Recovering from Google’s Panda & Penguin – A Case Study

Flexiscreens.com is an Australia business based in Tasmania, producing flexible insect screens from premium-grade materials and delivering them world-wide. They’ve been plagiarized and targeted by copycat competitors but have always enjoyed good rankings and resoundingly good client testimonials due to exemplary products and services. The past year has been a roller coaster ride. First they were rewarded by Google, and then they were punished.

• Monthly “Visits” grew rapidly from 12,800 visits in Aug 2011 to 36,000 in Nov 2011.

• From Dec 2011 traffic declined to 17,000 visitors by May 2012.

It was the same content, basically unchanged in that time and its initial wild variations seem to predate the Panda release. In contrast, over the year, the primary copycat competitor now ranks higher than they do, despite having inferior products made from lower-grade components.

There are some peculiar aspects:

• Nov 2011 – at the peak of visitor traffic, 54% of visits were from Australia!

• By May 2012, that had dropped to 17% of visits from Australia!

The website has been hosted in Melbourne, Australia since early 2011. The site appeared in “Search: pages from Australia” so was clearly associated with the Australian dataset. The downward variation in Australian visitors was inexplicable. In June we launched into a major revision of the site, tackling each and every aspect that could conceivably be harming the site’s rankings.

Eliminating Low Quality Content

In looking at the site from a critical perspective, it appeared that the 89 glowing client references might have been an Achilles heel. These were individual posts in a Testimonials category, with a widget that selected one at random and displayed it in the page sidebar.

However, bereft of unique Titles and Descriptions, lacking headings, and with only a couple of sentences and no images, they would certainly appear of very low quality and minimal value!

By way of remedial action, these posts were inserted into a Testimonials Manager plugin, retaining the random Testimonial display in the sidebar, but with a single consolidated page instead of 89 individual posts. 301 Redirection was put in place on those posts, to the new Testimonials page.

Doorway Pages

In addition, there were a dozen or so Authorized Distributor pages, half of which contained nothing other than a State or Country variation and with a standard layout/content format. This may have appeared suspiciously like Doorway Pages, specifically referenced in Google’s Webmaster Guidelines.

These were all combined into a single distributors page, with an index hyperlinked to State & Country sections, and 301 Redirection applied.

Impact of Initial Remedial Action on SERPs Results

There was a prompt response on the rankings for multiple keyword search phrases:

• flyscreen distributors – jumped up +19 places within a week to #5 on Google.com.au

• flyscreen distributors – jumped up +55 places within a week to #10 on Google.com

• fly screen distributors – jumped up +3 places within a week to #9 on Google.com

The consolidated Distributors page with its larger and unambiguous content definitely gained rapid traction in the SERPs! Unfortunately, since then those gains have been offset by further ranking losses across the board, but traffic volumes have remained relatively steady over the last few months.

Aside from the first round of remedial action on the posts and pages, we’ve also systematically reviewed page content to sharpen the focus, reduce vagueness, and eliminate other potential problems.

Google Says Be Careful about Links

Any external links were converted to NoFollow to prevent loss of link juice, or any suspicion of involvement in link spam activities.

Google Says Avoid Duplicate Content

Tags: all “tags” removed because these were not used correctly, and in the main applied to the 89 Testimonials posts. These tags consisted mostly of Australian state and/or international country, and the window type the reference related to. The consequence of that was dozens of almost identical Tag pages, all bereft of Titles and Descriptions.

Extraneous / Boilerplate Text

At some point not too long ago, the owners had inserted 3 lines of H2 and H3 headings into the top of EVERY page! That was what amounted to a Unique Selling Proposition and a Call to Action, but of course it severely diffused the effectiveness of the real first headings & first paragraphs across every page on the site. All iterations of extraneous text were replaced with an image on the bottom of each page, with a relevant link title & image alt text.

Further vague first paragraph content occurred in the About Us page which had been topped with some “warm fuzzy” content that had little to do with their Products / Manufacturing / Distribution. That extraneous content was displaced downward by something a little more useful.

