301 redirection loop to generate duplicate content

As many of you know, a 301 redirect is the solution for many SEO issues. You may wonder the reason for this article then.

Very recently, I provided support to a user in the Google Webmaster Forum claiming the 301 was not work correctly. Particularly, he was complaining on the GWT reporting some pages previously redirected having duplicate meta tags title and description.

Before further carrying on with the article, a distinction needs to be done: not all the tools around the net are reporting information in the same way.

More in details, Google Webmaster tools reports on a URL basis, whereas Analytics for instance reports on a page basis.

Which is the difference? Simple! Not all the URLs have a corresponding page, but all the pages have at least one URL.

What was happening?

Despite the redirects, GWT was claiming duplicate tags for the redirected URLs / pages. How was this possible?

The old web site – a forum - internal links’ structure details were necessary, but the user was not able to provide it. So we have tried to reconstruct what the previous web site was doing.

Pages on the site – all of them with a user friendly URL - were containing a link to another no-longer existent URL, which with the new configuration were redirecting back to the original URL with an hash fragment to the end.

To be clear: www.test.abc/page-with-URL-A -> link to www.test.abc/inexistent-page-with-URL-B -> 301 to www.test.abc/page-with-URL-A#somewhere-on-the-page

As named anchor are normally ignored from Search Engine. However, for whatever reason, the new 301 structure in place generated a loop, and by the time Google re-crawled the page(s) it assimilated the two URLs (technically the same) as different pages thus reporting a duplicate content.

I don’t want to comment on the error per se; it is appreciable the fact that even though the information is incorrect, at least something was reported thus helping us solving an issue that could have degenerated somehow, and which up to a degree was already affecting the ranking of some previously high-performing pages.

The solution has been found

It was impossible to ascertain since how long the loop was in place; however, by the time the issue was resolved the errors in the GWT console disappeared and the ranking was progressivly reinstated.

Therefore, my advice at the end of this story is always to pay attention to both your internal link strategy as well as to all unnecessary 301 redirects.