SEO

6 Months with Panda: A Story of Complacency, Hard Decisions, and Recovery

In the fall of 2011, I was contacted by the Director of Marketing for a B2B company.  The company’s website had been hammered by Panda, and he didn’t know what to do.  I could tell very quickly that his team was truly baffled.  The company and website have been around for a long time, the site contains a boatload of ultra-high quality content, and used to rank for thousands of keywords.  The Director of Marketing made sure to point me to their top articles, whitepapers, blog posts, etc. after our initial conversation.  I can tell you that he was right; they had a ton of great content.

In addition, the site’s link profile was not only clean, but it was ridiculously impressive.  They had earned tens of thousands of links, many from relevant and powerful sites in their industry.  Needless to say, I was fascinated by this story, and I was eager to begin assisting them.   Although the company will remain anonymous, I received approval to write this post covering the details and key learnings.  Everyone involved agreed that there are some great points here for others hit by Panda, so they were cool with me covering what happened.

The Plan of Attack

Since I was contacted after their first Panda attack, I had a lot of research to do.  I wasn’t familiar with the company, website, key players, content, etc.  I began my work by heavily analyzing Google Analytics and Google Webmaster Tools data to determine the drop in impressions and traffic.  I also focused on determining which content was heavily impacted.  Comparing the timeframe before and after the Panda attack revealed some alarming findings.  I tracked those metrics back to the content on the site and began to see some serious problems.  From an outsider’s perspective, there was a trend occurring with the quality of content being published in the months leading up to getting hit by Panda.  I’ll get to that shortly.

They Experienced the Classic Panda Drop in Traffic:
panda drop ga1 6 Months with Panda: A Story of Complacency, Hard Decisions, and Recovery

Before I move on, if you believe you were hit by Panda or Penguin recently, definitely check out my post about how to determine which algorithm update impacted your website.  You want to make sure you focus on the right algo update before taking action.

Hi, I’m Glenn… Let’s Talk Panda

Once I had a wealth of analytics data, I began to interview key stakeholders.  I wanted to know the types of changes the website experienced leading up to Panda from an execution standpoint.  This involved interviewing the technical players involved, as well as the people driving content.  As the interviews went on, the evidence began to grow.  I started to get a good feel for what happened.

Complacency Can Be a Killer

As I mentioned earlier, the site in question held a lot of high quality content.  There were articles, blog posts, whitepapers, etc, and the company had built this up over years (legitimately).  The site was rewarded with outstanding search engine power and performance and ended up ranking for thousands of target keywords.  But that’s when a critical problem started to creep in.

To put it simply, the company became complacent.  I noticed a big drop-off in the quality of content leading up to the Panda attack.  The posts and articles were thinner and didn’t really provide the level of knowledge and thought leadership that they used to provide.  Some were only a paragraph or two that linked out to other stories on other blogs.  It’s also worth noting that there were times some of those thinner posts linked out to partner websites (companies that had a business relationship with the company I was helping).  That said, it was a very small percentage of their content (less than 3% of the content I analyzed).

panda content quality1 6 Months with Panda: A Story of Complacency, Hard Decisions, and Recovery

Inherited SEO Power – The Silent Killer

Unfortunately, those thin articles and posts that were published leading up to the Panda attack inherited search power based on the domain authority of the website.  Thin posts began ranking for competitive keywords when they had no right to do so.  A lot of traffic was coming through to the content that simply wouldn’t answer questions.  Matching search queries with the content easily revealed this problem. As thousands of visits arrived from Google to these low quality posts, user engagement dropped drastically.  Bounce rates were high, dwell time was low, and this was not good for the company’s organic search situation.

Thin Content Was Ranking Highly, Surging in Traffic (when it shouldn’t be):
panda seo inheritence 6 Months with Panda: A Story of Complacency, Hard Decisions, and Recovery

The “Your Baby Is Ugly” Conversation

Once I determined the main content problem, I set up a meeting to cover my findings.  I had to explain that their complacency problem manifested itself in months of lower quality, thin content.  Upon showing them my findings, comparing older high quality posts to the newer posts, explaining the inherited search power problem and surge in traffic, they started to get it pretty quickly.  The room grew eerily quiet.  They now realized how the massive drop in traffic occurred, and it wasn’t a technical problem.

