<rel="canonical" href="http:://www.domain.co.uk" />
<link rel=”alternate” href=”https://example.com/en” hreflang=”en” />
<rel="canonical" href="http:://www.domain.com" />
See Gianluca's comment - the combination of rel=alternate/hreflang and rel=canonical is tricky, and I'm not an expert on it. Google also recently announced the default option for rel=alternate, but I haven't seen much data on it.
A nice canonical refresher, thanks !
Haha, the second time I've heard the phrase 'secret sauce' recently - had a client said they didn't want to know our 'secret sauce' - like we have some!
Anyway getting distracted before I've properly replied to this - great reference and one I'll put in my reference documents. This issue only tends to crop up with larger and/or e-commerce clients so will help to explain things better.
Two questions (I think):
1. There should be no way a no implementation of rel=canonical on a small site should cause problems is there? (excepting people copying content).
2. Can you use canonical to prevent an outside site cause you to be penalised for duplicate content i.e. your site along with authorship+canonical? I know this should be an extreme case, but just thinking from a content-protection perspective.
Thanks!
(1) Are you asking if a small site will be ok without any canonicalization in place? Yes, typically, it's fine. Many small sites worry a bit too much about technical SEO, and most problems are on larger sites. I almost always put one on the home-page, just in case, but that's just a minor precaution.
(2) Unfortunately, it probably won't help. If someone scrapes you, and for some odd reason, they scrape the entire , then it could help you, but if that gets put in the body of their content, it won't be processed. I've heard people suggest this, but I've never seen it actually work. It's harmless, but you can't really use rel=canonical to force Google to see your site as the source of a piece of content.
Yeah that's what I thought (in both cases)... and I agree on the home page thing as I've seen sites we've inherited with links to both /index.html and /index.htm and just root and although G is usually smart, links coming in aren't and could cause future problems.
Anyway enough of thinking out loud for one day! Thanks again :-)
That's the most common issue - people link to "/index.htm", etc. in their navigation and instantly spin out a duplicate. ASP/.Net sites have a common issue with default.asp(x), too, that can be maddening.
Good article. I happen to be researching this very issue for a website which has products whose only difference is color (advertising flags), but the description is very similar. I am interested to see the answers to some of the questions. Our website is not run by wordpress but I do manage a few that are so that question is of particular interest.
Dr Pete?
What do you do when only a portion of the content is duplicated?
Scenario: A site sells books by various authors, and each page has the author's bio prominently placed on the page. The same bio is posted for each of the author's books before the synopsis of the book.
Since you can't use a canonical on a section of a page, should we put the bio in each book's detail page in an <iframe> from a dedicated bio page on the site?
A broader question: Is this even considered duplicate content?
There's no great tool/answer for that. In cases where there's still rich content on the page, I'd leave it alone - it's not really a duplicate. If you start trying to cloak the content, put it in iFrames, etc., you tend to do more harm than good. You could try a few things:
(1) Only put the start of the bio, with a link for more information to the profile, reducing the amount of duplicate content.
(2) Put the author name/pic at the top, but then the bio at the bottom of the page, pushing up the unique content.
Hi Dr. Pete,
Thanks for the "simple advanced" breakdown of the many applications of Rel=Canonical. You've saved me "10k" hours of mastering this hard and technical subject.
Thanks for the post. In depth as usual. Any tips on managing canonical on very dynamic sites where pages could vary considerably in 'uniqueness' ?
I've got a very important question that till that moment I'm not sure regarding the answer-Let's say I have a pagination product pages. Every pagination page has the same exact article.Would the rel/prev attributes will solve the problem in here, or should I remove the article from all the other pages besides the first one?
