Duplicate content via indexed query string

[CainIV]

Hello guys. I recently underwent some serious ranking fluctuations on a website I own. After investigating further, I had found that good ol G had indexed my home page both as http://www.example.com, and as
http://www.example.com/?page=about_us

I have always been a firm believer in plugging as many holes as possible using 301 htaccess (index.html - > root, non www to www etc, but had never encountered this before.

Naturally, this caused duplicate content, right at the homepage. Worse off, the page is not a traditional index or default page where Google may accustomed to better understanding that this is the same page.

I have since added a rewrite rule which addresses this page, and the last known cached is Feb 26.

The website, which formally had quite good rankings, has now slipped very suddenly and fast out of the top 10 pages in Google. I have combed up and down to find issues, but this is the only item left. Sherlock's theory would tell me this is the problem, but this doesn't help my confidence in Google finding the very obvious correct page that owns that content.

Since then, the homepage has been reached, but still no return for funland.

Has anyone else seen this sort of thing with their website, and if so, how long until this might return, if this is indeed the issue?

[edited by: tedster at 2:38 am (utc) on Mar. 17, 2009]
[edit reason] use example.com - it cannot be owned [/edit]

[tedster]

Yes, I have seen this kiind of canonical problem take down rankings. It's one of several dozen possible canonical issues [webmasterworld.com] that we've catalogued, and many of them are extremely difficult if not impossible for webmasters to address technically, depending on their hosting situation. That's why I am very pleased with the new canonical meta tag.

I've seen two incidents where a query string issue was resolved, and rankings returned within a week or so in both cases. If this really is the source of your troubles, you have every reason to be hopeful.

[CainIV]

Thanks Tedster. The 301 has been in place for about 5 days. If this is the issue, it literally crippled rankings....fast.

I will keep everyone updated. My sense is that if this is the case, it was accidentally indexed by G, but it could also be that another webbie pointed a link here since I am in a very competitive genre.

This is one of those canonical redirects I tend to forget to cover in my approach to plugging holes.

Would you advocate using the rel=canonical tag on the homepage of the website to help further cement against potential problems, in addition to using first sensible upstream logic in CMS and url creation as well as 301 redirection?

[tedster]

Well, the canonical tag is still pretty young, but so far there's been a lot of good reports about it. I'm currently overseeing a new site build on an IIS server -- and I'm recommending they use the canonical tag from the start, especially because they have not yet installed ISAPI Rewrite. IIS has a lot of canonical problems right out of the box.

This might be a good place to point out jdMorgan's masterful thread in our Apache forum: A guide to fixing duplicate content & URL issues on Apache [webmasterworld.com]