Hosting Multiple Domains
As I was performing a review of a recent new client, I came across an interesting issue which I thought I’d share.

Over time, the client has purchased multiple variations of their domain name and simply parked them on their web server and pointed them all to their main site’s content.

While this is a common practice among site owners for various reasons, the issue becomes how you park those domains.

Before I get into the proper parking techniques, I should explain why some site owners would want multiple domain names.

There are many reasons: One would be to protect a brand. If you were to perform a Whois lookup on Google you would see that on one of their IP’s there are 78 domains hosted. Your first question might be “wait a minute – how can they have 78 domains on one IP yet penalize others for doing the same thing?” Well that’s a whole other topic and one I’m not willing to get into right now.

Instead, if you perform a lookup on the IP with 78 domains you will see variations on the spelling “google.com” such as “ggoogle.com” “gogole.com” and even “google.biz” to name just a few.

What Google is doing, besides protecting their brand, is also anticipating spelling errors. So if you type in ggoogle.com you will go to google.com.

Now to the good stuff – how to properly redirect these domains:

When I was reviewing the new client’s site, I came across all these domains so the first thing I did was connect to them with my browser. And of course I could connect to most of them, and I saw the same content as the main site.

The problem arose, however, when I started digging into the more technical details, such as checking server headers.

Server headers are part of the system which tells the browser (and of course search engine spider) how to deal with a site. When the browser requests a site, they also receive a header which tells them if the site is OK (200) or has been redirected (as indicated by a 301 or 302). There are many other header codes, but for now I’m going to focus on these three.

When I checked server headers for those other client domains I got 200 server headers on them all. While this may seem like a good thing, in terms of search engines it is not.

You see, when a spider comes and receives that 200 code, it assumes it is safe to index all the pages. It has no way of knowing that it shouldn’t be indexing this content. If, down the road, the engine finds that this is actually duplicate content (which it is – the only difference is the domain name) it could take action against one or all of the domains up to and including banning them all from the index.

Granted this is a worst case scenario, but one to be aware of. We’ve all heard of duplicate content penalties and of the problems associated with them, such as trying to get a site back into the index which has been banned. In the past I’ve heard of sites which have changed domains only to have the new domain penalized for duplicate content even though the old domain was no longer being used.

Duplicate content is a tricky subject because one never knows how a search engine will react. They may overlook the duplicates, or they may remove entire sites from the index for duplicate content.

So when I saw that most of these domains all resolved properly, and displayed the same content with a 200 server header I knew there could be problems.

The next thing I did was check some of the domains in the search engines to see if they had in fact found them. And I did find that there were domains which had been partially indexed by the search engines. Both Google and Yahoo! had found some of the aliases. As I mentioned above, this could lead to a penalty imposed on the main site.

So how does one properly handle multiple domains?

The best advice is to place 301 redirects on all domains. That way, no matter how the engine finds any of the aliases (which could come through a shared bookmark, a forum post, or a blog entry) they will know not to index the site and instead index the domain to which they are redirected.

(Further to this and as a side note, you should also place this redirect on your non-www alias. In other words, google.com should be redirected to www.google.com. I can’t tell you how many sites I’ve seen where this isn’t the case or worse, the non-www alias doesn’t even work and you get a 404 error).

A 302 will not work in this case – and in fact can do more harm than good. Because then what happens is the domain from which you want to redirect actually inherits the content of the real domain. This is what is referred to as 302 domain hijacking: When another domain gets credit for your content.

So if you own multiple domains, but only have them parked, consider placing 301’s on them until you need them. That way you don’t have to worry about someone finding one, bookmarking it or blogging about it, and then before you know it your site is banned from the engines.