Google Does Not Like Low Quality Pages

The informational pages needed to be there, but contributed little to defining the site’s genre, products or services. All non-important pages were set to NoIndex using the “Ultimate Noindex” WordPress plugin, including Login, Admin, Logout, Author, Search, Privacy, Terms etc.

• Gallery: this comprised three pages of various insect screen formats, but there was zero explanatory text. A picture may well be worth a thousand words in some circumstances, but not to a search engine! A preliminary paragraph of explanatory text was added, and the owners are preparing more to boost the intrinsic value of these pages.

• Headings: additional supportive sub-headings in H2 or H3 added on pages as appropriate.

• Images: all illustrative images had relevant explanatory Alt text inserted.

• Text Emphasis: A significant amount of non-relevant BOLD text was reverted to Regular to prevent emphasis confusion. As a general rule, Bold should only emphasize important page content, not entire paragraphs!

• Header Tags: All Titles, Descriptions & Keywords were revised to be within recommended length parameters, and made relevant to the specific page/s.

The Google Rollercoaster

Those who have engaged in close-quarter combat with Google’s relentless algorithms will understand that during the past 6 months ranking charts look like an 8.5 earthquake graph. Between Google’s tweaking, and our own remedial efforts, it was sometimes hard to ascertain if what we were doing with the website was making things better or worse! However, with the objective of “good content” firmly grasped, we persevered.

By and large, the efforts that were taken resulted in significant Google.com.au ranking improvements, sustained as Google’s algorithms have stabilized.

Local vs Global Search

For US-based readers, this is an issue that many SEO’s won’t encounter. However, if you are located elsewhere on the planet, it seems Google has drastically altered the way in which it rewards websites within their local market, at the cost of their global prominence.

Referring back to the “visitors from Australia” issue mentioned previously – as things have settled out, Google HQ, rankings across Google.com.au have improved to 2011 levels. The volume and percentage of visitors from Australia has increased back to ‘normal’ levels. Sadly, as global rankings have declined, the diversity of international visitors has faded.

Prior to 2012, Flexiscreens drew many of their clients from around the world due to good global rankings. In the past year, monthly visitor traffic (Webaliser stats) has declined from the 2011 high of 36,000 to a low of 8,700. October 2012 shows a modest improvement to 10,600 visitors.

In all other respects, rankings have clearly responded to ministrations on the website aimed at meeting Google’s Quality Guidelines. This is a family business, totally dependent on remaining prominent in global SERPs, and its fall from grace has been sudden, and painfully expensive. Recently, the company decided there was no option but to resort to Google AdWords to gain some additional exposure in targeted international markets.

The most striking aspects in the year of Panda and Penguin have been:

• the inexplicable downturn to 17% of visitors from Australia by June

• the equally inexplicable upturn in Australian visits thereafter, to 54% by October

• the total loss of international rankings

The more cynical amongst us might consider this is a Google “carrot and stick” strategy, giving local business better results in local markets, but beating those who need international exposure into the AdWords corral.

The New Normal

There does seem to be an across-the-board improvement in search relevancy. Google seems to have worked hard on ensuring that the search engine results pages give prominence to the source site of a business name or brand. As an example, a search for a Bed & Breakfast or hotel by name is now far more likely to show the specified business in first place. Previously, it would have been buried under a pile of B&B and hotel booking sites, all creaming commissions off the legitimate business owner / operator.

For that we can be grateful and I guess we can accept that we are now living in the ‘new normal.’ It’s not perfect but overall, it’s getting better!

The Bottom Line

If your business website was badly affected by Google’s Panda and Penguin updates, it is possible to recover fully from the impact. In fact, it’s usually a straightforward process of improving the website’s content, structure and functionality. Results of such improvements are invariably positive, rapid and enduring.

Ben Kemp, a search engine optimization consultant since 1997, is a specialist in website redesign, and a veteran of 25-plus years of experience in the IT industry.

