Skip to main content



Testing Google’s Panda algorithm: CNET analysis

http://news.cnet.com/8301-31921_3-20054797-281.html

Google continuously makes changes to their search and pagerank algorithms.  Most of the time these changes go under the radar and are incremental, but on occasion the update has major effects in the internet search industry.

The above article (linked) discusses one of Google’s major and recent changes to their searching algorithm, referred to as “Panda”.  CNET ran a test on the new algorithm to investigate the effects these changes had.  The goal of Panda was to address a recent increase in sites referred to as content farms, which generate abundant content based on what people are most commonly searching.  Content farms typically have poor content and can occupy the top searches in many categories, thus making it difficult for internet users to actually find quality pages.  Websites such as eHow that copy or automatically generate content addressing questions people pose to search engines are a major part of content farms.

The results of CNET’s test show that the top sites such as Amazon, Wikipedia, and YouTube maintain the top positions.  Also, some sites exhibited a significant boost in rankings such as government websites, which generally have high quality but specific content.  Websites like WikiHow and eHow slipped considerably in ranking.  It is interesting to note that the websites which were most concerned with search engine optimization (SEO) slipped the most, since they are most focused on tailoring content to what people are searching rather than focusing on improving the overall quality of site content. The Google algorithm pays attention to what users do while at a site to better develop an understanding of their “evaluation” of it a being useful or not.

SEO attempts ton increase a site’s visibility by making it appear more often in search results, which will generate more site visits.  This is achieved in many ways.  One method is to increase the number of cross links between pages.  Another method is to contain content that peope often search for.  The latter is often the method of choice for content farms, which most obviously take the form of sites like eHow, WikiAnswers, and EZineArticles.  Many people search for answers to their questions, and these websites often have poor, limited responses to people’s answers.  Just the mere fact that their content exists directly addressing the searcher’s question places the content farm as a top ranked site when searched, but provides low payoff to the internet user.  The content farm is primarily focused on generating site visits to increase advertisement revenue and visibility on the net.

Panda works to reverse the effects of SEO on content farms.  Overall, Panda improves Google’s search results to users giving much more usable and worthwhile results while sifting out the websites that are more focused on “beating the game” and exploiting existing, well understood searching and ranking algorithms.

Comments

2 Responses to “ Testing Google’s Panda algorithm: CNET analysis ”

Leave a Reply

Blogging Calendar

October 2011
M T W T F S S
 12
3456789
10111213141516
17181920212223
24252627282930
31  

Archives