Was Panda applied at sub-domain or root-domain level?

[TomCritchlow]

Does anyone have any case studies or examples of sites where a specific sub-domain was hit by Panda while other sub-domains were fine?

What's the general consensus on whether this was applied at the sub-domain or root-domain level?

My thinking is that Google already knows broadly whether a "site" is a root-domain (e.g. SEOmoz) or a sub-domain (e.g. tumblr) and that they use this logic when rolling out Panda.

I'd love to hear your thoughts and opinions though?

[TomCritchlow]

Hmm, I definitely think there is a page level element but Google have gone on record to say there's also a domain level impact.

I imagine that the definition Google holds of a "site" is used in this case rather than slapping the whole domain.

[DanDeceuster]

This may sound weird, but I think the answer to your question is no. Panda and its subsequent rollouts hit a few of my websites. But it seemed to me it was a page level thing, not a domain level thing. For example, Panda seemed to target some affiliate websites. I have lots of affiliate websites. Some of their rankings were untouched, while other pages suffered. Granted, any site that got hit by Panda got hit pretty hard. But oddly, several pages on those sites were completely unaffected.

To your point there, blogspot.com, wordpress.com and tumblr.com all have several sub-domains that were affected without the root domains getting affected, so what you are asking is if blog.mysite.com got hit, then did mysite.com get hit as well?

I think that just as pages rank and not domains, so were pages affected by Panda and not whole domains or sub-domains.