Starting in February 2011, Google began releasing what was to become a series of game-changing updates designed to weed out websites with low-quality content in high quantities, spam-techniques, black-hat SEOs and any webpage that was conning the algorithm into getting a search ranking that it did not deserve. A majority of the websites, hit severely in terms of their SEO visibility, were content-farm websites that were badly dependent on search engines for traffic and revenue, which is perhaps the reason that another name for the Panda update was ‘Farmer’.
The rise of Google Panda
Google Panda update made headlines not just in the world of SEO but internet publishing as a whole, going to the extent of forcing million dollar websites out of business overnight or to the brink of bankruptcy, and literally being forced to rethink their practices and business models.
In their own words, Google wanted to restore the rightful places of high quality websites and get rid of what they call ‘thin’ websites that were taking advantage of clever SEO techniques and loopholes in the algorithms; that contained sub-standard content, were full of intrusive and irrelevant advertising, had ugly user-interfaces and did not bring any real value to the internet. Some excellent and telling examples of the kind of websites that Google was looking to target when it rolled out the first Panda update are www.ezinearticles.com and www.hubpages.com amongst a dozen or so other examples.
The former are, or in some cases were, websites dedicated to farming links through hosting shoddy marketing articles written for purely SEO reasons rather than any content value or the benefit of users and are notorious for passing off marketing spam as informational pieces. The CEO of HubPages who took over after the Panda onslaught, admitted in surprising candor the SEO based business model they had been following pre-Panda and that the administration had clearly encouraged the submitters of content to target the search engines rather than produce content for users or any informational value. Lastly, these websites also contained a vast and overwhelming amount of saturated advertising that covered a massive portion of the user-interface.
Although Google had been making changes to its algorithm on a very regular basis, with the frequency rising to 500-600 changes per year, most if not all of these changes were minor in comparison to what began in February 2011, and they did not have a significant impact on the traffic and revenue of websites that were deeply dependent on Google for their survival. Compare this with the top twenty websites in the losing boat post-Panda, that lost SEO visibility ranging from a minimum of 65% and many highs of up to 91% loss in search visibility.
Only two websites from a daunting list of the affected losers were able to make any sort of recovery, while the rest continued to slump in their SEO visibility, traffic and revenues, not even being able to return to the position they held after weeks of Panda rolling out, since their descent into SERP oblivion did not cease. After the first Panda update, Google continued fine-tuning the reforms in algorithms it had started in February 2011 by releasing more updates of Panda, with Panda 4.0 from September, 2016 being the last known update. It was also announced that Panda officially become a part of Google’s core ranking algorithm. Parallel to the Panda updates, Google floated some other changes in its ranking algorithms with Google Penguin and the EMD update.
Having looked at a brief history of the revolution in the SEO and internet publishing caused by the shockwaves of its series of updates, let us look in detail at what these changes in algorithms actually are, how they affect websites, and what they mean for you if you are an SEO professional, website owner or an online business.
Understanding the evolution of Google’s algorithms
We have looked at how Panda came to make an appearance on the internet and more specifically, the SEO scene and whom it affected the most. Let us now discuss in detail what Panda changed and how these changes in algorithms actually work.
There has been a lot of speculation as to the nature of changes and the factors, which led to such a large number of websites losing out in their SEO visibility in such a major way. The whole online publishing community was involved in presenting theories as to what the new algorithms constituted and what elements of a website’s design, structure and SEO strategies were to be used to win Google’s favor; or in some cases, win back Google’s favor and head towards recovery if you were part of the losing bucket after Panda.
Here is a list of factors identified in these new algorithmic changes.
Quality raters and machine-learning algorithms
For quite some time Google had been making use of a human element in its otherwise automated ranking system, and that was the use of Quality Raters. Quality Raters were staff hired for the purpose of reviewing and rating websites by answering a set of questions, which built up a site’s quality, authority and trustworthiness, and gave a subjective aspect to the objective and automated technical methodology of understanding websites.
