Multiple Sites - Same Code | Am I doing the right thing?

[kleward]

I have several sites (separate site for hotels in each city) which point to a single code/database, I've on my server. But each site have a search functionality embed with it which on search for other city shows the hotels of any other city on some other city domain.
and Google is caching & indexing those pages too.

Let me be clear my question with an example:
my sites are:
city1hotels.com
city2hotels.com
city3hotels.com
and so on.

now suppose there is a HotelABC in city1 so the hotel specific page in this site is: city1hotels.com/../hotelABC/105 which is in Google's index

however other sites like city2hotels.com & city4hotels.com don't have a direct link to this hotel But if someone searches through the search box given on the sites for some specific city all the hotels in that city/locality are shown. (data is fetched from a single database)

Now suppose 'm on city2hotels.com and searched for hotels in city1 so the city1 hotels' listing is shown on the city2hotels.com (Which I strongly feel is right for User's convenience) Now if i click on HotelABC on this search the same description page opens on city2hotels.com/../hotelABC/289 Google indexes this page too.

NOW the problem is except the URL everything else (hotel details, title, description etc) is same and can be considered duplicate, so I tried to Disallow these search results on my server's Robots.txt file (so that URL's being created by search should not get indexed)

But even after that Google is caching those links i.e city1hotels.com/../hotelABC/ & city2hotels.com/../hotelABC/ and for some queries both the pages come into google's search results.


Question 1: Is this the right thing to do?

Question 2: despite of disallowing through robots.txt file why Google keeps indexing those pages?


I'm afraid of being marked for duplicate content? BUT strangely after the complete roll out of Google Panda, my sites have started gaining good ranks in Google SERPs.

Does PANDA behaves differently for industry specific pages.. such as Travel !

I hope I'm clear with my questions. if not please let me know.

[tedster]

Welcome to the forums, kleward.

If you are using the proper robots.txt syntax, then you've done what Google asks you to do and are saying "keep my search results out of the index." Have you checked the syntax of your robots.txt with the tool in WebmasterTools?

[Planet13]

Kleward:

I know this isn't what you asked, but off the top of my head, this seems like it would be a very hard way to compete for rankings.

It sounds very similar to many real estate web sites; they all show listings for homes for sale that are in the Multiple Listing Service (MLS) database, which is available to all members of the National Association of Realtors. So in essence, all those real estate sites show nearly identical content.

I don't know if you would be showing the exact same content as a lot of your competitors. If so, it is going to be tough to compete in the SERPs. Especially because you seem to have LOTS of different domains, so you will need to build quality links to ALL those different domains.

anyway, maybe I am misunderstanding what you are doing, and if so, I apologize. But it seems to me a lot of webmasters would benefit by looking at their sites the way that any CEO would look at his or her own company.

They need to ask "what makes my product / service unique? What are my core competencies? What can I offer that no one else can?"

anyway, I hope this helps.

[Shaddows]

I tried to Disallow these search results on my server's Robots.txt file (so that URL's being created by search should not get indexed)

Just to clarify- you are blocking the search results but not duplicated URLs?

So if G "knew" the final URL, it could legitimately fetch the page? As URL paths beyond the domain are "the same", I assume this is the case?

What you might want to consider is using a script to compare if a returned hotel is hosted on the searched site, and insert "noindex" if not.

[johnhh]

I have had this problem

If "city" in database field = site name then "index", if not then "noindex nofollow"

However that will only work for a page featuring one hotel. If you have a search results listing short details per hotel the hotel from the other city may still appear in the lists.

[kleward]

Hi all !
Thanks, tedster, Planet13, Shaddows & johnhh for your kind replies. :)


tedster,
Yes I talked to developers and as per them the path that generate search results have been disallowed in the robots.txt, I checked it in the Google webmaster tools and it seems to be alright.

Planet13 The hotels' details/descriptions we have acquired form the hotel owners/representatives, so yeah that might not be unique (they may have provided the same descriptions to other sites too).

Yes, my team is working on the links part and for each individual hotels they are doing social bookmarking thing too (I hope this is the right way?)

Shaddows & johnhh

Yes, I've asked the development team to work on this possibility to put noindex meta for the other city hotel on a different city domain.
However I didn' find except the search whereelse the G can find the URLs of other city hotels.

I hope putting meta noindex will work?

But I didn't understand Why after this update my ranks for good keywords have improved? and search engine traffic has grown.

[Shaddows]

However I didn' find except the search whereelse the G can find the URLs of other city hotels.

I assume you meant, "where else can Google discover the URLs apart from the search".

Well, there are lots, including
- Toolbar
- Inbound links
- Google "testing" (they try nonsense URLs occasionally- I imagine they could test cookie-cutter sites like that too)
- Already crawled, never forgotten. Although as an orphaned page, it might cause you problems in future (there is/was a theory I haven't heard for AGES that orphaned pages are bad for ranking)

Don't put "noindex,nofollow", just "noindex".
***

Edit- misread text, removed irrelevant information

[kleward]

Thanks Shaddows,

I was busy working with the development team to fix all these issues, as you suggested, I'm implementing meta robots noindex for the other city hotel pages.

Also worked with the article section on these sites which used to go on all the websites. Now travelogue submitted from one website will be displayed on that website only and not on others.

This will increase the Unique content pages on the websites. I hope this will help too.

Thanks for all the help, I will write back again in few days, once all the changes are live.