Google Panda Penalty and its History

Google Quality Signals

In this series, I am going to look back at the history of the Google Panda Update, what it is, how it affects websites and what are some of the things to avoid to stay on the right side of Google. Finally, what are the key things you need to do to recover from a Google Panda infringement.

Each blog post will build on the previous one, so please revisit to read the next instalment. More importantly, I would love you to offer your views (and any corrections) through the Disqus comments at the bottom of this page.

Panda (aka Farmer) update on 23rd February 2011

This was Google’s first major algorithm update which affected many websites hard. It was suggested by Google themselves that up to 12% (11.8% to be precise) of search results would be hit.

At this juncture, it was believed that Google’s Panda Update (or Farmer update – more on that later) was designed to target thin content sites, websites with high advert to content ratios, content farms and various other poor quality signals.

It was launched in the USA first and moved across the Atlantic reaching the shores of Europe in April 2011.

Who was affected at this Panda Update on the 23rd February 2011?

Some of the notable websites alleged to have been affected at this time were:

• Ezinearticles.com down 90%
• Fixya.com down 80%
• Articlesbase.com down 94%
• Business.com down 93%
• Doityourself.com down 77%

As you will agree, the losses in traffic are quite substantial. Of course, there is a degree of “outside eyes” looking via 3rd party software and estimating traffic before and after the 23rd February 2011 date to calculate these losses, and only each of the five website owners will know the exact figure.

Who were the winners of the update?

As with all changes, there are winners and losers. From research conducted at the time, there were a few well-known websites which appeared to get a boost from this Panda update, namely:

• Wikihow.com up 79%
• Facebook.com up 12%
• Ehow.com up 14%
• Techrepublic.com up 16%
• Instructables.com up 17%

Again, some big differences, perhaps the wikihow.com percentage is a bit high, but time has shown that the “wiki” related websites have dominated the top of Google SERPS.

What did the Panda/Farmer update look at?

At the time, this was one of the biggest changes to Google’s algorithms that marketers had seen. And in typical Google fashion, they didn’t reveal what they are about to change, most of the time they don’t give any advance warning, and rarely do they give any detail.

Marketers generally have to dissect what happened afterwards and formulate their own conclusions. This is what marketers had to do on this occasion. Some key observations when comparing the winners and the losers were:

Observation 1

Website pages with fewer and less obvious advert blocks (and fewer ads towards the top of a page - which we later found out was one of Google’s pet hates as they launched a separate update to attack this) as a whole fared better after the update. The websites with numerous adverts which tended to overwhelm the page (and therefore user experience) tended to drop down the rankings.

Observation 2

User interface and user experience (UI/UX) appeared to be a factor, with websites that appeared well laid out, gave a feel of higher quality and correctly rendered to the end user tended to rise up the rankings. Whereas those sites which, on the face of it, failed to invest in the user experience and looked like a dog's dinner, tended to be a loser in the update.

Again, this would later appear to be true as Google asked webmasters to consider the user, have websites load quickly and notably, to give Google bots access to CSS and JavaScript files (normally sites blocked these in their robots.txt file) so that the Google bots could properly render web pages to see how they looked.

This point would later be confirmed when in Google Webmaster tools (now Google Search Console) they offered the fetch and render option to webmasters, so site owners could make sure their site looked similar to Google bots as it did humans.

Observation 3

Website content seemed to be one of the most noticeable points. Websites thin on content and of lower quality tended to lose out. These were sites that end users could post content on for free and who often reused the same content across multiple websites.

Whereas websites which were well edited, had unique and authentic content and also well laid out content which the user could easily consume, tended to gain ranking positions after the Google Update.

Confirmation from Google’s Matt Cutts and Amit Singhal

The headline mentions confirmation, but in reality, it is a “sort of” confirmation. During an interview at a TED conference in 2011, both Matt and Amit were asked a series of questions from which we have to deduce the underlining algorithm factors. The following points summarise their answers:

• After the Caffeine update in late 2009, we received a lot of good fresh content, and some not so good. The random gibberish was gone, but now shallow content was being indexed.
• We wanted to keep it strictly scientific. We basically used outside testers to rate questions on our standard evaluation system.
• Questions included, do you consider this site to be authoritative? Does this site have excessive ads? Would you be comfortable giving this site your credit card? Would you be comfortable giving medicine prescribed by this site to your kids?
• Based on the answers, we defined what we considered low quality. We also used the Chrome site blocker to compare the answers and found an 84% overlap.
• We came up with a classifier to evaluate each website and assigned them to quality websites vs. low quality websites.
• We received an email saying that content farms had been ranking higher that government websites and now the government website are ranking higher afterwards – so the update has done what we said it would do.
• Some people say you should be more transparent and your changes are only favouring those who advertise with Google. This was categorically denied. Matts Cutts mentioned that they would never be 100% transparent as the bad guys would know how to get back up the rankings.

If you analyse the answers yourself (the full interview can be viewed here) it is clear to spot the “quality content” undertones. Website owners who originate highly authoritative, researched and well-written material, will rank higher than those who don’t put in the effort and are of poor quality.

The interview provided some “confirmation” of what the marketers had deduced themselves after the Google Panda Update on the 23rd February had occurred.

Finally, where did the name Farmer (later known as Panda) come from?

Before it was officially announced by Amit Singhal, one of Googles top web spam team members, those in the industry thought it was because they were fighting “content farms” and by default associated it with farmers.

As we know now, it was named after Navneet Panada. Navneet is from Odisha in India and graduated from IIT KGP and has a PhD from UCSB. He was an engineer on the Google team who devised a large part of the machine learning algorithm which made the Panda/Farmer update possible, and as a result, the name stuck. The farmer update lost its name and the Panda update was born.

Next time…

I will look at how the Panda update evolved into Panda 2.0 on the 11th April 2011.

About the author

David Reid

David previously worked with Europe’s No.1 Marketing Agency & the World’s No.6 biggest Integrated Marketing Agency, seeing first-hand how £150 million+ campaigns are executed to create household brands. Now he is the CEO of a specialist marketing agency helping companies of all sizes.