All websites on 1 server or across multiple?

[fulleffect]

Hi everyone,

If my client owned a car cleaning business and had a primary website of cleanmycar.com and he wanted to create other websites (instead of creating sub domains or normal folders) such as

carcleaning.com
car-cleaning-england.com
carvaleting.com

.....instead of ......

cleanmycar.com/carcleaning
cleanmycar.com/car-cleaning-england
cleanmycar.com/carvaleting


Would those additional websites be better scattered across multiple servers, as aposed to being on the 1 server?

Reason i ask this, is because i want to know if Google realises, all of these related car sites that point to each other, are hosted on the same server, so therefore it does not rank them as much as if they were on multiple servers?


Thanks
Daz

[kgun]

Originally Posted by fulleffect

Reason i ask this, is because i want to know if Google realises, all of these related car sites that point to each other, are hosted on the same server, so therefore it does not rank them as much as if they were on multiple servers?

Depeinging on where your market is, the concenus view as I interpret it seem(ed) to be that for SEO purposes it is best to

1. Have a .com verision in USA.
2. In your case have a co.uk verision in UK
3. Other regions if that is an important part of your market.

After this

Aaron Wall gets special treatment from Google?

thread I am far from sure any more. Aaron Wall uses subdomains on the same server.

In my opinion.
Content is more important than digital structure.

[fulleffect]

Thanks for your comments Kgun, however i think you have misinterprated my question.

I realise content is king, thats why my client wants multiple websites with different content on each.

My question is, "Is Google clever enough to realise the 10 websites all about car cleaning on the 1 server are related and gives negative points for this, as aposed to them being on multiple servers" ?

The more websites that point to 1 primary website, the more positive points Google gives that website, however, if all of those websites that link to the primary are on the same server, surely Google will realise this and not give as many points, or give no points at all?



Thanks
Daz

[spiderbait]

I believe Google discounts IBLs from domains residing not just on the same server, but in the same IP range.

To avoid that, I think you'd need your websites to use different IP C Blocks.

However, if you're planning to create duplicate or near-duplicate content it isn't going to make much difference whether you host them on the same server or on the moon. IBLs from duplicated content are likely to be greatly discounted no matter what.

[kgun]

Originally Posted by fulleffect

My question is, "Is Google clever enough to realise the 10 websites all about car cleaning on the 1 server are related ...

That should not be difficult to figure out for Google.

Originally Posted by fulleffect

and gives negative points for this, as aposed to them being on multiple servers" ?

Ordinarily yes according to what I regard as the concensus view, but study the link I gave you above. I am not sure any more.

Originally Posted by fulleffect

The more websites that point to 1 primary website, the more positive points Google gives that website, ...

Yes, especially stable external semantic links.

Originally Posted by fulleffect

..., if all of those websites that link to the primary are on the same server, surely Google will realise this and not give as many points, or give no points at all?

Linkfarms and similar, have you heard about that? Unique content on all sites is the primary foccus IMO.

Personally I distribute my more or less related content (47 sites) accross 4 IP's and three hosters. The primary reason is security, not depending on a single hoster and semantic linking (there are some overlap because of different foccus).

[cw1865]

Originally Posted by fulleffect

Hi everyone,

If my client owned a car cleaning business and had a primary website of cleanmycar.com and he wanted to create other websites (instead of creating sub domains or normal folders) such as

carcleaning.com
car-cleaning-england.com
carvaleting.com

.....instead of ......

cleanmycar.com/carcleaning
cleanmycar.com/car-cleaning-england
cleanmycar.com/carvaleting


Would those additional websites be better scattered across multiple servers, as aposed to being on the 1 server?

Reason i ask this, is because i want to know if Google realises, all of these related car sites that point to each other, are hosted on the same server, so therefore it does not rank them as much as if they were on multiple servers?


Thanks
Daz

You're better off across a couple of different hosts.

[jestep]

At this point Google is looking at registration information as well as IP C blocks, and the network that your server is being hosted on. Apart from that, if you create a big link network, it will be blatantly obvious that all the sites are owned by the same entity.

You may be OK with a major TLD for a few specific regions. Make sure you host those sites on a server that is on one of those areas as well (IE: .com -> US, .uk -> somewhere in the UK, etc..) If you do this method, would still keep cross-linking to a minimum and make sure the sites are all high quality and as unique as possible.

[edhan]

Personally, I do believe that it can be in the same server but the ranking usually goes to one main site. If you do it for multiple servers (different ip addess) with unique content on them, then I believe that the ranking will be evenly distributed (depending on IBL).

[SEO]

Originally Posted by fulleffect

Reason i ask this, is because i want to know if Google realises, all of these related car sites that point to each other, are hosted on the same server, so therefore it does not rank them as much as if they were on multiple servers?

You will more than likely get your clients site banned if you interlink all of the sites.
Webmaster Guidelines

For maximum SEO benefit, you will need to have them all hosted on different servers and all pointing to the main domain.
Interlinking them all is Very risky.

