Blog tags are creating excessive duplicate content...should we use rel canonicals or 301 redirects?

[VanguardCommunications]

We are having an issue with our cilent's blog creating excessive duplicate content via blog tags.  The duplicate webpages from tags offer absolutely no value  (we can't even see the tag).   Should we just 301 redirect the tagged page or use a rel canonical?

[evolvingSEO]

The easiest way to resolve issues with tags is to noindex them. I wrote a post about how you can safely do this: http://www.evolvingseo.com/2012/08/10/clean-sweep-yo-tag-archives-now (you basically just double check to see if they are receiving traffic, and leave the few that receive traffic via search indexed).

But at the root level it comes down to knowing how to use tags correctly on a blogging platform to begin with - and knowing how they function, and what happens when you tag something.

First off, tagging any post creates a new page called a "tag archive". The only way someone can get to tag archives by default is if you allow some sort of navigation or links to them on the site itself. This is usually in the form of a "tag cloud" (sidebar or footer) or at the bottom of posts when it says "tagged in....." and links to the tags.

Then if they are internally linked to, they will get indexed (unless you noindex them like I have suggested above). They are typically low to no-value pages because most bloggers just tag everything, and use lots of tags per post. Then you end up with hundreds of pages (tag archives) with no value.

So noindexing them is the safest way to go, except for very extreme cases where a blogger uses them 100% perfect (which is rare, so I always assume most people asking should just noindex but use my post to check for traffic to any of them first).

[evolvingSEO]

Thanks for chiming in! Just to reiterate something - canonical tags are only a suggestion, not a hard directive. Google can and does ignore them. The canonical tag and also pass noindexing directives to the page you point them at. So with tag archives, if they are set to noindex and you canonical them to posts, you might deindex your posts.

And finally, canonical is only something that should be used that can't be solved via indexation, crawling or architecture solutions. In the case of tags in a blogging system (probably wordpress) the easiest and 100% definite way to handle tags is just to noindex them. Then you don't need to worry about canonicals or duplicate content.

Also, tags are no harmful because of duplicate content per se, but just that they add a lot of unneeded pages to the index.

[Travis_Bailey]

You can set tags to noindex/follow. If you're using WordPress and one of the more popular SEO plugins, this could be done with a couple of clicks. But are these tags actually generating duplicate content? Usually a snippet of the tagged posts isn't considered duplicate.

Anyway, noindex should be more effective than it was in the past. And as Highland has said, setting a canonical would be a good idea as well.

If the tags aren't really helping out site users, they aren't using them - etc., and they don't have any link equity - you could just 410 them. Plus you could submit the tag URLs for removal in GWT.

So check the referral traffic and backlinks for those pages and go with either removal or noindex follow and a canonical.

[Highland]

Canonical hands down. This is what canonical was made for anyways: duplicate content you can't remove.

Canonical simply lets you tell Google which duplicate content should "win" the indexation race and Google will take it into consideration. I can think of many reasons why you'd have overlapping tags but would not want to remove them (which is what a 301 would do)