The majority of SEOs out there know that you should avoid duplicate content – avoid copying content from other domains, as well as your own. But what happens if you are forced to use duplicate content from your own domain? There is a tag that will solve all of your problems:
<link rel=”canonical” href=”http://www.example.com/original-page/” />
First of all, check out the image on this post! That is awesome. A real life example of dealing with duplicate content.
OK now, so you should make all the content on your site unique – but what about A-B testing? If you are going to test a landing page, and were testing something other than page copy, then it wouldn’t be scientific to vary the content up. What if you have products that fit in a variety of categories, and will end up with duplicates of your product page? What if you have query parameters in your URL that are possible to pull the same pagedata?
These are just a few examples of when it is appropriate to use the rel=canonical tag. But what does it do?
The rel=canonical tag is a signal to the search engines that you are fully aware that you have content on your domain that is duplicate or similar to content found at another URL. It contains an href attribute that lets search engines know which is the most important version to give credit to. This way, if someone links to an alternate version, Google will know to which page to attribute that credit. Here are some examples of proper implementation:
So let’s say you are testing variations on landing pages for PPC or for SEO with website optimizer and you want to make sure you aren’t splitting your credibility with the search engines on the original page. Take all the variations that you have created for testing purposes, and put the rel=canonical tag referencing the original in the href. It will preserve all of the link credit, allowing you to conserve your rankings during the test.
Multiple Paths to Products
Let’s say you sell rugs. You have a rug that doubles as a bath mat as well as a Shamwow (I know this doesn’t exist, but humor me for this example.) It is likely that you’re going to have the product listed under your ‘bath mats’ section as well as your Shamwow section. If you don’t make two unique entries for this one product, you’re making Google choose which category is the most important. Why make an algorithm choose for you, when you can choose intelligently yourself?
Choose the product that you would like to come up in searches. I would start by doing some keyword research and consider your current rankings. If your rankings for ‘bath mats’ are doing well, and gets more search volume, make that your canonical. If your rankings for ‘shamwow mats’ are phenomenal, and you can work with the amount of search volume that is going on, canonize that page.
Different Query Parameters to the Same Content
Let’s say you sell used exotic cars online (again, bear with me for a minute). You can sort these cars on the website by several query parameters that pull from the database. Your preferred method would be to access it by this URL:
But let’s say you also tagged your URLs with a query parameter to track certain kinds of traffic. That means you (and search engines, too) can access the same content by going to this site:
If for whatever reason you have multiple paths to take users to the same content, you can use the rel=canonical tag on the page to reference http://example.com/product.php?item=Ferrari599 to avoid any duplicate content issues.
Hopefully this has been informative in understanding the rel=canonical tag. Drop any questions in the comments and I’ll respond.