Web: www.ComAuth.co.nz + www.Website-Redesign-Company.co

About the author


Ben Kemp

Ben Kemp, a search engine optimization consultant since 1997, is a specialist in website redesign, and a veteran of 25-plus years of experience in the IT industry.
Web: www.ComAuth.co.nz + www.Website-Redesign-Company.co Email: support@Website-Redesign-Company.co


Click here to post a comment
  • Wow so that where all our ranking went. we still have a pr6 but totally lost ranking in the serps!
    we dominated and the loss of tags is irevelant now, but google is still using tags in the serps, as ive been editing the site for the last two days and rankings are all over the place still and tags are still getting indexed. makes no sence. Thanks for your post as its critical.

  • Matt Cutts said long ago that Google does not want websites to rank in multiple countries, “if they gave users a country select box they would just select them all”. The best suggestion was to own multiple top level domains co.uk .us .fr etc…

    Google also wants to make money from adwords so it makes that info unclear.

    Recovery is complicated because Google is scared of giving its inside info out to manipulators, however they are first to work it out leaving us genuine business owners stuck. All I have to say to that is, Google you are a bunch of idiots for taking that view.

  • “Any external links were converted to NoFollow to prevent loss of link juice, or any suspicion of involvement in link spam activities.”

    Since 2009 a nofollow will NOT help you losing pagerank. And a Nofollow will actually be negative for external links doe to pagerank.

    So it would be smarter, to only nofollow those links you dont whant be affiliated with.

  • I’m currently seeing this right now. My retail outlet is intended for the UK market, but over the last month, although traffic in general has increased, UK traffic has become a smaller % of it.

    I don’t want to be promoting to Australia (today I’ve had the same number of organic hits from Australia as the UK) but I can’t find a solution to this.

    There are no real opportunities for me to add UK locations to the site to bolster regional traffic.

    I am at a loss. It seems a little pointless to have the ability in Webmaster Tools to assign a region or country if they are going to completely ignore that request. Typical Google though, only thinking with half their brain.

  • Happy to see a recovery story from Panda/Penguin. We were hit by Panda in March 2012 after 13 years of working hard to create good content and not engaging in any black hat seo. We have had some experts look at the site and could not really see anything that would cause a fall of 90% of Google traffic. We have gained back about 20% of Google as of December 2012.

    Your point about the testimonies was interesting, we wonder if ours is too much content in the left and right sidebar of each page that is the same although I have been told that Google can tell what is header/footers and sidebars and should not affect the ranking.

    Sounds like even this site is not up to where it once was, even though it sounds like a very quality site.

  • I am building a website myself. Thanks for posting this now I know what mistakes I can’t make so I don’t get slapped by Panda and Penguin. Great post.

  • Brilliant Article Ben. I’ve been wrestling with some of these issues myself so it’s helpful to read this.

    One has to wonder if its worth bothering with WordPress tags as the damage they do if setup incorrectly seems so bad.

    You might want to think about putting together an ebook on how to setup your WP site so that it conforms to Googles latest commandments and puts in place clean, effective techniques going forward. I would buy such an ebook.


  • Thank you for sharing your experience Ben, I appreciate this very much, as site owners and seo’s can only benefit from comparing notes with each other. However, coming from a scientific background, I much prefer studies where variables are isolated and studied one at a time. In that way the findings are more valid than if we do all different changes at once. Even if the effect is positive, we still would not know what did the trick and what contributed most. My question is: as you mentioned these changes that you have made, did you do them one at a time or all together and what % of changes of traffic have you observed from each change? With regards to traffic changes, which traffic was affected? Your article implies that it was the search traffic, but still would like to confirm. In my opinion, the types of traffic mean very different things, and that is the first thing we need to look at – which is the type of traffic that is reducing. If it is the search traffic, then I would like to ask what was happening with the impressions at the time when traffic was reducing? Were they staying the same number, or lower? Then after you made the changes and you saw the keywords gaining momentum, what was happening with the impressions? Were they going up as well or staying the same. I will be grateful if you share this info, as it will help me compare notes with my experience, and consequently draw some meaningful conclusions. I am happy to share my experience if you are interested. I have been doing similar things as the ones you describe, and have observed interesting events which don’t conventionally go hand in hand. Happy to share if anyone interested. Love,

  • One of the things to think about in the wake of the Google changes is that Google SEO should not be your only promotional strategy, it should be part of a bigger marketing mix which includes a broad range of advertising and PR tools.

