Google Panda 4.1: How it Impacted Sites Organically

Google’s Panda update, which is aimed at filtering out low-quality content from the Google search results, has yet again rolled out, this time affecting 3 percent to 5 percent of search queries.

The news hit on September 25, and early analysis of the results here at BrightEdge yielded some interesting findings. More on that in a bit.

You may remember in 2013 that Google said it would no longer be announcing Panda updates, as it would be baked into the algorithm as a more regular, gradual rollout.

But this announcement comes with a note from Google that the infrastructure to Panda has changed a bit.

From the announcement on Google+:

Based on user (and webmaster!) feedback, we’ve been able to discover a few more signals to help Panda identify low-quality content more precisely. This results in a greater diversity of high-quality small- and medium-sized sites ranking higher, which is nice.

Historically, Panda went after sites with different types of weak content such as thin content, which is basically a lack of content; duplicate content, usually at scale; and machine-generated content, like spun content.

But according to BrightEdge’s initial analysis of sites that have been impacted after Panda 4.1, we’re seeing some interesting observations that may go beyond the initial criteria of Panda and could hint at the new signals working together to surface higher quality sites in the search results.

How Panda 4.1 Impacted Sites Organically: 3 Patterns

BrightEdge tracks a large set of independently researched queries across several verticals on a daily basis, and its professional services team has been monitoring movements closely over the past week to identify trends.

We observed some interesting patterns amongst the sites that did not fare well after Panda 4.1, and wanted to share that data along with some tips with readers. We also want to stress that we have not yet had time to form definitive conclusions about what we’re seeing.

Our starting point was to pivot our data in different ways to see which sites experienced dramatic gains and losses in first page rankings and Top 100 rankings, plus loss or gain in Share of Voice. We then analyzed the heavily impacted sites for notable characteristics.

Here’s what we found in terms of types of sites and patterns that emerged for those websites that did not fare well.

1. Infotainment and Media Sites: Image Heavy, Ad Heavy

Several negatively impacted sites, particularly in media, dedicate valuable real estate above the fold to ads, and also packed a lot of images on the page – and sometimes a lot of ads as well.

Just some of those sites included:

  • EOnline.com
  • Essence.com
  • HollywoodLife.com
  • Independent.co.uk

Generally, the editorial-to-ad ratio was low, and the image count was high, which could impact page load time and thus user experience, especially on mobile devices.

Screenshot
EOnline.com
Screenshot
Independent.co.uk

Remember Google’s page layout algorithm has set the tone for sites that use a lot of ads at the top of their pages; pair that with the fact that Google is placing a high value on sites that offer a good mobile experience in general, and you can see how sites like this might be negatively impacted.

With the latest Panda rollout, we think your content-to-ad ratio and image heaviness may count more than ever.

While visuals are great for engagement, use them with the right intent and mind their placement. Invest in high-quality images, but not at the expense of high-quality copy. Ensure the page loads quickly, and that ads and images augment rather than obstruct the site’s experience.

On that note, there were some image-heavy sites that actually seem to have gained ground like USMagazine.com (search is a zero sum game, after all), and they do seem exercise at least a little more moderation with images and ad units.

Screenshot
USMagazine.com

2. General Informational Sites: Generic Content

Once again, we see a set of mostly informational sites amongst those most negatively impacted by Panda. The topical reach of each of these sites is extensive, but the length and quality of the content is not always uniform.

Some of those sites include:

  • Answers.com
  • HowStuffWorks.com
  • HubPages.com
Screenshot
Answers.com
Screenshot
HubPages.com

Of those that benefited were a combination of highly recognizable media brands and more targeted niche media (these would be some of the small and medium websites that Google mentioned would benefit):

  • GlassDoor.com
  • NYTimes.com
  • OrganicGardening.com
  • Slate.com
Screenshot
OrganicGardening.com
Screenshot
PopularMechanics.com

Share of Voice continues to shift to rich informational sites, even for queries that are more transactional in nature. To compete, brands need to get creative, leveraging proprietary data and new ideas, and putting their communications team to work to create differentiated, compelling content.

3. eCommerce Sites: Thin Content

Again, we see rich, vertically oriented information sites competing with more commercial sites for the eCommerce Share of Voice. Thin content comparison sites and e-tailers did not fare well in many cases.

Some of those sites included:

  • CheaperThanDirt.com
  • FindTheBest.com
Screenshot
CheaperThanDirt.com
Screenshot
FindTheBest.com

That said, richer content around products is typically rewarded. Smaller sites can compete if they focus on best-in-class content and experience that aids in the shopping experience, like the following sites:

  • Cabelas.com
  • CNet.com
  • Nike.com
  • Otterbox.com
Screenshot
Cabelas.com
Screenshot
CNet.com

Odds are that the comparison shopping site will have to invest in substantial content (detailed reviews or articles, for instance), in order to provide a significant value and compete.

Google stated that the latest Panda update should be rolled out by the end of this week. So if your site has been impacted, you should be able to tell right around now. We’ll keep a close eye on the data here at BrightEdge, and let you know what we find out when we’ve been able to analyze the data further.

Thanks to Justin Thomsen at BrightEdge for driving the data analysis for this piece.