The input of the Quality Raters and their patterns of answering were then fed into a machine-learning algorithm, which you can understand as a kind of artificial intelligence or software capable of evolving and learning, and thus becoming capable of imitating the human Quality Raters in their functioning. Google Panda changed this in a major way by making significantly more use of these machine-learning algorithms, thus indirectly giving Quality Raters a much bigger say in what websites would be considered high quality and which would be not.
Now obviously, when a Quality Rater is asked questions like: Is this a website you might trust with your credit card information? And the sites in question are obviously shoddy content-farms or ‘thin’ websites like HubPages, Ezinearticles or suite101; the answer would most definitely be an almost unanimous “no.” And since humans are more capable of distinguishing between high quality/trustworthiness and low-quality/marketing stunts, due to years of online experience and subjective reasoning, the strong partnership between Quality Raters and machine-learning algorithms, that came to significant prominence and importance as a result of the Panda updates, made it excruciatingly difficult for ‘thin’ websites and content-farms to fool the system. However, a slight downside to this new approach is precisely the subjective nature of opinion formed by Quality Raters, which could lead to a somewhat unexplainable criterion.
More refined content analysis
Another characteristic feature of the Google Panda update was its use of topic-modeling algorithms to conduct a much more in depth and human analysis of the content of a website with wider and more advanced parameters for rating. So whereas before, Google would only look at the relevance of a page’s content in comparison to the description found in tags, meta-elements etc., with Panda coming in to play, the topic-modeling algorithms came equipped with being able to distinguish between finer elements of content such as a page’s readability, the visual attraction of content presentation, how unique and solid the content was and other similar factors relating to content.
Duplicate and rehashed content
There are many types of situations that could be classified as duplication of content and just as many reasons for how it could happen, even without you doing it intentionally as an SEO trick. Let us look at some of these:
- True duplicates on the inside
This is a common form of duplicate content and it can happen without any clever designs on your part. An internal true duplicate is when you have two URLs leading to the exact same page. Google will think that you are trying to con the algorithm and quite possibly penalize you when you might just have been negligent in your URL structures. You should eliminate this content immediately.
- True duplicates on the outside
This is a problem usually caused by syndication of content across domains, which could be owned by you or somebody else. The danger lies in this seeming like SERP noise to the search engine and it might lead to website being penalized.
In case you own the other domains, you can use a simple canonical tag to redirect them to one URL of your choice as the source. However, if the domains and properties of the content belong to someone else as well, then you need to figure out how to solve this with the other webmasters.
Another even more annoying scenario could be mischief caused by scrapers who will scrape your content and republish it, resulting in a duplication of content, which then also becomes your problem sadly. Developing site-authority is the usual remedy to this situation, or in more extreme parasitic cases, you will need to bring out the big guns with a DMCA takedown notice.
- Quasi-duplicates on the inside
Now this might have been a clever bit of SEO before Panda began its reign, but it will only get a website into trouble now. Here is how it works. Suppose you have an original piece of content that sits well into a theme like Haute Couture Fashion and makes use of that keyword on a page that is about your online boutique. Now, if you make a minor change to the content of that page, by just adding a sentence about Prêt Couture, and make it a new page with meta-elements and tags that show up in search, as if that whole page was different than the first and all about Prêt Couture, what you will have is actually duplicate content with the exception of one sentence. Google’s algorithms might have not been smart enough to notice what you have done, but now you can expect to be penalized for this clever SEO stunt.
When Google crawls these pages it will instantly pick up what is going on and thus, rank it low. You might also be penalized even further, so although it may be a long arduous task to come up with new and original content for the other page with the same theme as the first, remember, being penalized is still a much more arduous and tedious affair.
- Quasi-duplicates on the outside
This will usually occur only when affiliates or partners pick up content from your website or copies of product pages from your site, which results in the appearance of content that is almost a true duplicate and sadly, it gets the same treatment by Google, which sees it as a true duplicate and is quite naturally not very impressed.
