multi sites as SEO strategy?

I have recently come accross a very strange strategy for search engine optimisation and I would love to hear your views on this.
The company intends to launch a new site (in a very competitive arena) in the first quarter of next year. As they have no established brand as yet, they decided to buy about 500 domain names, which all have interesting keywords in them. (based on topic related keyword searches on the overture and espotting networks). Only one domain will be used as the actual brand. The idea behind the other 499 domains is to develop them as information site which relate to the activity being conducted on the "main" site. The intention is to cross link all those sites between them and of course always link them to the main site, thereby creating a strong link network which should raise visibility on the search engines.
Over and above the fact that developing 499 original content sites is probably a daunting and expensive task in itself, can this strategy actually work? Will the search engines not pick up on this and consider this as inappropriate and possibly ban those sites? Are there other examples of similar strategies being used successfully? Even if it does work, is this not a ludicrously expensive and therefore inefficient way of getting search engine visibility?

Any views on this much appreciated. Please note that I am not an SEO expert, so apology if this question is totally naive.

Thanks in advance

Stephan Samson

The technique you mention is frowned upon by the major search engines.

There are many sites that have been banned from Google for doing exactly this. There are also many organisations who are 'getting away' with similar practices, and are appearing high in the search engine results pages.

However, any organisation employing such methods is walking a tight-rope, and should be aware that at anytime they may fall foul of the search engines and be completely dropped from the listings.

We would advise any clients to optimise a single site, and any necessary 'microsites', rather than trying to create an artificial number of 'different' inter-linking sites. Search engine optimisation has to be considered as a long-term strategy within the marketing and promotional mix. Doing something that may be successful in the short-term but can jeopardise the site's presence in the search engine listings over the long-term is rather short-sighted.

Search engines couldn't care less about the website owner. They are far more concerned about producing relevant results for the searcher (of course they would like website owners to pay for PPC ads, and here their priority is balanced between the searcher and the advertiser). A results page containing lots of links to the same organistaion who is flooding the results with essentially the same content is not in the best interest of the searcher - and therefore unlikely to be looked upon favourably by the search engine.

Of course, there will be other people who will disagree and positively encourage such practices. If you decide to undertake this approach, you should definitely be aware of the risks.

Hope that helps,

Daniel

>> Clue or warning?

A bit of both. The "big money" categories (adult, travel, gambling and historically pharmaceuticals) are where you will see the cutting edge of SEO techniques, and where the axe falls first when the engines work out an effective counter

The big players, especially in the travel and adult sectors are very fond of the "domain farm" approach, as described above. I remember one memorable breakfast with a certain notorious "Black hat" who was describing his monthly Google spamming cycle to me

"I buy about 300 domains a month, generate a few titles, <META> keywords and description tags, <H1> tags and half a dozen keyword and link stuffed paragraphs of text. I plug those into a mix'n'match script, which churns up about 30k pages overnight. Divide those up into 100 page blocks, upload them to the server, and wait for G'bot to come calling. If the linking structure is right, and the feeder domain is still clean, it takes about 3 days to grab them all. Come the next update they are all in, and pulling traffic"

You can buy domains for $5 a time, if you look hard enough. The fact that 90% or more of those domains will crash and burn is irrelevant, as the remaining 10% can more than pay for the lot

The reason that this works is related to how the PR (PageRank) system works. PR is NOT a zero-sum game, ie as more pages are added to G's index, the "total" PR in the world increases. A site that has more pages than a rival, all other things being equal, will have a higher PR value, and therefore outrank it. The domain farms are the logical exploit for this facet of the G algo, although there are some obvious counters (ID'ing and dumping the whole network once for instance), but there are other less drastic options (removing the ability of key PR distribution pages to pass on their PR for instance)

To relate this to your actual query, there is nothing inherently wrong in creating hundreds of stand-alone sites, and cross-linking them, provided that each site stands alone as a valuable resource in it's own right, and the cross-linking is not a many-to-many, "link every site to every site" pattern, but a considered, "natural looking" linking pattern that will lead users to other information that may interest them within (or even outside) your network.

Google is interested in providing the best possible service to their user base. If you help thm out by providing good, unique content, you will prosper in the long term