The last I heard, Google were linking subdomains with their – what would you call them? Canonical domains? So that if www.piggynap.com ranked for monkey toes, monkeytoes.piggynap.com wouldn’t rank separately. It would be indented and counted as one result and not two. This means that relevant subdomains still have a chance of ranking for a key term, (the main domain might even play the role of the indented result) but several subdomains couldn’t dominate the page.

I also thought that subdomains would find it a little harder to rank – given the choice Google would rather place a proper domain first. Obviously this isn’t the case with domains like wordpress.com where separate blogs are owned by different people – presumably Google make an exception for hosting solutions like these. They still have to be careful however and avoid showing many results from the same site so I thought that in general they held subdomains at arms length.

So where am I going with all this? Well, I first noticed that Google were ranking their own subdomain in the form of books.google.com. It seemed that Google had made an exception for themselves. Then I noticed that a lot of the results were subdomains, whether or not the parent site was an authority.

I know this is all a bit subjective, but I just wonder if Google have started to allow subdomains more power. If I had my tinfoil hat on I’d say that they’ve changed their tune since deciding to put books.google.com into the results. But I’ve got no evidence, just some observations and a wild imagination. In the Webmaster Guidelines they say “Don’t create multiple pages, subdomains, or domains with substantially duplicate content” so who knows. Maybe that line will disappear just like the advice they used to have about buying a listing in the Yahoo! Directory.

If you’ve noticed anything weird going on with subdomains, leave a comment. Let’s be confused together.