The way to avoid this form of duplication of content is to keep your pages stocked with fresh content and add an original and unique introduction to each product page or piece of content that has been borrowed. Another trick that usually helps is to introduce more user-generated content on such pages, with even a few pieces being enough to convince Google not to see your content and the affiliate’s page as duplicates.
Search pages on websites
Although this is usually relevant to websites of considerable size and shopping sites in particular, some medium sized websites might want to pay attention to this. Some websites will have an internal search feature through which visitors can make queries to find queries regarding the whole website’s contents, like individual product pages or specific articles or informational pages. This can result in those internal search pages showing up on Google’s search pages – a phenomenon that Google does not like at all. And the cause of Google’s discontent is pretty simple. As a search engine, they are interested in providing access to relevant pages to their users and not more search pages. The solution to this conundrum is quite simple although it takes a bit of time. You should instruct the spiders through the robot.txt option not to index these pages and block them from being followed.
Too many ads
Although originally not an issue with Google, it has become a major indicator of what a low-quality website is in Google’s dictionary, which is due to a pattern observed by the Google team, since a majority of the ‘thin’ marketing type low-quality websites we have mentioned in the first section had extremely high ad ratio, to the effect that they had started resembling cheap tabloid classified ads.
It is only natural for website owners to want to make their sites much more sustainable and easier to finance with as many ads they can get. But Google has started taking note of the ad ratio, thinking of a high quantity of ads on a webpage as noise and thus ranking it low or even penalizing it, even if the website might be providing high quality content. Remember, Panda has not only changed how algorithms work or a few technicalities here and there, but rather caused a deeper shift and reinforcement of Google’s philosophy, where now it is the users who are the measure of all things, making the search engine a user-centric enterprise. Since users would like to see fewer ads, so would Google.
A way around this issue, one through which you do not have to give up your advertising revenue and still not annoy users, is to concentrate on the ads that are performing well, the money makers and shedding the ones that are not relevant or high-earning.
An essential lesson to take from this issue in particular is to understand that the new SEO is all about the user, not the search engine. If you want to do well with the search engine, try to do well with the users sincerely, understanding what their needs are and what you would want if you were a user. The way to Google’s heart is now through the satisfaction of online users. This applies to the overall philosophy behind changing SEO and the internet-publishing world.
A chronology of updates since Panda 1.0
We will not present a rough chronology of the updates so you can place all these exciting, yet confusing developments, in context of where we stand today and how we got here.
February 2011 – Panda 1.0/Farmer rolls out
The first Panda acts as a heavy onslaught against content-farm type websites that have high quantities of low-quality content, mostly contributed by users.
April 2011 – Panda 2.0 is rolled out
Panda 2.0 enlarges the impact range of the earlier update from only US search activity to that of all English language search activity around the globe. Panda 2.0 is also the first time that the search engine publicly recognizes the use of data from blocked sites to affect ratings. However, only 2-3% of all searches are impacted by Panda 2.0.
May and June 2011 – Panda 2.1 and Panda 2.2 make an appearance
Both updates add minor tweaks to the previous 2.0 update released in April and impact even smaller percentage of searches. Panda 2.2 is designed to combat the issue of scraper websites, hijacking the rankings of sites with original content and rising higher on SERPs. However, Google is not pleased with the result.
July 2011 – Panda 2.3 is rolled out
According to Google, Panda 2.3 integrates new and more effective signals to discriminate between websites of high and low quality. Following the update, changes are noticeable as some websites begin to rank higher than before.
August 2011 – Panda 2.4 or the Language update rolls out
Panda 2.4 is designed to make algorithms do better with search results across a variety of languages.
September 2011 – Panda 2.5 rolls out
Panda 2.5 results in significant changes on search results in relation to large web outlets. FoxNews, Facebook and Android.com are some of the websites that rise to the top of search pages.
November 2011 – Panda 3.1
A pretty minor update influencing hardly 1% of search activity.
January 2012 – The January 30 Pack is introduced
The January 30 pack is a cocktail of little tweaks here and there to improve the overall user-experience of the search engine, and how convenient the results are, with regard to the searcher’s geographical and cultural context.