Canonical url issue
My site https://ladydecosmetic.com on seomoz crawl showing duplicate page title, duplicate page content errors. I have downloaded the error reports csv and checked. From the report,
The below url contains duplicate page content.
https://www.ladydecosmetic.com/unik-colours-lipstick-caribbean-peach-o-27-item-162&category_id=40&brands=66&click=brnd
And other duplicate urls as per report are,
https://www.ladydecosmetic.com/unik-colours-lipstick-plum-red-o-14-item-157&category_id=40&click=colorsu&brands=66
https://www.ladydecosmetic.com/unik-colours-lipstick-plum-red-o-14-item-157&category_id=40
https://www.ladydecosmetic.com/unik-colours-lipstick-plum-red-o-14-item-157&category_id=40&brands=66&click=brnd
But on every these url(all 4) I have set canonical url. That is the original url and an existing one(not 404).
https://www.ladydecosmetic.com/unik-colours-lipstick-caribbean-peach-o-27-item-162&category_id=0
Then how this issues are showing like duplicate page content?
I took a quick glance at the campaign, and off-hand, your canonical tags look ok. I think we may be wrong on this one. I'm wondering if the secure pages are giving us a little grief. You can email Support to find out specifically what we're seeing, but I don't see any reason to panic here.
Point # 5 & # 6 can become intertwined issues.
I think one of the most important comments from the post is "I do worry that excessive use of rel=canonical could cause search engines to devalue and even ignore those tags" (found in point # 6).
Oh what a shame.. imagine spending the time to set up every page on your site with canonical and later find the effort becomes devalued or ignored altogether.
In my opinion, canonical was created for one reason. I am hesitant to use the word tricks, but it may trick some webmasters into uneccessarily "reducing" the number of pages being measured by the engine.
For example, Joe is losing traffic for over a year, He is desparate to find a solution. Although his pages have entirely unique content, he would be tempted to canonical the following three pages all to on the news.html page.
news.html
online-news.html
industry-news.html
Again, if all three pages truly had unique information, and Joe combines them, the canonical allows Google to cut their investigation by a third. Multiply this out by the number of pages on the net and Google has a HUGE savings on resources. Although the original pages had been valid and of worth, Joe told Google that news.html was the only important one of the group.
In the short term, canonical may benefit both google and the webmaster. Eventually (after everyone uses canonical), google will have to abandon the concept and determine a new way to cut down on the number of pages to review. Unforatunately, google will not be their to help us gain back the time lost creating all those canonical links.
Hai Everyone I have Question Anybody plz replay
I have a site name
https://www.example.com. and
https://www.example.com/index.html
both the pages are same.. i wrote canonical as
https://www.example.com/" />
Is it right or Wrong ???
Has anyone used rel canonical tags for large e-commerce sites without google thinking you are page rank manipulating
Hello Pete,your stuff is really good.may i translate it to chinese?looking forward to your answer.tksFigu
Hi! Please send an email to [email protected] and we'll send you the information about translating blog posts. Thanks!
Thanks for the offer to answer questions - I hope I'm not too late.
If an article exists on a website, and then this article becomes a duplicate at a new URL on the same website - rather than new to old, is it okay to canonical the old version to the new version, if this is a regular occurrence? The new version also has a slightly differently designed page. This duplication cannot be prevented in the short-term, and it'd be a bad UX to 301.
Hello Dr. Pete,
Today, I was reading 5 common mistakes with rel=canonical on Google webmaster tools' official blog. They have given ideas on mistake about rel=canonical to the first page of a paginated series and You have given very clear ideas on Near Duplicates pages. But, I have similar question regarding Narrow by Search or Shop By section pages which are creating on Ecommerce website.
I would like to give very clear ideas on it.
1. Can I set Canonical URL https://www.gunholstersunlimited.com/airguns.html on following page. There are thousands of pages on website which was compiled due to Narrow by Results section.
https://www.gunholstersunlimited.com/airguns.html#!/p=clear&manufacturer=228&order=name
2. Can I set Canonical URL https://www.knobdeco.com/cabinet-hardware/cabinet-knobs.html on following page. There are thousands of pages on website which was compiled due to Narrow Shop By section.
https://www.knobdeco.com/cabinet-hardware/cabinet-knobs.html?line=edwardian
I am looking forward for your valuable reply on these two questions.
