PageRank and the Downfall of Quality Content
We’ve all heard of the infamous PageRank algorithm that Google employs to rank the webpages returned by it’s search results. Of course, Google intended to make it foolproof, fair and useful to the user, but did it in fact turn out to be so?
PageRank works on the principle of Link Analysis, mainly taking into consideration the following two factors:
(a) The number of links that point to a particular webpage – if a page has more links pointing to it than any other page on the same topic, it will rank the highest. This is natural as it means the page is more reliable.
(b) Relevancy/value of their content – besides having a high number of links pointing to it, it is also essential to consider the quality or value of the content of this page.
It was found that factor (b) weighed a lot more in the algorithm than factor (a), meaning that if a page had 10 high quality links, it would rank much higher than a page with a 100 low quality links – this was popularly referred to as rewarding “keyword-rich content”.
However, when PageRank first came out, the web was nowhere like it is today – content in general was limited. Once companies and businesses figured out how PageRank worked, they also figured out how to manipulate it to get themselves more hits with higher ranks. To have PageRank work for you, all you needed to do was satisfy factor (b). While factor (b) at first appears to be something desirable and a good filter, we can see that it is in fact easy to find a workaround to it. The algorithm simply relies on “keyword-rich content”, so if lazy marketers simply published bad keyword-rich content that made absolutely no sense, they still succeeded in ranking highest. In fact, for maximum manipulation, some even created garbage content on external blogs/content farms and linked them back to their websites which helped with factors (a) and (b).
Of course, this resulted in rankings for various webpages soaring which resulted in others following in their footsteps, focusing on ridiculous “keyword rich content”. This had an adverse effect on writers of online content at the time. Since the value of the content was no longer the focus of online writing, it became a layman’s task which meant that good-quality writers were no longer in demand. Their wages went down immensely and they found themselves frequently unemployed unless they were looking for meaningless work that just involved churning out posts that were attractive to PageRank’s algorithm. Another problem was that finding quality content on the Internet became near impossible even though they were plenty of websites and blogs with legitimate, valuable keyword-rich content.
In February 2011, Google finally addressed the massive problem with the Panda update it rolled out. Panda worked using a Machine Learning principle – they fed it a relatively small training data set of sites Google’s quality raters liked and sites they didn’t, which sites they would trust with their personal information, which sites they thought were credible, had quality content, etc. They then scaled the results of the training data set to a larger data set using a machine learning process.
And voila! All the malicious, manipulative, horrible web pages were no longer allowed meaningless high ranks. The best part? Panda ranked entire websites and not just web pages, which meant that even if a section of their website was deemed unacceptable by Panda, the credibility of that entire website was lost and it no longer featured in the high ranks.
Of course, there are still various ways to manipulate Panda and rank high on search results. But with all the great Panda updates Google has been releasing over the years, it is becoming more and more likely that the only ways to manipulate it are by producing actual high quality content.
http://www.copyblogger.com/page-rank-vs-author-rank/
https://moz.com/blog/how-googles-panda-update-changed-seo-best-practices-forever-whiteboard-friday