The Panda Update, 1 Year Later: A Guide to What We Know

It’s hard to believe, but tomorrow will mark one year since Google launched the first “Panda” algorithm update. Originally impacting 11.8% of US queries, Panda sent SEOs and “Panda-lized” webmasters rushing to find the fix.

One year later, we know a lot more about Panda, with several clients recovering from the update. I’ve compiled what now know about Panda-citis and how to cure the affliction.

Lone man walking abandoned streets56 Weeks Later: Where are all the content farmers...?

Panda Is More Than a Penalty, Affects Everyone

Those who were largely unaffected by Panda have largely ignored it. This is unwise; however, as Google have made it clear that the update “will provide better rankings for high-quality sites—sites with original content and information such as research, in-depth reports, thoughtful analysis.” Along these lines, Richard Baxter explained how Panda represents a new ranking factor.

We have already seen examples of Panda feeding high-quality sites on the remains of content farms. For example, Cyrus Shepard shared some SEOmoz data showing that SEOmoz had seen a 49% surge in traffic since the first panda update. The only sites that can safely ignore the implications of the Panda are those sites that don’t want additional traffic.

SEOmoz Traffic through Panda updates

 SEOmoz Sacrifices Adorable Animals and Quality Posts to Panda

Diagnosing Panda Penalties

Mark Nunney has written the Panda survival guide, which remains up-to-date as one of the best resources on Panda. The guide also includes a step-by-step analysis to seeing whether a drop in site traffic was actually the result of Panda.

Quality or Die

Google elaborated on the Panda update, releasing what is now known as the Panda Survey. The search giant also hinted here that the focus on quality will continue, and reach far beyond Panda updates. Not to be left out, Bing’s Duane Foreseter wrote a post on high-quality content, revealing some of the signals that Bing watches. Quality signals from both search engines fall into the following categories:
  1. Quality content site-wide
  2. Quality design and positive user experience (UX)
    1. Engaging, thoughtful, and interesting (generally not all text/all images)
    2. Limited duplicate/thin content
  3. Social and branding signals
  4. User metrics
While quality SEO consultants have been encouraging the above for years, these items are no longer optional.  Let’s take a moment and look at what we know about each of these elements.

1. Quality Site-Wide Content

Panda’s greatest immediate effect was the demotion of low-quality content and the promotion of high-quality content. A year later, we have seen businesses rapidly change their content strategy.

Dr. Pete wrote two pieces about making content more Panda-proof: one about duplicate content, and another about thin content. These pieces are both worth a read, because, to use Dr. Pete’s words, “It used to be that duplicate content could only harm that content itself …  Panda made duplicate content part of a broader quality equation – now, a duplicate content problem can impact your entire site.” I especially recommend the post on duplicate content, as it gets in-depth on diagnosing and fixing common problems.

Rand put the need for high-quality content best in a now- famous Whiteboard Friday on post-Panda best practices when he said,

Rand Fishkin“So a lot of time, in the SEO world, people will say, ‘Well, you have to have good, unique, useful content.’ [pause, sigh] Not enough. Sorry. It's just not enough.”
Rand and other industry experts have made their position clear: to make site content work, it must now be exceptional and engaging.

2. Quality Design, Positive UX

Five or six years ago, it was actually possible to get almost anything to rank, regardless of how the site looked. That’s simply not true anymore, and Panda is simply another nail in the coffin of poor-quality site design. Much like space-filling content, a poor-quality site design just doesn’t cut it anymore.

In a Wired interview and Vanessa Fox’s recap of SMX, Google’s Matt Cutts seems to confirm what had been suspected since day one: that having a low content-to-ad ratio is one of the factors Panda considers. In any case, sites loaded with excessive or intrusive ads infuriate users, and should be avoided. Cyrus later suggested how to place ads in a more Panda-friendly manner.

Conversion optimization and UX improvement has become a serious part of Distilled’s own offerings since the Panda update, as integrating design and user experience improvements with traditional SEO and social optimization makes more sense every day.

3. Social and Branding Signals

We have seen Google’s ability to differentiate between brands and “generic” sites grow rapidly. While we’re still speculating as to what search engines use to determine the existence of a brand, it has been clear that sites with strong brands are at a much lower risk of penalty, and Panda was no exception. Matt Cutts said the following in a Wired interview about Panda:
Matt Cutts“We actually came up with a classifier to say, okay, IRS or Wikipedia or New York Times is over on this side, and the low-quality sites are over on this side.”
To see that in action, look at the initial list of initial Panda winners and losers, none of the losers are truly strong brands. That’s less interesting, since the majority of these sites are also content farms with overly-aggressive ads.  More interesting, though, was that the vast majority of winners have strong branding. If ehow.com stands out, note that it did get hit by a subsequent Panda update.

Empty about page

Clearly not a brand.

The message is clear: make it obvious to Google and users that the site is run by real people. Register your business, establish brand pages, and do what you can to get people talking about your unique brand.

4. User Metrics

It has become clear that Google relies on user metrics for the Panda update. In the recently-mentioned Whiteboard Friday, Rand explained the basics of how user metrics work in combination with machine learning to allow Google to aggregate and test adjustments based on user behavior.

Site design and quality content are our tools, and improving the experience for users (as far as it can be measured) is the goal. Unlike users, Google cannot detect quality directly, and must rely on quantifiable actions to infer usefulness.

Before we dig into our analytics programs, can we begin where Google begins, and gather data on how users feel about our site? How can one tell if a low time on site is due to a terrible site design or poorly-written content? Our own Will Critchlow spoke with Rand about replicating the Google survey referenced above. Mechanical Turk (or similar) surveys are a fast and easy way to get feedback from users. Having run a few of the prescribed tests on sites hit by Panda, we have been able to diagnose problems of trust, design, content, and more.

Every site owner should also dig into analytics looking for potential problems. Where are the pages with the worst bounce rate and lowest time on site? What queries have the worst metrics? We should start by weeding out the lowest performing pages, as Richard Baxter suggested.

95% bounce rate vs. 40% average

Clearly there is a problem with this page.

no backWhen it comes to tracking pages with high bounce rates, are these users going back to Google, or are they clicking on outbound links on your page? Joost de Valk has explained one way to track outbound clicks. By subtracting these events (outbound clicks) from the bounce rate, we can find the true bounce rate, which is at least a more consistent number to work with.

For a winning formula, combine that sign of query relevance with what Dr. Pete argues are the two user metrics that really matter: SERP CTR and dwell time.

Inbound Marketing Means Writing to, for, and by Humans

In the “totally missing the point” category, many article sites are apparently requesting articles of 400+ words, with no serious changes to quality review. Simply creating longer fluff pieces is likely to have little to no effect in recovering from Panda (or similar, upcoming tweaks).

The age of quality is quickly approaching, and it is sheer foolishness to throw every possible bit of content to the wall and wait to see if it sticks. For those who are not convinced on the need for top-quality content, I invite you to first read the posts and guides linked to above. Improving content quality and site design will:

  • Improve your rankings and site traffic
  • Increase the likelihood of receiving natural links
  • Increase the value of your current traffic
  • Improve the author’s reputation in the industry
  • Make the world a better place filled with less crappy content
For those who feel like they do not have time to create quality content, you may wish to revisit your content strategy and its place in your marketing plan. Inbound marketing strategies hinge entirely on quality content, and there’s no longer much point in writing sub-standard blog posts every other day – or ever, for that matter.

Get blog posts via email