Rel=canonical vs noindex/follow - tabs with individual URLs

[Digitator]

Hi everyone

I've got a situation that I haven't seen in quite this way before. I would like some advice on whether I should be rel=canonicalzing of noindexing/following a range of pages on a clients website.

I've just started working on a website that creates individual URLs for tabs within each page which has resulted in several URLs being created for each listing:

Example URLs:

• hotel-downtown-calgary
• hotel-downtown-calgary/gallery?tab
• hotel-downtown-calgary?tab
• hotel-downtown-calgary/map?tab
• hotel-downtown-calgary/facilities?tab
• hotel-downtown-calgary/reviews?tab
• hotel-downtown-calgary/in-the-area?tab

Google has indexed over 1500 pages with the "?tab" parameter (there are 4380 page indexed for the site in total), and also seems to be indexing some of these pages without the "?tab" parameter i.e. ("hotel-downtown-calgary/reviews" instead of "hotel-downtown-calgary/reviews?tab") so the amount of potential duplication could be more. These tabbed pages are getting minimal traffic from organic search, so I've got no issues with taking them out of the index - the question is how.

There are the issues I see:

• Each tab has the same title as the other tabs for each location, so lots of title duplication.
• Each individual tab doesn't have much content (although the content each tab has is unique).

I would usually expect the tabs to be distinguished by the parameters only, not have unique URLs - if that was the case we wouldn't have a duplication issue.

So the question is: rel=canonical or noindex/follow? I can see benefits of both.

Looking forward to your thoughts!

[CleverPhD]

Well, you are on the right path with thinking on how to reduce the amount of unneeded pages.

Here is how I would approach it.

1. Check the query volume levels on those specific queries "Map to XXX"

2. Check the search volume on those pages.

See if you can detect a pattern that there is search volume to justify those searches, do they result in significant traffic to those pages.  Then try and determine, what is the content on all those separate pages and is it any good?  Are they making extra pages to make extra pages?  Sure, in theory you could do a page per query, but I would bet if they have a ton of hotel, all the info on those pages is a bunch of boilerplate crap copied from somewhere else.  Even if the search volume was there, do they have a good enough page with good content to rank for it?

Now that we have hummingbird type algos in Google, it reduces the need to get so specific for matching on queries on a page by page basis.  Build a single, awesome, page that is really helpful to users and has original content, that is how you win for the big queries and then fill in for the rest. You can then use the title, description and H1, H2 headings to show the important information.

Remember that the rel=canonical will help Google understand what your main page is and what your secondary/duplicate pages are in this specific case, but I am not sure that Google would see it as the consolidated awesome single page.

Rel=canonical is more for showing how the parts are just parts of a whole page that is already there, it is more to help clean up duplicate content

http://googlewebmastercentral.blogspot.com/2013/04/5-common-mistakes-with-relcanonical.html

As I read your post originally, this seemed to be more of the issue.  This was why I was encouraging you to take the content in the tab (that did not seem substantial) and put it on the main page and use the canonical or if the content was junk then it did not matter as much.

Hope that makes sense.

[Digitator]

Hello CleverPHD!

Thank you very much for your response - that really clears a few things up. Most info around rel=canonical revolves around duplicate content so I just wanted to make sure!

Regarding your separate question, your assumption is correct, each hotel on the site is set up this way (and there are hundreds of them) which is why I am concerned.

I've just come into this project but I assume that the initial thinking was going to be that they wanted to rank for a range of search terms around the hotel i.e. "Xxx Hotel reviews", "map to Xxx Hotel" etc and thought that dedicated pages for each term would help them.

What I am hoping is that by rel=canonicaling these pages to the main Hotel URL we will in effect be creating the "really awesome" page you refer to, while avoiding any potential penalties!

Once again, thanks for your in depth response - it's very much appreciated!

[CleverPhD]

The rel=canonical is what you need to do right now to help fix this.  Google has already indexed all of those tabbed pages and you need to make sure that Google knows that they are subsections of the main page and how it all fits together.  The canonical is treated like a 301 redirect.  Doing that should take care of the extra pages indexed.  If you can 301 those tabbed pages that would be a good option as well, not a good option if you do not want to lose the content on those tabbed pages though.

Right now, I would not use no-follow on links or no-index meta tags on the tabbed pages. If you use no-follow, you are telling Google to not follow that tab link and so it is not able to crawl over the the tabbed page and see the canonical link.  Second, if you have a noindex meta on that tabbed page, it tells Google to take the page out of the index and would probably conflict with the canonical link as well.

The only way you would want to use no-follow or noindex meta tags or blocking in robots.txt is if 1) the content is not worth indexing and/or ranking and or 2) it is not in the index already and you want to keep Google out of this stuff.

Ultimately, I would try and get the stuff on the tabbed sections onto the main page and then use the noindex meta tag on the tabbed pages so that you can get rid of the tabbed pages all together and not work about losing the content that is in the tabbed section.  If you can do that, it would be a better approach than the canonical option, but it looks like you may not have that option at this time.  All those extra tab pages are just wasting Google's time crawling pages that do not matter (most likely).

If you did not care about the content on the tabbed pages, I would just 410 them right now and get rid of the links to the tabs on the main pages.

Hope the above makes sense, tried to answer the "it depends" scenarios.

Feel free to ignore this next part, but I have a separate question if you drop the whole tab pages issue and look at this at a higher level.  Looking at this subsection of URLs from above

• hotel-downtown-calgary
• hotel-downtown-calgary/gallery
• hotel-downtown-calgary/map
• hotel-downtown-calgary/facilities
• hotel-downtown-calgary/reviews
• hotel-downtown-calgary/in-the-area

Would not all the above pages really need to be about a single hotel or a category page on all the hotels in downtown calgary? Unless you have enough search query volume to support a separate page for each of them, seems like if you put the content on all the pages listed above into a single page, you would have a really awesome page about either a specific downtown calgary hotel or all the hotels in downtown calgary.

[Digitator]

Thanks for that Dana - appreciated.

At the moment it is currently set to Let Googlebot decide and I need admin access to look under the hood which I don't have, so I'll have a look when I gain administrator access. I can see one potential issue through - some of their landing pages don't exist without the "?tab" parameter which I think might make this a non starter unfortunately.

Given the possibility that this method doesn't work, does anyone else have any thoughts on  rel=canonical vs noindex in this situation?

[danatanseo]

Hi Damon,

I believe there may be a third option. In Google Search Console, there is the option to tell Googlebot not to crawl and index URLs containing certain parameters. In this case, it seems like it would be easiest to indicated to Googlebot that anything with the ?tab parameter should be excluded from crawling and indexing.

You can find this in Google Search Console by navigating to "Crawl" and then "URL Parameters," then click "Configure URL Parameters" and Google will show you a list of potential candidates for exclusion.

Hope that's helpful...it's my understanding that Yahoo/Bing have something similar but I've never used those.

Cheers,

Dana