Google Panda Update

Panda update is a new ranking factor introduced by Google into its algorithm in early 2011, and tweaked continuously since then. The aim of this update is to surface better quality websites higher in Google’s search results. In this text we’ll provide some background and SEO context to this fundamental change in Google search.

From Caffeine to Panda

In August 2009 Google, the most popular search engine in the world, announced a massive update in its infrastructure – the Caffeine update.

With this update Google actually changed the way it scans and indexes the web. Up until this update, mapping was carried out in a linear chain of several steps. These steps would take different amounts of time to complete (one of the steps was performed once every two weeks) and were executed on large groups of pages which were fed to the search engine by its scanning system. Pages were only introduced to the database after the last stage.

The Caffeine update did not affect the rankings themselves nor did it affect Google’s search speed. It did however effectively replace the old scanning system and shortened the network mapping process dramatically. From that point on pages no longer needed to complete all phases of mapping in order to be entered to the database, so the indexing process accrues almost in real time (up to seconds after the actual scan).

This change significantly increased the information held by Google data centers at any given time (index size grew by 50%) thus increasing the amount of both current and relevant data. The Caffeine update laid the foundations for larger and more frequent updates in Google’s algorithm, allowing the implementation of changes in the same line.

By December 2009 it was possible to stumble upon “tweets” from twitter, news articles and other topical content, in the Google search results, minutes after it was published.

“May day” update

In May 2010 another large update known as “May Day” came into effect, a wordplay that derived from the date of the update’s release and emphasized the panic it caused many websites owners. The update primarily affected sites that were ranked on long phrases (Long tail) – strong sites were hardly affected.

The purpose of the update was to degrade lower-quality websites, placing more weight on quality signals and lowering weight of pure textual relevancy signals. This development can clearly point out the new approach Google has adopted that led later to Google’s famous “Panda update” (more details on this later on).

The May Day update was a direct result of the Caffeine update, due to the fact that Caffeine significantly enhanced Google’s index, thus flooding the search results with poor quality sites which managed to slip into feasible ranking positions for medium and long phrases (Mid Tail, Long Tail). Spam has always been Google’s number one enemy but at this stage the problem became the top priority and received the full attention of Google’s engineers.
June 2010 – Google finishes the embedding of the Caffeine in all its servers worldwide.

As of November 2010 Google allows users to preview websites right from the search results pages (SERPs). This addition came in the shape of a little magnifying glass near each result. This is another display of Google’s anti-spam policy which basically provided the user with the ability to choose the relevant result based on design and usability – 2 none measurable criteria from Google’s point of view.

Over time we can see that Google emphasizes user behavior, giving it a higher weight in its calculations.

Social signals introduced to the algorithms

In December 2010 Google officially confirmed the usage of social signals in determining site ranks. The main sites that were mentioned in this context were Facebook and Twitter, the world’s leading social networks. The logic behind this direction was that if the site truly contains quality content, it will be mentioned, talked about, and shared, regardless of the keywords the site has in its article headers.

The panda is coming

In the beginning of 2011 some clear signs of an uncompromising war on spam stood out as Google performed selective punishment of spam sites and sites which made use of promoting techniques that are being considered “unethical” according to Google’s terms of service (also known as “black hat SEO).

In late February 2011 the online world has trembled when Google released its “farmer update” (later renamed to Panda). The main purpose of this update was to hurt “Content Farms”, or sites that contain huge amounts of content of poor quality in order to rank on as high number of keyword combinations as possible, hence the name Farmer Update. The update was named later after the senior engineer who was involved in its creation and thus received the unofficial, but more renowned name “the Panda update”. According to Google’s official figures, the update has affected 12% of all online searches and countless sites were damaged or even completely disappeared from the search results.

Panda has revolutionized the concept of search engine optimization by placing the emphasis on user experience, and for the first time web site owners and developers had to think about how to make the site better for users and not just for search engines. This was a completely new perspective as far as SEO was concerned, and Google began using different parameters like average time spent on the site/specific page, bounce rate, CTR (Click Through Rate) and more, to measure and quantify user experience.

At the same time Google continued to nurture two rising trends:

• The social trend – “+1” buttons was added near each result. In addition, this move had set the ground for the “Google+” social network, which would be released later on that year.
• Personalized Search – According to this principle there is no one right answer for everyone and each user should get different results for the same search, based on what would be the most relevant results for that specific user. The changes in results between users could arise from geographic differences (different countries, cities and even different regions in the same city), daytime changes (user would get different results in the night then what was presented to him in the morning). If the user is logged in to his Google account the results would be adjusted even further since Google’s servers collect information about the user and his browsing habits.

The above can clearly show the enormous amount of resources Google invests towards improving its user’s experience.

Since “Panda” first raised its head in early 2011 there have been a total of 6 publicized updated to the Panda algorithm (and probably more smaller ones that went below the radar):

• April 2011 – Panda update receives some additional features and released as Panda 2.0 in some countries.
• May 2011 – Panda 2.1 (consisted of some minor changes that didn’t cause much movement in rankings).
• June 2011- Google releases Panda 2.2 as well as its anticipated social network, “Google+”.
• July 2011 – Panda 2.3.
• August 2011 – Panda 2.4. For the first time the update affects non-English quarries (except for Chinese, Japanese and Korean). Also an upgrade was done to the “Site links” in the SERPs, empowering brands and large sites.
• September 2011 – panda 2.5. The nature of changes is still unclear.