SEO best practices has traditionally stated that subfolders are the ideal choice for most websites, but when the Google Panda update hit, webmasters all over the web were thrown into a massive suckhole of swirling obscurity. And, ironically, a lot of the sites affected by this update weren’t publishing a lot of low-quality content. So when Matt Cutts came on the scene Post-Panda and told HubPages to “try subdomains” among “other things” to recover from their 50%+ drop in traffic, SEOs started to question conventional wisdom. Should we use a subdomain or subfolder in our site architecture?
 
I looked all over the web for relevant information on the topic, but, surprisingly, not many people have written about this argument since Panda allegedly turned the tables on subfolders. In frustration, I tweeted about it, looking for some resources. I ended up having a very enlightening discussion with Alan Bleiweiss (@alanbleiweiss) from Click2Rank about the merit and potential new-found value of subdomains. Let’s start by answering the questions “what is a subdomain and a subfolder?” and “What are the differences between the two?”


 
There’s no easy “one size fits all” answer for the subdomains or subfolders question, and which one you use depends highly on your site’s individual needs and niches. Here’s a quick overview of what subdomains and subfolders are and how they’re different.
 

Subfolders
A subfolder is a container that resides under a parent folder on a server. It represents one step in the site hierarchy, generally beginning with the home page as the main parent with extra pages added on as subfolders. The URL structure of a subfolder generally follows this format:

 

http://www.maindomain.com/subfolder/

 

Most websites use subfolders in their site structure, and until recently, subdomains have seen relatively limited exposure as an SEO strategy. Open-source content management systems like WordPress and Joomla use this structure as a default.

 

Subdomains
A subdomain is a domain that is a smaller part of a larger (or root) domain. In a search engine’s eyes, a subdomain behaves more like its own website than a subfolder, and not as much (and some SEOs say none at all) authority is transferred from the root domain as with a subfolder. A subdomain’s URL structure is as follows:

 

http://subdomain.maindomain.com/

 

or

 

http://www.subdomain.maindomain.com/

 

Most SEOs argue against using subdomains except under very specific and applicable circumstances. Since it’s essentially like managing an entirely new site, inexperienced webmasters are counseled against using them.

 

Has this traditional belief changed with Panda? Let’s take a look at the HubPages “Panda recovery process.” HubPages is an online publishing platform that allows their users to easily monetize their content. They provided a space for creating the content (much like Blogger or WordPress) and added a monetizable infrastructure. And, until just recently, they used a subfolder of hubpages.com for each user.
 
When the Panda update hit, HubPages saw a steep drop in traffic—reportedly over 50% of their pageviews vanished—and they were at a loss as to how to recover. They appealed to Matt Cutts, Google’s spam department head and Google’s general liaison to the SEO community, for ways they could land back in Google’s good graces. Cutts advised them to “try subdomains among other things.”
 
HubPages implemented Cutt’s advice by moving all of their users from a subfolder(hubpages.com/user/) to a subdomain (user.hubpages.com), much like the structure seen on WordPress and Blogger sites. In the above linked post on HubPages’ blog, they state that doing so “should allow each author to be judged by Google separately” and not damage the overall integrity of the site.
 

What Worked for HubPages?

It would appear at first glance that by pulling all of the users into their own subdomain helped HubPages resolve their Panda-related woes by allowing Google to judge each user’s HUB as its own entity, without damaging the integrity of the entire site. By giving each user their own subdomain, HubPages is telling Google “each of these pages reflects a separate author” so that, when Panda finds something worthy of a smack-down, it damages only the content on that subdomain.

 
So when should you use subdomains instead of subfolders? Before you go out converting your entire site to subdomains from subfolders, let’s take a look at this argument with a post-Panda perspective. I peppered Alan Bleiweiss with questions about subfolders and subdomains. Here is an excerpt from the email I received from him:
 

 
Subfolders, he points out, are best for most cases. If your site has a fairly narrow topical focus (movie reviews, an SEO blog, a strictly news site) then subfolders allows you to transfer a lot of the authority that your root domain has to these pages to help them rank quickly. It also makes navigation a whole lot easier.
 
If, on the other hand, your site covers a wide variety of topics (product reviews, in-depth book reviews) or has a lot of user generated content (like HubPages, WordPress or Blogger), then separating your content into subdomains will net greater results, assuming you have the SEO resources to do so. In the former case, a tighter topical focus shows Google that your site is focused and relevant, allowing it to rank higher for associated keywords. In the latter, separating user generated content into its own subdomain allows Panda to penalize low-quality content and leaves the quality content producers untouched.
 
What do you think? What does this mean for you? Are you thinking of changing your site structure as a result of Panda and HubPages’ impressive recovery? Are you content with subfolders? Let me know what you think in the comments.