One of the most important processes I took the team through during this meeting was pulling some of the thinnest posts and articles and simply asking:

  • Objectively, do you think this is a thought leadership piece?
  • Does it answer your questions about the topic?
  • Would you share this on Twitter, Facebook, LinkedIn, and Google+?
  • Should users from Google searching for x, y, and z be driven to this content?

Nobody answered yes to these questions.

Taking Action – Gut, Rewrite, and Revamp

Now that we knew the core problem, we formed an action plan, and I made sure that everybody was on board with moving fast.  With regard to the thin content situation, I explained that their best options were to nuke them or rewrite them. We were able to go through the list of content and discuss the posts and articles.  If a piece of content was flagged as “filler content,” then it was nuked, but if it was a topic that would help prospective customers, then the content was rewritten.  In addition, any thin post that linked to a partner organization was nuked (to be safe).

panda decision flowchart 6 Months with Panda: A Story of Complacency, Hard Decisions, and Recovery

6 Months of Hard Work

I made sure to explain to the company how Panda worked, so they wouldn’t expect immediate change.  Since Panda rolls out periodically, we needed to move fast, make serious changes, and then track everything we could, leading up to the next Panda update.  After the initial gutting and rewriting of content, I made sure the company kept pumping out high quality content.  My recommendations to the content team were, “Focus on building outstanding content. Don’t look at Google Organic traffic and just keep going.  Make believe you can still rank well, and that you’ll be getting a ton of Google Organic traffic.”  That worked.  The posts that followed were some of the best they had written in a long time.  Quality was back, which was great, but Google Organic traffic was stuck, which was expected.

3 Months In – A False Alarm or a Test?

About three months in, traffic from Google Organic jumped again.  This was across older posts that used to rank and newer posts that should be receiving traffic.  This stuck for just a few days and then dropped back down again.  I was ready for this type of behavior, but I was still extremely disappointed (and so was the company I was helping).  Was this some type of test Google was issuing to see if the quality of content was high enough?  Was it testing user engagement?  Regardless of the reasons, the post-Panda bump was gone for the time being.

A Temporary Panda Recovery…
panda bump 6 Months with Panda: A Story of Complacency, Hard Decisions, and Recovery

A Cross Channel Effort Emerges

After both Panda and Penguin, the need to effectively utilize other channels has become extremely important.  For example, social media marketing, e-mail marketing, paid search, building high quality referrals, etc. Well, during the Google Organic drought this company was experiencing, I mapped out a cross channel plan.  Some of the components were already in place, but they just weren’t being utilized to their fullest extent.  For example, they had a pretty strong Twitter following and LinkedIn following.  They needed to pump up their Facebook following and activity, Google+ activity, and revitalize their e-mail marketing program.  They definitely put a lot of time into this, and it worked well.  The only problem was that social and e-mail were “ spikey.”  That’s really where you see the impact of strong organic search traffic.  It’s residual.  Posts that were written a year or two ago can still receive traffic on a daily basis.

6 Months Later, The Panda Has Left the Building!

I don’t know if this is going to scare the heck out of you or make you feel better about your own situation, but it took six months for my client to shake the Panda.  And that’s from a company with high-quality content and a long track record of doing the right thing.  In addition, the company moved fast to gut all their thin content and rewrite some of that content into high-quality, valuable posts and articles.

Ironically, I was analyzing paid search for the company when I noticed the spike in Google Organic traffic.  The posts that were rewritten were receiving a good amount of Google Organic traffic, the newer content was getting a lot of traffic, etc.  Life was good again in Google Organic land.

I finally noticed the first signs of recovery:
panda recovery1 6 Months with Panda: A Story of Complacency, Hard Decisions, and Recovery

Key Learnings

So, based on this six-month Panda recovery story, I wanted to list some key learnings.  My hope is that the following bullets can help you determine what’s going on with your own websites and how to best address those problems:

