When optimizing a subdomain for search engines, you must decide if the sub-domain should be viewed as part of a bigger mega-site or as a separate, standalone site.
Unfortunately, many people are misled by SEO fallacies.
In many cases, creating effective sub-domains is simple and worthwhile.
Every time someone claims that subfolders are better for SEO than sub-domains, Wikipedia bursts out laughing at the top of the search results.
As ridiculous as the sub-domain-versus-subfolder debate is, individuals continue to wage war on sub-domains with the most illogical of justifications.
Now, I’ve addressed those flaws in previous posts, so I’ll get right to the point with this one.
Subdomains are not for the weak of the heart.
They are used for crowd-sourcing sites (such as en.Wikipedia.org), large Websites (such as money.cnn.com/business), Web hosting sites (such as Blogspot and WordPress), personal blogs (such as blog.michael-martinez.com), corporate blogs (such as blog.adobe.com and blog.twitter.com), and just about anything else you can think of.
You can’t manage subdomains if you can’t handle any of those types of sites.
It is not shameful to concede failure.
But don’t worry, whatever SEO technique you develop for a subfolder will also work for a sub-domain.
It makes no difference to search engines whether you use subdomains or subfolders.
Sub-domains are distinct hosts.
This is true for browsers as well as search engines.
That doesn’t make them distinct websites.
Before you start throwing cheap jabs about toying with semantics, consider what makes a website a website.
Websites are deceptions.
Each URL is its entity.
Sub-domain URLs now seem differently from subfolder URLs.
The sub-domain name comes before the root domain name, whereas the subfolder name comes after the root domain name.
That’s all there is to it when it comes to the variances in URL structure between subdomains and subfolders.
Subfolders can be nested in both subdomains and subfolders.
As a result, they are rather unique.
In a single subfolder, you may host a million URLs, each of which can go to a distinct Website.
They’d utilize various themes, work with different writers, and have their logins.
In truth, there are a few single-page website hosting that offer this service.
I’m sure you’ve heard of at least one of them: Twitter.
If you’re providing encrypted material, each host must have its SSL certificate, or you’ll have to pay for an expensive wildcard certificate.
Subfolders might be distributed across several IP addresses, with distinct SSL certificates assigned to each.
But how many of you do that?
Are there any volunteers?
I can’t find any.
Any sub-domain might be hosted on a server other than the original domain.
All you have to do is add an “A” record to your DNS zone, and the nameservers will redirect everyone in the right direction.
Hosting a subdirectory on a separate server is a little more difficult.
You’ll need to set up a reverse proxy, which is a difficult and ugly process.
That is something I never want to have to do.
But it is possible.
So, there isn’t much that separates a Sub-domain from a subfolder — you can put up a DNS entry for one but not for the other.
You can pretty much do anything else for either of them.
Documents are linked together using links.
We develop navigation widgets like menus, tabs, and boxes with link lists, which we utilize to connect the pages of our Websites.
The search engine is oblivious to the fact that the page it is crawling is part of a “site” (that is, a collection of pages).
It will deduce this from numerous signals such as linkages between papers, metadata (maybe), page design (perhaps), and relationships between documents.
It is feasible to build a Website such that each page has a distinct appearance.
That’s something I’ve done.
I accomplished it with a rather large website.
If I ever try that again, shoot me.
So, as far as you and I are concerned, the navigation is what distinguishes the Website.
A search engine may be more demanding, but you and I know what a Website is because we look at the navigation in the browser and say, “Oh, this is a Website with a home page, an about page, a blog, and so on.”
Websites, in other words, are abstractions that we create in our minds by using methodical criteria.
As a result, a Sub-domain may be as much a part of a site as a subfolder.
You may also separate a Sub-domain or a subdirectory from the rest of the site.
That’s something I’ve done as well.
You’re doing it incorrectly as long as you assume that Websites are hosts and hosts are Websites.
That is not how the Internet works.
People make it appear that way.
To keep up with our delusory perceptions of the Web, search engines need to perform a bit more effort.
For far too long, people have been listening to phony experts explain the fictitious distinctions between Sub-domains and “genuine” Websites.
You’ve been educated in a particular… well, school of thinking.