January 2012 – Search+ Your World is introduced
Google introduces personalization into its search activity, adding Google+ social data and profiles to search result pages, alongside a feature to switch personalization on and off. Critics say that Google+ is being given preference to relevancy in searches.
January 2012 – Panda 3.2 is rolled out
Panda 3.2 is brought out and Google says it makes no changes in algorithm. No one seems to understand the purpose for this particular update.
April 2012 – Penguin comes out
The Penguin update had been talked about for quite some time, as the update is designed to target spam and over-optimization. It picks up on things like keyword stuffing and abusive anchor texts.
May 2012 – Penguin 1.1 is rolled out
Google describes this update as ‘changes in algorithms designed to target websites that are not meeting quality standards and actively violating guidelines.’
The update also hits quite a few number of false positives and websites not guilty of any wrongdoing are also affected in terms of rankings.
June 2012 – Panda 3.7 comes out
Panda 3.7 has a bigger impact on search activity than Penguin according to some SEO experts although Google underplays the number and says only 1% of sites in the US and worldwide have been impacted.
June 2012 – Panda 3.8 rolls out silently
Panda 3.8 is a data only update with no changes in algorithm. Impact is minimal and somewhat unidentified.
July 2012 – Panda 3.9 is rolled out
Rankings seemed to fluctuate for 6 days while Google said it affected 1% of websites.
August 2012 – 7 results on SERPs and DMCA action
Google announces major changes in the month of August. Firstly, the main search page will only contain seven results as opposed to the earlier top ten. Secondly, Google will actively penalize websites which were repeat offenders when it came to copyright violations, using the DMCA takedown information as a tool.
September/October 2012 – Google puts out overlapping updates (EMD and the 20th Panda)
EMD, or the Exact Match Domain update, was designed to target just that – exact match domains. It became difficult for search engine online watchdogs to understand whether the spikes in traffic they were seeing were due to a new Panda update or the announced EMD update. Shortly, the next day Google announced that a new Panda update had also been rolled out and the SEO community decided to put an end to 3.X naming system and name it Panda 20, since it was the twentieth update to come out. Panda 20 was a major update, with an impact on 3-4% of search activity including non-English searches.
October 2012 – Penguin 3 rolls outs silently
Penguin 3 was a minor update and had an impact on 0.3% of search activity that was in English.
November 2012 – Panda 21 and Panda 22 roll out
Having a smaller impact than Panda 20, Panda 21 affected around 1.2% of English-based search activity. Panda 22 was a data-only update of a minor nature.
December 2012 – Panda 23 rolls out around Christmas
Having a greater impact than the previous Panda 21 and 22, Panda 23 impacted 1.3-1.4% of search activity in English. Although Google quietly maintained that it was merely a refresh.
January 2013 – Panda 24 comes out
Panda 24 is officially announced by Google and immediately affects about 1.2% of search queries as per official sources from the search engine itself.
March 2013 – Panda 25 rolls out
Panda 25 is considered a minor update. By this time, critics, and the SEO community in general, have started to lose patience with Google’s seemingly endless entourage of updates, which they always have to be on the lookout for.
May 2013 – The Mysterious ‘Phantom’ Algorithm update is spotted
With its exact details unknown and close to no word from Google on what the changes in algorithm entail, a substantial amount of websites report heavy traffic losses around the time the Phantom update is seen to make an appearance.
End of May 2013 – Penguin 2.0 or Penguin 4 rolls out
Although there is a lot of talk about the new Penguin before it comes out, once it does the impact is quite modest. The precise details of the changes brought about by Penguin are not clear but there is some proof to suggest that it was more focused on the page level.
May 2014 – Panda 26 (or 4.0) rolls out, too much hue and cry
Panda 4.0 becomes the most high-impact update since the first Panda (the Farmer update) with its impact range being 7.5% of English search queries according to official figures and around 9% according to SEO monitor websites. It seems to be an algorithm update plus a data refresh too.
