Friday, July 20, 2007

Content in Adsense

It should come as no surprise at this point that, content is of the utmost importance as it relates to AdSense monetization. Constant attention needs to paid towards your content, regardless of the kind of website your are running. Content can and will affect the following parameters:

• Ranking on Search Engines.

• Time it takes to get indexed and ranked by search engine bots.

• Overall visibility over long periods of time.

Understanding how the search engines rank your content is crucial to orchestrating your attack on them. While proprietary information not made public, there are strong indications that Google, and possibly other search engines, judge web content using a method based on accessing a sentence sequence consisting of 12 to 15 words. This sequence is referred to as a "shingle." Now, if this shingle appears anywhere else in the search engine's index, it considers it "duplicate content." When that happens, based on a combination of other parameters, a snap assessment will be made to decide which of the duplicate pages should be ranked higher.

It is imperative that the search engine results not display 10,000 duplicate pages across 10,000 different sites. If it did, the search engine wouldn't be that popular. To help clarify this, go to PRWEB.com and pick out a press release at random. Then, still at random, highlight a sentence of around 15 words and run a Google search with double quotes "USING THE 15 WORD SENTENCE LIKE THIS."

Very likely, PRWEB, having established itself as the authority when it comes to press releases, will be #1 in the results. Now, considering that there are thousands of spam sites scraping PRWEB for their content every minute of every day, it's a pretty impressive feat.

So, it's obvious that web content, and its relevance, without a doubt, are extremely important to our search engine objectives. But I don't believe it to be the most important.

           Astute readers may be wondering how Google, which is a completely automated system using artificial intelligence, could possibly have known that PRWEB is the press release authority and, in a matter of micro seconds, rank them accordingly?

The question is so brilliant that the answer might be a let down. But it is the very reason why I do not consider web content, no matter its relevance, to be the most important factor in search engine placement. One element is even greater: Back links.

A "back link" is a fairly simple concept to grasp. It's a website that your site is links to, that when the spider follows it, finds a link back to your site from. So if your website is linked by really good websites, yours will rank better.

Even the age of a domain plays a role, which is why a seasoned, well-aged domain, while appearing useless to some, is so highly sought after by others. Google even holds a domain registrar status so they can access any information about any domain at any time. Even information not made available to the public.

So, while it might seem questionable as to whether Google can use such data to influence their search result, the best guess is that of course they do. So a domain with proper registration information, which is a few years old, would probably rank better than a newer domain, based on age to determine relevance over all the other parameters involved.

So, how can we sum this all up?

1. The more back links you have coming into your site, the more
relevant it will appear to the search engines.

2. Furthermore, the more back links to your site from similarly
themed websites,
the more relevant your website is to the search
engines compared to point 1.

3. Finally, the more back links from similarly themed websites with
tremendous notoriety, the more relevant your site will be to the
search engines compared to point 2.

In the BlackHat world, we must generate a lot of readable, yet constantly unique content every day. It is not an easy task, but there are a lot of weapons we can use to assist us in our mission.

Markov Chain

Markov Chain algorithm can be utilized to superficially generate very "real-looking" content, given just one seed document. This means that if you feed it a text-string, it will instantly be returned as a completely rewritten article, yet will still be reasonably readable. This is of supreme value to us because that's just what we want: unique content! By throwing a bunch of text into the Markov Chain algorithm, we will get just that. Usually the Markov Chain algorithm is used to generate hundreds of articles on the fly using only one established article as the seed.

Peter Bray's Markov Engine is one of its kind content generator that uses this algorithm and brought content generation beyond what you can imagine. It is very user friendly and you just need to give it some seed document and templates, and it will immediately generate the pages out containing unique and readable contents in matter of seconds!

No comments: