The old question: Domain, Subdomain or Subdirectory

[bluemi]

Hi, I'm running thge website of a travel related company which - so far - offers trips to two different countries. We now need to make a few decisions for the future. Currently we are using the main domain www.ourdomain.com for the main destination and a subdomain destination2.ourdomain.com for the other destination. Question is, shall we stick with this? Or use subdirectories? Or two completely different domains, as the destinations are quite different? The company would like to have everything on only the main domain but I'm hesitating a bit since we have a good search engine standing for the main destination and I'm a bit afraid to "dilute" it by adding a second destination. What are your views on this?
Thanks!

[Andy Langton]

A subdomain does provide some degree of separation if you want to "protect" the current site. However, I can't see that it makes sense for the longer term. I would expect to be able to find a travel company's destinations all on one site. If it was a totally separate product, then a separate site or subdomain might make sense. Similarly, in terms of "dilution", I think you may be looking at this the wrong way. If I link to "Travel Company Ltd", I'll link to the homepage. This equity should then be distributed across all the offerings - that's not dilution, it's just the natural way of things! You won't find many travel companies that run multiple sites or subdomains just for destinations.

In a purely practical sense, it's unlikely to make a huge amount of difference in the short term. You'll still be passing the same link value to a subdomain as you would be to a subdirectory. You might anticipate a subdomain taking longer to rank, but potentially being stronger in the long term (if it can attract a greater number of links that a subdirectory would). But, all things being equal, you can make any set-up work. It's really a question of what the most logical approach is for the long term, and the approach that makes the most sense to visitors.

[bluemi]

Thanks for this, Andy. Interesting food for thought. I have probably not explained my fear of "diluting" enough. The company has been active since more than 20 years in the first destination and has built an excellent reputation there, in many ways. Therefore it does very well in Google for all kinds of search terms containing this destination. The second destination has been added only recently and is quite different. So by adding some amount of text about destination 2 to the main page I'm a bit afraid the search engine positions for destination 1 will suffer.

[EditorialGuy]

I'd just leave well enough alone.If it works, don't risk screwing it up.

[Andy Langton]

Thanks for the clarification, bluemi.

So by adding some amount of text about destination 2 to the main page I'm a bit afraid the search engine positions for destination 1 will suffer.

To play devil's advocate a little, how will customers know about the new destination if it isn't mentioned on the main page in some capacity? In general, it seems like the approach is likely to hamper the potential performance of the new destination.

Your current structure seems to be like this (assuming you added an extra destination):

travel.tld (destination 1)
---destination2.travel.tld
---destination3.travel.tld

But assuming that all destinations would eventually have equal weight, you'd be likely to be aiming for this:

travel.tld
---travel.tld/destination1
---travel.tld/destination2
---travel.tld/destination3

That said, you may well be right to be cautious, as EditorialGuy suggests. You would likely have to accept that, at some point you would need a longer-term solution, but perhaps the risk might be too high at the moment?

[dipper]

I agree with EditorialGuy - if it ain't broken, don't fix it.

I've personally fallen foul of acting when it was not required - every move *could* result in a loss of traffic, and if you have not suffered a penalty or got on the wrong side of an algorithm, it's not a consideration to worry about.

Stick with what you have and if required add a new sub-domain for any new (3rd+) destinations.

[thedonald123]

I'm facing a similiar situation, except I already went with the decision to add the new content for the similiar but new niche on a subdirectory. The new content was added slowly starting about 9-12 months ago. Google has in the past considered the new niche added to be a distant synonym to the old established niche.
We were hit by Panda 4.2 "slow rollout" on 9/2. Because of the slow rollout I didn't realize until recently.
As an attempt to recover I have added meta robot noindex tags to all the new content, plus removed most internal linking to that subdirectory and adding rel=nofollow to any internal links still left. Tiny improvements so far, but too early to tell.

Now, the reason I was hit by Panda is probably unique to my situation with the new content being initially treated by Google as duplicates of the original since they are distant synonyms. Thus Google showed the original authoritative page in the SERPS when the new page was the appropriate choice.

However, I think the takeaway can still be that adding the new content on a subdirectory will dilute the sites authority for your original topic and hurt your rankings for the original topic.

However, the solution is not simple at all. A subdomain also has problems. I believe that there is no real difference between subdirectory and subdomain in Google's algorithm unless you treat them as 2 separate sections. If by your internal linking you indicate to Google that the new content is an integral part of the website then Google will consider it the same website regardless of it being on a different subdomain.
I am not sure if the same can be said about a subdirectory. (Would love to know as it effects which direction I go)

So your choices are
1. new domain.
Pros: no real risk in hurting original sites traffic.
Cons: Maintaining 2 sets of code, New Branding for 2nd website. Need to be careful that interlinking the 2 websites doesn't attract the wrath of Penguin. But, if you don't interlink them then new website starts from scratch and building traffic will be difficult.