(1) That looks like an AJAX-based implementation, so it theoretically shouldn't need canonicalization.
(2) I'd have to understand the site better to give an answer I'd be confident about, but given that you're not changing the page title, copy, etc., those filtered searches look pretty thing. My gut reaction is that I'd probably canonicalize or NOINDEX those variants. They aren't doing much but diluting your Google index.
Honestly, though, search sorts, filters and pagination get complicated very quickly, so you want to make sure you've got a strong structure and approach that covers all of the variations on the site.
Hello Dr. Pete,
I really like that you are checking each and every comment on your blog post and reply as fast as you can! This is big reason for me to love SEOmoz blog...
I really appreciated your quick response on my comment and going to make it happen on live website. I will share my experience on this blog post so other users can get more ideas for Canonical, Robots.txt & Meta Robots NOINDEX, Follow.
Hi Pete,
First off, I'd like to thank you for writing this article. I've found it particularly difficult to find recent posts on this topic at an advanced level.
My question is on cross-domain canonicalization. Are you able to cross-domain canonical three different domains?
One of my clients has three different sites with the same products due to different audiences and stakeholders. Ideally, I'd like to only implement the canonical on the product level and create unique titles and descriptions for categories.
Thanks in advance!
In theory, you can cross-domain canonicalize multiple domains. If it gets excessive, Google may not choose to honor those tags, and if you've got duplicates across many domains, I think it's often time to rethink your strategy, but the tag can be used in a many-to-one fashion. Just make sure that the canonical domain is consistent. If you canonicalize domain B->A and then C->B, Google is going to get confused.
After spending a day to understand what is canonical tag and how it works, this article has made me clear. In middle of the day, I was thinking of using canonical link and 301 redirect at the same time.
Actually rel=canonical is used to show the uniqueness of webpage. it finish confusion about webpage to search engine.
I have an ecommerce site which has a variety of different types of duplicate content:
Examples:
Duplicates via /sort, /search, /category, /product
There are just a ton of avenues to the product pages and it has created a bit of a mess. Personally, I would like to use the canonical tag. However, the client is on a small retainer and I don't think I have the time to assign a proper canonical to all the pages in question.
I was wondering what the difference would be between a decision to use rel=canonical vs. meta robots no index, follow and robots.txt.
What are the plusses and minuses.
Thanks.
Christian
Great article, Dr. Pete. Thanks for your insights.
I was wondering how you handle a site where you create the link and the CMS creates an additional automated link? That means you have hundreds of duplicates. Is there a way to automate the rel=canonical tag or is there another best practice to handle this problem?
For example: site.com/about and site.com/about/tabid/12/default.aspx
It additionally creates site.com/about.aspx
Ideally, you fix the CMS, but I understand that's not always an option. Usually, there's a programmatic way to do it (in other words, a few lines of code could create proper rel=canonical tags across every instance of this problem). Unfortunately, I can't tell you what that is, because it depends completely on the CMS in question. If the CMS can generate the new link, though, it can probably generate the canonical tag.
Hi thnx for the article ik have a question google sees duplicate content at my siteat www.domainname.nland www.domainname.nl/index.htm
or www.domainname.nl/folderand www.domainname.nl/folder/index.htm
the index.htm exists on the server. can i use a canonical url like : www.domainname.nl?(btw the PR is the same)
Yes, rel=canonical is generally a good solution for that problem. Generally, though, this occurs because of internal links, so if you can change those links to the root path, that's a better long-term solution. In other words, don't link your "Home" button/icon to "/index.htm" - use "/" or an absolute path.
Dr. Pete, Is there a need for a rel=canonical on a mobile site that is responsive or dynamically served? Do we need to at a rel=canonical just for the Googlebot-Mobile if we already have one on the page for the normal google bot?