  • Any site can get hit.  Don’t think that your well-aged, high-quality site can never get hit by Panda.  It absolutely can.
  • Moving fast doesn’t mean the site will come back quickly.  It took six months for this site to come back, and they did everything right.
  • Don’t be afraid to gut content or rewrite it.  Be objective when analyzing the content. Know when your baby is ugly.
  • Have a strong analytics strategy in place NOW.  That’s before any problems occur.  This will make your post-Panda analysis much easier and straightforward.
  • Complacency is a killer.  Don’t sit back and think your older content is so great that you don’t need to keep pumping out high-quality content.  This company did, and they got hammered.
  • Low-quality and thin content can inherit SEO strength, based on your domain authority.  This can end up driving a lot of traffic to thin posts, but that traffic will bounce faster than you can say, “Panda.”  More traffic isn’t always a good thing if that traffic is landing on low-quality content. User engagement matters.
  • No Panda recovery is a full recovery.  The reason for this is simple.  If you are gutting content, and that content received a ton of traffic in the past (but shouldn’t have), you can’t include that traffic in your comparison.  Basically, you shouldn’t have had that traffic in the first place.  Understand this when presenting a comparison of Google Organic traffic before and after Panda.

Summary – Scarred, but in a Better Place

And that’s, hopefully, the end of this Panda story for the company I am helping.  Their Google Organic search traffic is back, but their scars run deep.  I can tell you that nobody I’m dealing with at the company takes that traffic for granted now.  The content they are producing now is consistently high quality.  I am still helping them on a regular basis, which helps everyone stay on track.  Although six months with Panda seems extreme for a company that moved quickly to rectify problems, at least it shows you can recover.

If you leave this post with just one key learning, it should be that you shouldn’t let complacency sink in.  Like I said earlier, it’s a killer.  Avoid it at all costs.

 6 Months with Panda: A Story of Complacency, Hard Decisions, and Recovery
Featured SEO Writer for SEJ Glenn Gabe is a digital marketing consultant at G-Squared Interactive and focuses heavily on SEO, SEM, Social Advertising, Social Media Marketing, and Web Analytics. Glenn has over 18 years of experience and has held leadership positions both in-house and at a global interactive agency. During his career, Glenn has helped clients across a wide range of industries including consumer packaged goods (CPG), ecommerce, startups, pharmaceutical, healthcare, military, education, non-profits, online auctions, real-estate, and publishing. You can follow Glenn on Google+ here.

Comments are closed.

43 thoughts on “6 Months with Panda: A Story of Complacency, Hard Decisions, and Recovery

  1. Great stuff thanks Glenn Gabe :)
    More proof of what Google continues to claim about it’s algorithm. I am not sure why more companies don’t go to Google directly with questions. They are very helpful believe it or not!

    1. Hey, thanks Katia. I’m glad you found my post valuable. Although it was a long road back, it did eventually come back. That’s the good news. The 6 months of hard work and waiting was frustrating, though. :)

    1. Thanks Jennifer. I appreciate it. It was a fascinating project to work on, to say the least.

      It was amazing to analyze the data, present my findings, and then move quickly with an action plan. The mixture of complacency and inheriting seo power was a lethal combination. I fear many others are going down that path and not realizing a giant panda is approaching. Hopefully this post can change the outcome. :)

  2. Glenn,
    Thanks for writing this. I was hit by 3.6 (in April), and I’ve been busting my butt ever since to beef up all of my content and make every page world class.

    It took you guys 6 months from start to finish. At what point during that process did you feel like most of the shallow content had either been re-written or removed? 3 months before recovery? 2 months?

    I know that every site is different and there are never any guarantees in this business, but I’m trying to get some kind of rough idea of what I can expect to see after I’m done with the rewriting process.

    1. Thanks for your comment Kevin, and I’m sorry to hear you were hit by Panda.

      We moved very quickly on addressing low quality content. I’d say that we had 99% of it nuked or rewritten within 8 weeks of the initial drop in traffic. That’s why I was hopeful when we saw the temporary recovery 3 months in. I was hoping it wasn’t “temporary”. :)

      Have you also been adding new content while revamping the low quality content? That was an important aspect to what we were doing. We kept pumping out extremely high-value content, sharing via multiple channels, etc. We ended up gaining a nice amount of powerful inbound links while doing that. I’m sure that helped the situation. Let me know what you’ve been doing. Thanks again.

      1. Cool, that sounds about where I’m at. I’m hoping to finish the rewrites next week, which would be me right around the 8 week mark. My rewrites have increased the content each page to anywhere between 500-800 words.

        I have still been adding new content, but not as frequently as I was before (about one piece a week, instead of five). This is strictly an informational resource site, so it’s not hurting my readership to slow down the publishing. Hopefully it isn’t hurting me with Google.