    Relying solely on Google SEO puts your business’ success 100% in the hands of other people and that’s never a good strategy, hence the outcome we’re seeing in some of the comments here. If you have a broad promotional mix you will minimize the impact from any Google changes. (same goes for social media, get your contacts onto your opt in list as fast as you can, and own that customer…)

    Simon – InternetBusinessKickStart.Com

  • We constantly reviewing many aspects from panda and penguin updates from google. Our website has not affected any rankings. But many sites that are not updated so far and have some link juice practices are effected with these updates.

    revamping with new strategies really takes lot of time but need to make these changes to see the sites once again in the google for good SERP’s.

    Thanks for the post. I got clear picture, what to do after these updates.


  • I am currently working on recovering a site impacted by Panda in late Sept 2012. I am curious about your “Impact of remedial action on SERPs” and over what timeframe did you see these changes? Were you able to see any changes immediately (within 24 hours) or only after algorithm updates? Do you think this varies between Panda and Penguin penalties?

  • This year has been a particularly difficult one, with some of the most significant changes ever on the search landscape.

    Bilka: studies of variables are all very well when you have the luxury of time to test theories and measure results. That’s not the case when you have multiple old and new clients, all desperate for help because their business is evaporating in catastrophic fashion.

    Basically, you must reply on experience and instincts to interpret the new rules. You must urgently implement what your judgement determines to be the best remedial actions. On top of that, results must be delivered before those clients go out of business! 🙂

    Gavin: tags and categories can do you a great disservice if used incorrectly. My advice is, if you don’t know why you are using them – don’t! Or at least, use a mechanism to ensure that the pages they generate have NOINDEX applied.

    • For sure, intuition is always handy, but if you have to answer in one sentence – what actually helped – what would that answer be? You gave a good account of your experiments, but no clear answers as to what extent each of those endeavours worked.


    I’ve read this over a few times because I couldn’t see much in here that would help with Panda and it appears to me this site has NOT recovered from Panda. It says traffic in October 2012 was 10,600 visits but in November 2011 it was 36,000 visits. So where’s the recovery????

    The only things that stand out to me are removing boilerplate text and the fact that this is an ecommerce site but it has lots of external links. Why would a site selling something have lots of external links??

    I don’t think they’ve found the real cause of the problem yet. I strongly advise that this article is taken purely as a story of what they have tried, not as a ‘how to solve Panda’ story, which is the impression it gives.

    • Agreed! All too often there are articles like this that have a misleading title. It is a case study of what they have tried so far.

    • Hi Panda Expert,
      that was my point as well, Thank you Ben for sharing your story, but what is the conclusion and how accurate is that conclusion. I prefer to address just one issue and see what happens, then another one and another one, rather than doing everything at the same time. How would we know what helps and what doesn’t and to what extent… Panda Expert – I will be grateful if you could advise me how my ecommerce site stands in relation to Panda.

    • Did you submit a Reconsideration requests once you were done or did you just wait? Did google email you back? Please give more details about the whole process. Thanks

  • I have seen a few sites recover from thee Google updates. But it takes a LOT of time and effort. Unfortunately I am unable to see a confirmed and guaranteed process of getting Google Love back.

  • […] Recovering from Google’s Panda & Penguin – A Case Study […]

  • Thanks for the above stated points, I think there should be some other points to consider,including Blog bounce rate,less number of ads,Social media impact….

  • There is no shortcut to recover from Google panda. All you can do is write unique content, Titles and remove old and useless content from your blog.

  • Thanks for the post. We are quite aware of the panda and penguin updates from google. Our website has not been affected, but I have a clearer picture now. Will wait for the next post.