How Much do Canonical URL Issues Impact Ranking?

[JoyHawkins]

I often come across businesses that have canonical issues, meaning that example.com and Example Domain might both return 200 codes instead of having one 301 to the other. It generally happens more with sites that have https.

I was wondering what impact you guys are seeing this have these days for SEO? All the articles on these topics are super old and I was curious to see if you find it's one of those things that isn't a huge issue anymore since Google might be "smart enough" to figure it out.

I know in the past it was said that having 2 different versions of your URL could hurt ranking juice flow or Page Rank flow which hurts ranking but honestly haven't seen much on this topic in a long time.

 

[Chris D]

Hey Joy,

One of most surprising effects of canonical I experienced was when I first stared to use it on ecommerce store of my client. Those days we had a lot of useless product pages which were gathering a lot of traffic from specific keywords but not convert because the products were out of stock. I used canonicals on these pages and point them to relevant category pages. Surprising effect was that Google switched the traffic from out of stock product pages to these category pages and in many cases even with better positions.

 

[Linda Buquet]

Hey Joy,

This is a little out of my depth, so pardon if too far off topic.

But since you mention SSH, I think this info from Helmut in the doc is very interesting.
https://www.en.advertisercommunity.com/t5/News-Updates/Meet-the-Social-Team/m-p/824801#M405

I mean I realize his findings are re not being able to get a site to stick in GMB, but it also sort of points to the fact that Google sometimes gets confused with SSH sites. (If I remember right, been a couple days since I read it.)

 

[DanLeibson]

Canonical URL issues can be huge in terms of both organic rankings and pack rankings. Usually one of the first things a site should resolve.

 

[HoosierBuff]

I don't see much of a difference. I have some very anecdotal information:

1. This past year I have fixed 3-4 sites that were resolving both at the www. and at the root domain. I did not notice any major change in traffic. Sometimes that tough, because we do so many changes. I have one that changed about 30 days ago, and we have made no other substantial changes. The traffic is roughly the same.

2. (this is vaguely related): I had a site that screwed up the canonical tag on the home page. The canonical tag was mistakenly edited to www.www.example.com that DID have an impact. The home page basically disappeared from search. I know that's not the same thing, but there are some similarities.

 

[JoyHawkins]

I don't see much of a difference. I have some very anecdotal information:

1. This past year I have fixed 3-4 sites that were resolving both at the www. and at the root domain. I did not notice any major change in traffic. Sometimes that tough, because we do so many changes. I have one that changed about 30 days ago, and we have made no other substantial changes. The traffic is roughly the same.

2. (this is vaguely related): I had a site that screwed up the canonical tag on the home page. The canonical tag was mistakenly edited to www.www.example.com that DID have an impact. The home page basically disappeared from search. I know that's not the same thing, but there are some similarities.

I have actually found the same thing as you in both scenarios. I also spoke to someone once who had a page unindexed because they had a canonical pointing to a non-existant URL.

 

[MarketingGiant]

Canonical tags are very important, even more, important nowaday. Web sites create many web pages, and some create dup by itself and with a canonical tag it allows you tell search engine" hey this it the main page" so they index it instead of the dup or non-main web page


Not using canonical tags can also decrease rankings due to canalization, having the same canonical tag on multiple pages defeats the purpose and search engine most likely will ignore the tag. Which cause confusion and in turn lower rankings.

Rule of thumb:
Use Canonical tags when appropriate
Use for pages that have duplicate content, and point to main source.
Use one per page.

 

[Eric Rohrback]

I'm late to the game here, but canonicals can make or break you. I've seen cases where they're completely botched and organic traffic suffered. Soon after they were fixed, it was like a flip of a switch and things started working right. Like others have mentioned, canonical tags are very important for ecommerce sites.

 

[dank402]

having the same canonical tag on multiple pages defeats the purpose and search engine most likely will ignore the tag

Can you elaborate on this?

I have an ecommerce site that I am optimizing currently and we have added a generic FAQ page that incorporates the content from a number of other pages (i.e. installation, return policy, etc.). I was wondering if using canonical linking would be appropriate to point from these specific question pages to the main FAQ page as the source?

The FAQ page is new, the other pages rank well and receive decent amounts of organic traffic.

 

[Eric Rohrback]

The purpose of a canonical tag is to specify the best page on the site for that topic. It's a way to consolidate duplicate pages, or in the case of eCommerce, a way to make sure your parameters and session IDs are not being indexed by search engines.

Your FAQ page should really be unique content if you want it to rank or have any weight in Google (if that's the goal). If you point from all your other pages to that FAQ, you aren't really using the tag in the correct way. You might end up hurting the site in the long run. Think of a canonical as a "soft 301" in the way that you're not making a server side redirect, but search engines still acknowledge it as a redirect (to some degree).

Do you want those other pages to exist or do you want to send users looking for all your other pages to a singular source (the FAQ page)?

 

[DanLeibson]

Though there are some exceptions to what Eric is rightly describing.

For instance, we regularly come across CMS systems that will populate a URL for a "widget" in the FAQ. Basically where each question and answer will have it's own page. These are things that you can resolve by canonicalizing back up to the FAQ as the primary driver for bringing the canonical tag into existance was to use it to help eliminate programatic duplicate content.

 

[dank402]

Thank you. This helped me wrap my head around it.

Basically the site owner went overboard with an SEO years ago who told him he needed more more more pages on his site. These pages were not created with the user in mind, and created a rather painful user experience for an online shopper trying to learn about their product. However, a couple of them do get a lot of traffic and I want to best position those pages when the new site launches.

Unfortunately we are not being paid to write any new content at this time, so we proposed keeping the pages live so they can be found from search, but removing them from the main navigation in place of a single FAQ page that contains the same information as the pages in question.

I thought canonical linking might be a way around the duplicate nature of the content on those pages vs the FAQ page but your recommendation to make the FAQ page unique and avoid canonicals sounds like a better route.

 

[dank402]

Though there are some exceptions to what Eric is rightly describing.

For instance, we regularly come across CMS systems that will populate a URL for a "widget" in the FAQ. Basically where each question and answer will have it's own page. These are things that you can resolve by canonicalizing back up to the FAQ as the primary driver for bringing the canonical tag into existance was to use it to help eliminate programatic duplicate content.

Interesting. I do have the ability to do this to my FAQ "widgets" with #anchor links.

 

[DanLeibson]

Interesting. I do have the ability to do this to my FAQ "widgets" with #anchor links.

Not quite what I am talking about. These are seperate URLs (no # etc) that ONLY have a specific part of the page content on it.