        This whole thing made me reassess what the goal of the site is, and I took a chainsaw to every piece of content that didn’t work towards that goal. I actually deleted probably 1/3 of the content (some of it was very good, but didn’t fit with the direction I want to move in).

        I’ve also continued my link outreach. Thankfully, acquiring high quality links is pretty easy for me in the niche I’m in. My content happens to fit a particular need that government agencies, colleges, and libraries like, so it’s not too hard. The links send consistent traffic on their own too, which is nice.

  3. When you nuked the shallow content, did you just delete that page, and 301 it somewhere else (home page, or another article), or did you just let that page 404?

    Thanks for the great insight on this, very good example!

      1. Glen, would noindexing the thin content pages yield the same result as a 404? In terms of algorithm evaluation?

      2. Mike, good question. We wanted to make sure the pages were gone, with no possible way for users to engage the content. That would be a good test, though.

  4. This ties in nearly exactly with an article I wrote talking about the original Panda of 2011
    Apologies for adding a link but I hope you find this interesting – http://www.v1seo.co.uk/11/new-seo-tips/

    I was talking about a theory I had that I was calling Panda Balancing in which I believed too much thin content could have a negative effect on good content and the whole site in general.

    I came up with free options or re balancing the site;

    1. delete poor quality pages
    2. leave old pages but replace with new ones
    3 update and improve old pages

    Option 3 is obviously the hardest of the three, the most work and the most time consuming.
    Option 3 also appears to be the one closest to what you did that produced good results.

    As you say, it can be a long slog, with lots of work involved, but those prepared to put in the work appear to benefit most. Unfortunately most people don’t look at the long term and just want the quick fix though :(

    1. Thanks for your comment Terry. I’ll check out your post today.

      The problem I mentioned in my case study had to do with thin content inheriting SEO power, based on the domain authority of the site. The thin posts then ranked well and had a boatload of traffic coming to them (when they shouldn’t have). Then user engagement was horrible and made the site susceptible to Panda. It’s a dangerous problem… Like I said in my post, it’s a silent killer. :)

  5. Absolutely fantastic job and write-up – the best post-Panda recovery story I’ve read.

    It goes to show that if you put the work in and churn out regular quality content you can still rank well in Google :)

    Dan

    1. Thanks Dan. I appreciate it. I’m glad you enjoyed my post.

      This case study definitely shows that if you move fast, make hard decisions, keep driving forward, and be patient, good things can happen. I wish it didn’t take 6 months, but better late than never. :)

  6. I know of a lot of sites that still haven’t recovered from Panda and was then hit by Penguin. Some have just decided to start over.

  7. Thanks a ton, Glenn! One of the best articles I have read on Panda attack so far. Got great insights. We’re also in the process of purging out content on our client sites to combat panda attack. What we noticed is that, we had experienced ranking dip for many of our keyterms and there was a gradual dip in overall traffic. We monitored virtually all aspects and decided to remove 4 pages from our site which was mirroring content not fully of other pages. It worked and we’re regaining ranking position.

    1. Thanks for your comment Krish, and I’m glad you found my post valuable. I’m glad you were able to analyze your content and find the right pages to nuke. Were there only 4? That seems light, but I’m not familiar with the site. Thanks again.

  8. Glenn you have done a brilliant job

    Its been over an year for Panda updates there were no such post I’ve found detailing so clearly. I would like to say Thanks for your clients who gave you the permission to share the details.

    1. Hey, thanks Raviraj. I really appreciate it. I was really happy to receive clearance to write the post. There are some great Panda learnings, based on their situation. I hope it helps others that were hit.

  9. “Any site can get hit.”

    I think that’s the most important takeaway. It doesn’t matter how old and established your website is, it’s still at the mercy of the algorithm. Don’t ever assume it can’t happen to you. You can hope it doesn’t and do everything possible to prevent any issues, but always be prepared for the fallout.

    1. Right on Nick. I think too many companies think their well-aged sites can’t get hammered. This example shows they can get hammered. Never take it for granted! :)

  10. Hey Glenn

    Great article, I thoroughly enjoyed the detail you put into it.

    I’m fairly new to SEO and I am still developing a process for it, but it’s encouraging to see that amazing results are not instantaneous and that like anything worthwhile in life, they take time to bloom.

    1. Thanks for your comment, and I’m glad you read this before getting too deep into SEO. There are some important learnings from this case study that can help you as you build out your SEO strategies.