July 2014 – The Pigeon takes flight
Google releases what it calls the Pigeon update, which is meant to provide a better local search results experience by building a connection with Google Maps. It has a massive effect on business and SEOs immediately, since it creates very strong ties between the local algorithms and the core, adding to the list of various signals already used by Google to achieve ranking.
September 2014 – Panda 4.1 (or Panda 27) is rolled out slowly…very slowly
According the search engine giant, this is a major update with a strong algorithmic portion. The approximate impact range predicted by Google is 3-5% of search queries, although those numbers are already starting to seem understated based on the feedback of webmasters and SEOs coming in.
According to Google, the latest Panda update will be the final one for quite some time and it is designed to create wider diversity in results by helping small to medium sized websites with high quality achieve better visibility. It is also supposed to help Google weed low-quality websites with more precision.
December 2014 – Pigeon expansion Pigeon update starts spreading from U.S. to other English-speaking countries, including the United Kingdom, Canada, and Australia.
April 2015 – Mobile update This update is related to mobile-friendliness. For the first time, mobile rankings will differ from desktop rankings, and the sites with responsive design will be given better rating.
October 2015 – RankBrain Google announced that machine learning is a part of the algorithm and one of the most influential factors for ranking websites.
February 2016 – AdWords Shake-up This update affected paid results mainly. Google entirely removed right-column ads, and rolled out 4-ad top blocks on commercial searches. Although aimed at paid results, this update did affect organic results as well.
May 2016 – Mobile-friendly 2 This is another update that favors mobile-friendly sites.
September 2016 – Image/Universal drop and Penguin 4.0 announcement Statistics reported 50% drop in SERPs with image results, causing ranking shifts. Google announced that Penguin is now part of the core algorithm. It was also revealed that Penguin 4.0 will have several phases as the changes are slowly introduced. Phase one devalues bad links instead of penalizing the entire website.
October, 2016 – Penguin 4.0, phase 2 The second phase of this update was aimed to reverse all previous Penguin penalties.
A majority of SEO experts are of the opinion that the latest Panda update is gentler, more empathetic and a result of Google learning its lessons from the previous updates and taking the feedback and complaints of webmaster, bloggers and site owners seriously. But only time will tell, as the wrongly affected websites post-Panda begin to make a recovery and flourish.
Now let’s move towards the last section, where we discuss what all the chaos resulting from Panda and other updates means for you as a webmaster, blogger, business owner or SEO professional.
Getting ahead in the post-Panda internet world
The world of online publishing and search engine optimization keeps changing and if you do not learn the new game and get ahead of the curve, you will not only lose out; you might not get to play anymore.
Brief lesson for a radically new SEO environment
First of all, you have to realize that if you are a webmaster or anyone in charge of building better relationships with search engines to achieve visibility, your job has changed in a radical way and probably for the better. You are now answerable to users much more than the search engines, because Google itself will only rank you as high as the amount of favor you hold with users.
After these changes, building better relationships with search engines now means building better relationships with online users . The way to do that is focus on user experience.
Try to bring about a paradigm shift in your work and design philosophy, where you are working to entertain, inform, serve and provide convenience to users, since happy users are the best signal that you can send to a search engine. Clever keyword densities and running after links are not enough anymore. The search engines have moved on and you need to as well.
As an SEO professional or webmaster, you are now responsible for:
- Amazing design
- Responsive design
- High-quality content
- User related metrics
- Community building
- Interaction with the users
- Networking with other websites/blogs
So, in a sense, you are more of a web consultant or strategist rather than someone who sits around figuring out how many times a keyword needs to be used and how you can convince somebody to spare you a link. The horizons of SEO have broadened immensely and in qualitative terms.
Another conceptual or cognitive change that needs to come about is that of ceasing to think of search engines as robotic machinery or wall of code that needs breached. Google is reinventing itself to reward high quality and well meant content that adds value to the internet and focuses on the needs and experience of users. If you can do that, you do not really need to worry about being slapped by a Panda update or getting lost in SERPs.