2. Put it on a subdirectory, accept the temporary dilution of authority for main topic and the loss of traffic which ensues, wait a few months and build links to new content, or if you believe in magic then wait for the new links to come by themselves because the content is such high quality. ;-) At some point, perhaps a few months later, you will be considered an authority on topic1 and topic2, your traffic will double from the original and it will be worth it in the end.
Pros: it makes sense. It is the natural choice. It's what you would do if Google didn't exist.
Cons: loss of current traffic may be intolerable. Risk in never regaining sites original authority for original topic.

3. Subdomain. If you interlink from first subdomain to second as it were a subdirectory then Google will treat it same as subdirectory. See above. But, if you keep it as separate subdomain with minimal internal links from one to the other then this may be best option.
Pros: it's natural. Minimize risk. Same code base.
Cons: requires minimal interlinking which will make traffic growth much more difficult.

These are my thoughts. I've some experience with all 3 options. It's a tough decision. I guess I'd go with new subdomain.
Good luck.

[bluemi]

Many thanks everyone for your thoughtful replies. It's not really getting a lot easier... :D But at least I have a few ideas now.

[engine]

I always think of subdomains in the same way that i'd use a different domain name.
From a user point of view, the main advantage you have of using a subdomain is the branding.

Yes, i'll echo the, leave well alone message.

[EditorialGuy]

I think the takeaway can still be that adding the new content on a subdirectory will dilute the sites authority for your original topic and hurt your rankings for the original topic.

I think it depends. In my experience, Google has no trouble recognizing our site's authority for major subbtopics. Using fictitious examples to illustrate my point, the structure of our site is::

MAIN TOPIC: Poultry

SUBTOPICS: Chickens, ducks, geese, turkeys

"Chickens" and "ducks" are our most popular [fictitious] subtopics, and pretty much anything we publish in our "chickens" and "ducks" subdirectories will rank well in Google Search.

When we started this site many years ago, I put everything under our "umbrella" domain mainly because it was the simplest and cheapest approach and I hadn't bothered to look at alternatives. Later, I was tempted to split the site into dedicated sites (one for "chickens," one for "ducks," etc.), but I'm now glad that I didn't, because Google has no trouble recognizing our authority for those subtopics and we probably benefit from the clout of incoming links to the site as a whole.

Note: These observations are about a site with an overall topic ("poultry") whose subdirectories are about subtopics of that main topic. If we added subdirectories for unrelated topics like "goats" or "soybeans," we'd look pretty stupid unless we changed the umbrella topic from "poultry" to "farming."

[frankleeceo]

Same domain to start, since the services are really equal to one another. You are selling travel packages. As long as the current domain isn't area specific like...(asiatravel.com), and now you are offering traveling to europe with that domain. That most likely won't work.

However, if the goal is to be completely safe and the packages are truly different. I would "protect" the original brand and destination, and straight out create a new domain. Since really, if the packages are total different, they might serve to different groups of users and demographics. For example, if your current original destination caters to 20~30 singles or couples, and your new package caters to family, that might not work with the same domain.

If put on same domain, you most likely will not dilute the main destination because of its history in the short term, but will have trouble getting recategorized for the second destination, need time. However, I would caution as the second destination is gaining traction it does dilute the original destination (whole site) especially and only if the second destination does not gain merit or points on their own. But it's a pill that you may have to take to grow for the long term if you want to stick with same site.

If you are only testing the water to see how your users may "bite" the new offering, I would noindex the new pages for the 2nd destination originally and see how that goes. You will be serving that package to your existing customers with lesser possible search engine implications. (You still dilute to a degree because of the newly added keywords on your main pages leading to the noindexed 2nd offering) But if there is strong synergy or connection with users responding to the new stuff, then it could be time to let search engines index the new destinations pages.

If you realize through testing that none of your existing customer is actually baited to the new offering (not even checking them out), then that might definitely merit a separate domain under the different brand. Since you are really serving different groups of customers. You are looking to create a stronger brand and synergy between your original and new destinations, and if the new destination does not help with that, then I see no short to mid term point on the same domain.

I think the key is brand and user satisfaction. And how that new package can further the travel planning experiences of your current existing user pool. That would be where I make my final judgement and decision.