      And you’re right, SEO is a longer-term process (that’s for both building up strong search rankings, and in this case, coming back from getting hit by an algo update like Panda). I’m glad you enjoyed my post! Thanks again.

    1. Hi Jonah. Unfortunately, all I can say is that the site got hit in the fall of 2011. I hope the details and key learnings were helpful. There are some important lessons there. :)

  11. These observations suggest that engagement metrics such as bounce rates are a key factor in the algorithm however we’ve been told time and time again that this is not a factor by Google, am I missing something?

  12. Whoa! This is a very comprehensive article Glenn! I’m new in the SEO industry (coming from tourism) and is working for a small SEO company (www.digitalmoz.com). My co-workers keep talking about the Panda and the Penguin and I was like “what?!?”. This prompted me to do my research and so far, yours has been the most helpful one. Our client’s rank significantly dropped after the Panda and right now, we are currently working to keep them up in the ranks. Every day, we are learning in the office as to how to adapt to this update and can I just say, so far so good. We are still crawling but I am positive that we can bounce back. I mean, SEO companies learn to adjust in changes in algorithms almost as fast as they change right. So while I am still a newbie in the industry, this will will serve as my personal guide next time if there are new updates.

    1. Hey, thanks Cory. I appreciate it. I’m glad you found my case study helpful. My recommendation would be to go through the process I listed in the post. It’s going to take time to come back, but you need to do the right things. Understand that Panda gets rolled out periodically and track everything you can. My hope is that if you do everything right (like this company did), that you’ll come back too. Good luck, and thanks again.

  13. Excellent case study. How often do we takes our websites for granted. To be honest I too have been complacent at times with a couple of my websites. Its time to open my eyes and stop being complacent. I have already seen a fall in my website traffic all thanks to my complacent nature. Your post has indeed boosted me to work hard now. Thanks mate

  14. I find this to be one of the most easy to understand evaluations of panda I have seen in the last months. Thanks Glen and also congrats to SEJ’s new “Featured SEO Columnist” ;-)

    1. Thanks Andrew. I appreciate it, and I’m glad you found my post helpful. 6 months was a long time, but better late than never. We were very happy to see the site come back… :)

      Regarding becoming SEJ’s Featured SEO Columnist, thanks so much! I’m excited about the opportunity.

  15. Hi Glenn. It’s been 3 months now I was hit by Panda and Penguin, specially most by Panda. I’ve tried everything and in this cases, impatience is the worst enemy. I’m glad of reading this article. It gave me a little bit of boost to don’t lose the hope. Can I ask you something? What I’m doing now with all the thin and low quality pages which are about 15% or 20% of the whole site, is using noindex,follow while I’m improving and updating them, once this pages have better and quality content I’m going to take off the noindex tag. Do you think this a bad strategy? Should I leave them alive and crawlable while I’m improving them?

    Thanks for any advice :-J

    1. Hey, thanks for your comment. I’m sorry to hear you were hit by Panda and Penguin. That’s a tough situation, to say the least. But, I’m glad my post gave you some hope. If you are going to revamp the thin content, then you can keep it up and running. I would just move quickly to update that content…

      Personally, I wouldn’t noindex that content if you are going to update it. I would just move as quickly as possible to revamp the content, and nuke anything you won’t be updating. I hope that helps. Stop back and let me know how it goes. Good luck.

  16. Hey Glenn,

    Very nice story and i’m happy it worked out for you. We are also dealing with a similar problem but hopefully we can manage to pull through.

    However I have a question: what is the recommended number of words a good article should have ? I know that it is kind of relative but I would like to hear your thoughts on this.

    Thank you and keep up the good work!

  17. Can you tell me what is quality content in your opinion? (600 + words articles, picture(s), videos, audio, comments, external links, internal links….) Where is the line between content hit by Panda and great content ?
    Thank you

  18. Glenn, this case study is worth multiple views. I believe it is easy to comment on revival strategies (when everybody is doing it), but when someone explains it with a case study like this , it is more meaningful and useful. I will take a lesson from it and apply to my company.

  19. Good Article, what you didn’t mention were any of the pages duplicated around other websites, I’ve had a similar problem with a site of my own. Removing thin pages helps, but you need to remove scraper versions or syndicated version to achieve a full recovery.