Some people recommend setting up a pyramid type linking system.

Rule of thumb:
Avoid tricks intended to improve search engine rankings.
Don't participate in link schemes designed to increase your site's ranking or PageRank.

[incrediblehelp]

Originally Posted by fulleffect

I realize content is king, thats why my client wants multiple websites with different content on each.

Why? To game Google? Don’t do it.

Originally Posted by fulleffect

My question is, "Is Google clever enough to realise the 10 websites all about car cleaning on the 1 server are related and gives negative points for this, as aposed to them being on multiple servers" ?

Yes of course. Sometimes they catch them, sometimes they don’t. Why not focus all your content on one main websites for all your end users to find and send back links to because it is such a huge resource for them. Start thinking in terms of your end visitor and not how to game Google to rank better.

Originally Posted by fulleffect

The more websites that point to 1 primary website, the more positive points Google gives that website, however, if all of those websites that link to the primary are on the same server, surely Google will realise this and not give as many points, or give no points at all?

Well that is not the right way to think of it. You don’t want to get in the habit of building a mini-network that points all to your main website. You want to create a website that attract links naturally.

[subsystems]

I would think since all the proposed domains are related to "car cleaning" that maybe they could be consolidated into 1 main site. Take the advantage of having more content on 1 site. Which is better, 3 5 page sites or 1 15 page site?
The exception would be the .co.uk and .com domains. Maybe taylor the sites to each market. Colors and style appeal as well as wording are not the same between US an UK.

My thinking is fewer larger consolidated sites gives a better chance of a more comprehensive site, a better quality site and may attract more return visitors and IBLs. They always say build for the visitors not the search engines. Of course SEO is important.

[amitgupta007_99]

In my view,

if u have independent content for each of them, why dont u just make a big big portal out of it.

So much quality content and subdirectories can give u more traffic than so many small websites.

Think about it mate..

Regards

Amit

[fulleffect]

Thanks for everyone's reply.

The logic behind my clients method, is down to IBL, he thinks if he creates multiple websites on multiple servers all with different content, and Google does not see all his sites as his, then Google will give his site more points because of all of the IBL's.


I think the risk here is too great for the reward that he may get out of it, so on that basis i will advise against doing this.


Thanks again everyone!
Daz

[dbramley]

what if you set up a dedicated ip for each site (even though they are on the same server) - would this make a difference?

[freecall]

Hi

I am always questioning this theory also as I have a property portal listing 1,000's of properties worldwide - then we have individual websites for each country all on .com's taking the property data from the main database.

Every individual website has its own content but some are listed on the same server but none are listed on the same server as the main portal although the majority of the domains are hosted with the same hosts albeit on different c block ip addresses

Am I heading in the right direction or is the consensus of opinion that I should point all these idividual sites into the main portal?

[iandoc]

Originally Posted by spiderbait

I believe Google discounts IBLs from domains residing not just on the same server, but in the same IP range.

To avoid that, I think you'd need your websites to use different IP C Blocks.

However, if you're planning to create duplicate or near-duplicate content it isn't going to make much difference whether you host them on the same server or on the moon. IBLs from duplicated content are likely to be greatly discounted no matter what.

I think Spiderbait set it out quite succinctly.

[pietpetoors]

If it might be a problem hosting all your sites on one server because they might know they all belong to you, is it then wise to submit all your sites to Google's Webmaster tools? If I read above it seems to me that submitting all your sites to Google Webmaster tools will do more harm than good.

[BradHart]

I think the situation is getting over thought. Different servers or not having all these domains pointing only to each other sends up a big red flag. You might score some quick points in the rankings if you aren't in too thick of a keyword niche, then again it might just take google a couple of months of looking at it and deciding to do nothing or lower the ranking of every site. You might beat the system in the short run, but unless you are going to spread your network a lot wider than a handful of domains across a couple of different hosts then you will never get very far in this game.

[Terry Van Horne]

Multiple domains is risky, though less risky then it was a few years back. Cross linking the sites whether on the same server or different IP blocks is even riskier. Remember, different servers and C blocks mean nothing if the Whois shows all sites are owned by the same person/company. Google has used the whois data for years. In fact locking your domain records likely raises a flag, or seemed to when I tried it.

Kgun, IMO, the subdomain issue you mentioned is not something that has fully played out unless Google doesn't care that one site can and are dominating the first page of a SERP. I feel this algo has another shoe to drop yet. It definitely would seem like the SD are still being treated as standalone domains. No way IMO, does Google want that. It also rewards what used to be regarded as a risky SEO technique.

[kgun]

Originally Posted by Terry Van Horne

Multiple domains is risky, though less risky then it was a few years back. Cross linking the sites whether on the same server or different IP blocks is even riskier.

That may be a very strong argument for human edited directories. That does not mean that we don't need Bot driven search engines.

Originally Posted by Terry Van Horne