“Canonical tags take precedence over 301 redirects in indexation decisions.”
301 redirects have long been viewed as the standard for canonicalizing a URL and is what I’ve primarily relied on to control indexation issues such as duplicate content. And, for the most part, 301 redirects did the trick. But what happens when it doesn’t? What happens when the redirected URL continues to be included in search results instead of the destination URL?
This exact thing happened and left me a little confused and frustrated as I continued to explain to the client week after week that sometimes it takes a while for the effects of a 301 redirect to be reflected in search results. Which, in my defense, is generally true. But I eventually grew tired of this excuse–as I’m sure the client did–and decided to investigate further.
To set this up, here was the issue: non-www versions of URLs were 301 redirected to the canonical www version, yet the non-www versions were indexed instead of the www version.
The first thing I did was check Webmaster Tools to confirm that the preferred domain was set to the www version. It was.
Next, I took a look to see if there were multiple redirects that could be causing some confusion for search engines. Nope. The redirects were one-to-one. No excessive redirect chains to be found.
I looked in the XML sitemap to make sure the www version of URLs were included and not the non-www versions. And they were. Nothing there either.
Then…I called for help.
I reached out to Jordan Kasteler, friend and Red Door co-worker, to get a fresh perspective on the issue. I explained that I had checked and doubled-checked everything I could think of–from the Webmaster Tools domain configuration to redirect chains to the XML sitemap. Then he recommended trying something rather simple that I hadn’t thought of: in addition to the 301 redirect, add a canonical tag to the non-www URLs to reference the www URLs.
I entirely overlooked this as a viable solution. Why? Because I was so deep in the belief that a 301 redirect was the absolute best you can do to tell search engines which page was the canonical.
I set off to implement Jordan’s recommendation with renewed confidence that this was all going to work out. That’s when I made the discovery–I had finally found the culprit.
In reviewing the pages, I learned that the non-www version of URLs, while 301 redirected to the www versions, also contained…A SELF-REFERENCING CANONICAL TAG! This was the “aha!” moment that led to this blog post.
The self-referencing canonical tag prevented search engines from indexing the destination URL. Meaning, canonical tags take precedence over 301 redirects in indexation decisions.
Now, of course, this is only one scenario and results may vary. But after this experience, I’m definitely starting to look at 301 redirects differently. Maybe it’s just not what it used to be–and, maybe, the canonical tag has a bit more authority than we all thought.