Rel canonical tag back to the same page the tag is on?

[activitysuper]

Very simple,

Why would a website (and I have seen tons doing this) link the rel canonical tag back to the same page the tag is on?

Example: somepage.htm has a canonical tag linking to somepage.htm

I thought the idea of this tag was to tell google if 2 pages are similar, this page is the original, and it's this page which should be indexed and the page with the tag on should pass all PR to the original.

Maybe im wrong and someone can help me out to understand this.

[te_c]

Hi,

I am not an expert, so please do not take my answer very seriously. What you mention, of making a canonical tag pointing to the same URL, looks fine. In my understanding, canonical tags were created to tell the search engines that a page is the right one, even if the system you are using creates address that could look like duplicate content. For example, if you are using a Content Management System like wordpress or Joomla, you could have the following:

• http://domain.com/page1

• http://domain.com/tag/page1

• http://domain.com/category/page1

• http://domain.com/date/month/page1 and so on.

Search engines (again, I am not sure, I am just a newbee), could think all this pages are duplicate content, and could penalize you for this. But if you indicate with the canonical tag that the right url is http://domain.com/page1, then you are safe.

I hope somebody with more experience could help you better,

Best Regards,

Daniel

[TomRayner]

There's a couple of reasons why people might want to do this (and why I do with all my websites)

First of all, the page/site might be scraped and replicated by a bot, particularly if it's an authority domain.  Having your canonicals in place to begin with will help reduce the chance of your content being seen as duplicate, should a bot scrape your site.

Another reason would be if a website might generate any additional versions of the page through queries, eg www.domain.com/page.php?query2 - Having a self referring canonical will also tell Google that you want to rank the URL without any other queries, which can help prevent any of those queries appearing in the Google index and/or SERPs.

[Highland]

Newegg.com uses this because they have affiliates, searches and numerous other things that affect their query strings.

Remember that ANY change to the query string is seen as a new page. So

domain.com?page=a&link=1
domain.com?page=a&link=2

are considered separate pages, even if they return the same content.

Canonical is used to determine which duplicate page "wins" the index race. All other versions are considered duplicate and, thus, devalued.

[activitysuper]

Thanks Tom (and everyone else for the replies),

So if someone linked to a page with a querystring Google wouldn't index that page because the canonical tag is pointing to a url which doesn't have that query in?

I like the scraped part as well, that in itself makes it worth while.

[Highland]

Not quite. Canonical (per Matt Cutts) is considered a hint as to what the real page is. It doesn't stop the duplicate page from being crawled or indexed (a page that isn't indexed will not show up anywhere in Google for any query), it prevents the duplicate page from winning the duplicate race (i.e. if you don't pick a winner, Google will pick one for you).

[Dr-Pete]

For all practical purposes, Google doesn't seem to index pages where it recognizes the canonical as legitimate. You won't find them in a "site:" query, "cache:" command, etc. Google may call that a "filter", but once it's reached that point, the URL is as good as de-indexed. There may be subtle, technical distinctions, but the end result is virtually the same.