I was reading at Google Developers and it shows that you needed to have it for a subdomain mobile site but i did not see any mention for the other two. Do you know what the best practices are for this or could you point me in the right direction?
Thanks,
If the URLs don't change, you shouldn't need canonical tags. "Dynamically" served could mean a few things, so it depends, but responsive design should be fine.
Thanks for your help!
We are looking to set-up Google Shopping feeds for English speaking countries from one site that will use the site URLs to trigger the specific currencies and prices for those countries i.e. USD, AUD etc
The URLs will be the only thing that will change; the content will remain the same which raises the question of duplicate content?
Our question is how best to canonical these pages to avoid any duplicate content penalties. For example, the existing site is a UK site; would it be appropriate to canonical the different currency URLs to the original product URLs on the UK site? Example:
www.example.com/product1
www.example.com/au/product1
Should these canonical to:
www.example.co.uk/product1
Grateful for thoughts on this.
John
Anyone have any thoughts on this?
A very descriptive article and definitely learned a lot about the Canonical tag. However, I have a question, I know you mentioned that Google might not honor Cross-Domain Canonicals. So in what circumstances will they honor it. Are there any specific examples? In this scenario, if another website with higher authority copied my content and decided to use a canonical tag to my site. Will Google honor that? If not, wouldn't it be duplicate content and since the other site has more authority, wouldn't my site not appear in the SERP? Any how, thanks for the amazing article.
It is generally used for syndication, so if another site uses your content and sets a cross-domain canonical back to you, that should be a signal that your page should rank and theirs shouldn't. That said, it's not a guarantee. The bigger problem is that, if someone is copying you without asking, they're not going to ever cross-domain canonical back to you.
Pete, thanks for very useful post. But I still have 2 questions about re=canonical:
1. Is it ok if my website has every single page with rel=canonical to itself? Our coder implemented this thing to prevent duplicates on Joomla.
2. Are there any rank improvements if I post few of my posts (fully duplicated from the main website) to my few secondary websites with DoFollow backlink and use rel=canonical for them to the main original website? (Cross-Domain rel=canonical with full duplicated content and DF links to the main original article).
Thanks in advance.
(1) Generally, yes, as long as they're appropriate. My only worry is when people set it up incorrectly. It's pretty common for CMSs like Joomla, though, and generally a good idea.
(2) On the one hand, if Google honors the rel=canonical cross-domain, then the dofollow link won't pass any equity. On the other hand, the canonical will keep the pages from being seen as duplicates and will consolidate any inbound link equity to the copies (if someone links to them instead of the original). So, it can have advantages.
Thanks for your answers. It was very useful for me.
That was a good researched post on Rel="canonical". I guess most of the confusion on this single meta tag has been cleared. Nice post Dr. Pete.
Like always, thank you for a great post on a topic not well addressed previously. I do have a question, though.
If I have two pages on the same domain, one page contains all of the information of the second page, but the first is a longer write-up on the subject both pages deal with, is it appropriate to use the re=canonical tag to point the shorter, second page to the main article?
Example: www.example.com/why-use-the-word-widget.html is a broader, elongated explanation containing, word-for-word, all content on www.example.com/what-is-a-widget.html. Is it appropriate to canonical the /what-is-a-widget.html page to the /why-use-the-word-widget.html page?
Great one Dr Pete. I hope everyone is reading this. And a perfect article topic. People needed to be over this confusion. Cheers for doing this one. I hope to see you at Mozcon. Trying to plan my travel around to make it to Mozcon this year.
Should the canonical tag ever be used to help Google determine the preferred domain (www vs non-www)? If not what do you recommend for that other than notating the correct option in WMT? Do you use HTaccess rewrites?
It can, in theory, but typically 301-redirects are a better bet for this. They're also generally a lot easier to implement sitewide, as one rule can apply to the entire site. The rel=canonical tags would have to be placed on every single page/URL.