It’s a subjective interpretation.
You can start developing effective Sub-domains once you stop persuading yourself (and your clients) that Sub-domains are distinct from subfolders.
If you can disregard the fictitious barrier between Subdomain and root domain that your educated mind wants to erect in this phantom of a site we’ve made,
you’ll be able to resist the impulse to add “rel=’nofollow'” attributes to all links on the root domain referring to the Sub-domain.
That was never a good idea.
It’s the same as using “no follow” attributes on links to subfolders.
Why would you want to do something to a Subdomain if you wouldn’t do it to a subfolder?
If you publish information on the Sub-domain that you want to be seen in search engines, don’t you want all the links pointing to it to help?
No, employing “rel=’nofollow'” attributes does not allow you to “hoard PageRank.”
“Link equity,” like “link juice,” is one of those SEO aphorisms that barely make a smidgeon of sense (to me).
Many other people understand it, but if you ask them what link equity is, you could humiliate them.
Don’t do that in front of the customers.
I’m not sure where this insane concept of “dividing equity” comes from.
It’s possible that it leaked over from warnings about generating duplicate material.
There is a situation in which you may post identical material on many URLs and get links on all of them.
Consider how people write their blog articles on their domains and Medium, or how they publish their blog posts on their domains and The Conversation.
If the copies of the material connect, they will exchange PageRank-like values (I believe – link equity).
Some will point out that Medium/The Conversation receive a portion of that PageRank.
That is correct.
But then our philosophical friends will wonder if the connections to Medium/The Conversation would still exist if you hadn’t syndicated the information.
In other words, if you have a Subdomain (Alfred) and a root domain (Doris) and marry them via their navigation, any PageRank-like value they collect would travel to each other in the same manner that Alfred would if Alfred were a subfolder.
PageRank and its ilk are determined as a value flow between documents rather than between hosts or sites.
Furthermore, if anybody (you, your syndication partner, or the search engines) is canonicalizing duplicate material to a single source,
Google claims that all signals are merged into the primary source’s profile.
As a result, no “link equity” is divided.
Everything is fine.
And I believe Bing is the same way.
You can only “split equity” if you sever the link between your multiple link-attracting papers.
That may happen whether they are on different root domains, Sub-domains, or all in the same folder in the same root domain.
That’s because it all boils down to navigation — the linkages between the various page URL.
So, if you’ve read this far and understand what I’ve stated, it should be clear that anything you do to make subfolders successful also works for Subdomains.
So, if you want a nice, easy-to-remember URL – and you want the content to handled as if it’s part of the same site like everything else on the domain
Choose a Subdomain if it looks better than a subfolder, and choose a subfolder if it looks better than a Subdomain.
That is, follow your instincts about URL aesthetics since that is all that matters.
That, plus you’ll have to do some more effort to set up the Sub-domain.
However, there is no reason not to set up a Sub-domain as a fully distinct site.
If you (or the customer) were thinking of starting a blog on a separate branded domain, you may save the hassle of acquiring a domain by creating a brandable Subdomain instead.
That’s exactly what en.wikipedia.org did.
So far, it has served them well.
It.wikipedia.org… and de.wikipedia.org, too.
They could be onto something with this Sub-domain business.
Sub-domains are excellent candidates for content differentiation based on language.
You are not required to do so, but it will make it simpler for you to sort things out in your head.
You have a root domain for your native language, and all other languages can have their Sub-domains.
Of course, some individuals just place the additional languages in subfolders.
It’s six of one and a half dozen of the other.
But, regardless of how you handle it, be sure to utilize appropriate hreflang markup.
Sub-domains can have their themes and navigation, but if they constantly link back to the root domain (and vice versa), they will recognized as both unique sites and as part of the main website.
That’s an amazing effect.
(Note: This also applies to subfolders.)
Sub-domain SEO is only as difficult and enigmatic as you make it.
While it might useful to provide the appearance of a different site by using a Sub-domain,
I promise there are numerous orphaned subfolders and root domain sites spread over the Internet.
They are also independent websites.
Or they appear to be fragments of larger locations that exist only in the human mind.
SEO should not feel any more hard, hazardous, or difficult when dealing with Sub-domains than when working with a single root host.