What's my best strategy for Duplicate Content if only www pages are indexed?

[kimmiedawn]

The MOZ crawl report for my site shows duplicate content with both www and non-www pages on the site. (Only the www are indexed by Google, however.) Do I still need to use a 301 redirect - even if the non-www are not indexed? Is rel=canonical less preferable, as usual?

Facts:

• the site is built using asp.net
• the homepage has multiple versions which use 'meta refresh' tags to point to 'default.asp'.
• most links already point to www

Current Strategy:

• set the preferred domain to 'www' in Google's Webmaster Tools.
• set the Wordpress blog (which sits in a /blog subdirectory) with rel="canonical" to point to the www version.
• Ask programmer to add 301 redirects from the non-www pages to the www pages.
• Ask programmer to use 301 redirects as opposed to meta refresh tags & point all homepage versions to www.site.org.

Does this strategy make the most sense? (Especially considering the non-indexed but existent non-www pages.)

Thanks!!

[jesse-landry]

This is an incredibly easy topic to address because you've already laid out exactly what needs to happen.

In other words, yes! That strategy is exactly the way you should go.

Good job and good luck!

[kimmiedawn]

Thanks for easing my mind Jesse! One thing still confuses me - the fact that the non-www pages are not indexed. They are not disallowed in robots.txt, there are no rel=canonical tags (except for pages in the blog subdirectory), they are not meta-refreshed, and obviously not 301 redirected. Could it be the doing of a sitemap (though I cant find one)? Or did Google simply decide all the www pages were more relevant? Am I missing something here? I don't want to ask the programmer to add a ton of 301 redirects (which I did) only to get a 'DUH!' response;)

FYI - site is asp.net - not sure if that matters, except when they redirect homepage to avoid creating infinite loop. (right?)

Thanks again!

[jesse-landry]

Well the reason Google has picked the www version as it's preferred version automatically is most likely because of all the links you mentioned that were already pointing to that iteration of your domain. Google can figure this out on their own. That said, it still sees both of the sites (non www and www) as duplicates of each other. Best practice is to 301 one to another.

I've waged this war with a programmer before so I know how it goes. The one I dealt with didn't think there was any reason and told me all websites work that way. So I asked him to go to http://google.com and tell me how it resolves. Repeat that step with every major brand you can think of until he/she gets the point and that might help you.

They should be able to 301 this one time, no matter whether they're using an Apache or IIS server. This should be a quick fix. If they're unsure how to do it, have them Google "IIS 301 redirects" if it's a Windows server or "htaccess 301" if it's a Linux/Apache server.

[kimmiedawn]

Very informative - thank you! It seems when I think I have a relatively firm grip on SEO, I stumble upon something new - like the dangerous potential for an infinite loop in a 301 redirect in IIS. (That was Chinese to me two days ago;))

Your response solved my concerns - hopefully it will help somebody else when they face the same problem.