Thanks Dr. Pete, you make it all so easy to 'get'... as usual :-)
I have a static site. About a year ago, I switched my newsletter service over iContact. They use tracking URLS similar to this:
https://www.example.com/news/may-2013.html?utm_source=iContact&utm_medium=email&utm_campaign=Example&utm_content=May+2013
I started adding rel=canonical tags to every page that is mentioned in the newsletter in case it's indexed. Do I need to? Can tracking URLS cause duplicate content?
Also, I have some pages that use anchor tags (ie https://www.example.com/news/may-2013.html#may5). Should these pages also have rel=canonical added?
Even with "utm" variables, which Google should be able to ignore, I have seen issues. If you know a page is going to frequently have tracking variables appended, then I think rel=canonical is a good bet. Keep in mind, that Google would have to crawl those tracking variables. So, if this were from a PPC ad or something Google never saw, it's probably ok. If it's an email campaign, though, and that campaign has an HTML version people can view, then the parameters might get crawled.
True named anchor tags (not AJAX or some other tech that uses them) are no problem. Google won't treat those as separate URLs.
Search Result Pages are usually low quality and thin pages according to me. Its Better to Noindex Search Result pages - Again an opinion instead of using rel="canonical"
Hi, I came across a situation that maybe you can help me with.
We have a client who has an well established, with high authority website A (e-commerce)
Our client allows another website (let's call it website B) using the content from site A, through a white-label licence. Website B is also an well established, with high authority e-commerce website (even higher than website A), who has millions of products of various categories. But until the white-label from our client, they didn't have this type of products, so it wasn't competing with website A.
When the white-label was implemented on website B, to avoid duplicate content, declare the original and avoid competition in SERPs (as agreed), the categories and products imported from the white-label where marked with rel=canonical, cross-domain, pointing to website A. (aprox, 2.000 product pages and 3 categories). The 3 categories are linked from the main menu on website B and can be accessed from every product page (around 1.3 million pages on website B)
What is confusing is that in client's A Webmaster Tools account, in the Links to your site sections, we suddenly can see 4.000.000 links pointing from website B to website A. The links point mostly to the 3 categories. Can the rel=canonical be interpreted by Google as a link? It doesn't look natural and after the Penguin 2.0 website A suffered a drop in SERPs, from 2nd place to the 4th for the main keyword phrase.
We looked at the back-links profile, but nothing else looked suspicious.We didn't engage in any spam links tactics, so...do you think it's possible that the rel=canonical from website B could have done that? And if so, would it be a solution to leave the rel=canonical only on the product pages and remove it from the category pages?
thank u
I suppose it's possible, but I haven't seen reports of that yet. Are those pages also linked - it could be that they're counting the links even with the cross-domain canonical in place. I doubt that links from just one domain would trigger a Penguin penalty - more likely they would just be devalued. At massive scale, though, it's always a bit tricky. Our the cross-domain tags being honored? Specifically, do you see the non-canonical pages getting indexed and/or ranking?
Thank you very much for answering.
1. Yes, the pages are linked from every product page on website B. The non-canonical pages are being indexed, but they don't rank.
2. Further, we tested and added a 4th category on website B, but didn't add rel=canonical on the category page, but only on the product pages. The WMT account started showing a few links to the product pages from the 4th category, but no massive links to the category itself.
Now, the dilemma is: if we remove the canonical from those 3 categories on the website B, given the fact that website B has a higher authority, wouldn't website B rank higher than website A?
thank you
Unfortunately, this is a complex enough problem that you'd need someone to take a deep dive, but my gut reaction is to leave it alone. I think removing the canonicals is a lot more risky than some odd numbers in GWT. A drop from #2 to #4 doesn't really fit the profile of a full-on Penguin penalty. I'd dig deeper and see if there's more to this picture.
Yes, I agree that's a lot more risky and we will continue the investigation. The tricky part is that website A has been developing constantly over the past year, including internationally, with various releases from various teams, sometimes daily and it's kind of tricky to correlate all the actions, but I hope we will figure it out. Thanks again for your time and input on this.