Google originally unleashed their pet Panda 12 months ago on 24th February 2011 and it has caused a whole range of carnage since then, with around 10 iterations in the past year alone.
In a way it has been a blessing in disguise, well maybe ‘blessing’ is a bit too much of a strong word, but it has certainly made webmasters wake up about what type of content needs to be on a website these days to rank well.
This post is going to be quite an epic, but if you are only interested in certain bits then here is what is going to be covered;
- What is the Google Panda algorithm
- How to know if you have been Pandalised
- How the Panda algorithm works
- Panda timeline
- Panda winners and losers
- Factors that could be used by Panda
- How to recover from Panda
- Quotes from other SEO experts
- Panda resources
What is the Google Panda algorithm?
When Google crawls the web, it basically scrapes every single website it can find and stores that in a massive index. When each page is crawled it is run through a sophisticated algorithm that contains over 200 ranking signals which determines where the page should rank within the search results for a specific query.
Well the Panda algorithm sits on top of this normal ranking algorithm and is designed to filter the wheat from the chaff so to speak. It’s aim is to punish low quality websites and promote websites with high quality unique content.
In the past a lot of poor quality websites used to scrape content from other websites and rank higher than them. Whilst this still happens on occasions, the Panda algorithm is aiming at promoting the original sources of information. When a website has been penalised by the update it is often referred to as Pandalised.
Some sources claim that the Panda algorithm is a machine learning algorithm which can detect the differences between high quality websites and low quality websites, although I am not so sure it works like that. In the past Google has been strictly against machine learning and instead using static algorithms developed from tests run internally. Whether it is a machine learning algorithm or not doesn’t really matter though.
Is Panda really ‘better for the user’ ? Well according to some independent research, yes.
“The results show that Google and Microsoft have won a major victory in the fight against such content farms.
To test how successful Google and Microsoft’s Bing have been at fending off content-farmed results, McCreadie ran 50 search queries known to be a target of content farmers, such as “how to train for a marathon”, in March and August this year. Then he paid people to examine the results for links to low-quality sites, where “low quality” was defined as uninformative sites whose primary function appears to be displaying adverts.
The results are striking. In the case of the marathon query, sites that contained lists of generic tips, such as “invest in a good pair of running shoes”, were present in the top 10 in March but had disappeared by August, while high-quality sources, such as Runner’s World magazine, now appear near the top. Similar trends were found throughout the 50 queries.“
What is it called the Panda update?
Because that is the Google engineers surname who developed the Panda algorithm. There are actually three people within Google who have the surname of Panda but the particular person who developed the algorithm was Navneet Panda.
Information from Google about the Panda update
Google have been kind enough to provide some reasonable information to webmasters about Panda and generally their ranking algorithms. One of the quotes from a Google webmaster central blog post was about how….
“Low-quality content on some parts of a website can impact the whole site’s rankings“
Quite telling really, if you have got a lot of crap on your website then best wake up and smell the coffee. Google (and users!) are looking for high quality information to rank well in the search results. If you have a lot of poor quality content on your site then best take another serious look at it and decide if you are going to (a) improve it or (b) bin it completely.
Another interesting quote from the same Google blog post…
“The “Panda” algorithm change has improved rankings for a large number of high-quality websites“
Or if you want to read that the other way, the Panda algorithm change has decreased ranking for a large number of low-quality websites. Panda is all about getting rid of crap from the search results, crap being websites who scrape content from others and websites who have lots of very low quality content about ‘how to tie your shoe laces’ (unless it is a very good website about that as we will see later in the post).
Google also offered 23 questions to help webmasters evaluate their websites in the same light as how Google tries to do so. The questions below aren’t specific things you can do but instead are more looking at the holistic approach of your website / online business. Have a good think about this information and answer honestly. Then do the same for the people now ranking at number 1 (i.e. your competitors) and ask yourself if you really deserve to rank higher than them?
- Would you trust the information presented in this article?
- Is this article written by an expert or enthusiast who knows the topic well, or is it more shallow in nature?
- Does the site have duplicate, overlapping, or redundant articles on the same or similar topics with slightly different keyword variations?
- Would you be comfortable giving your credit card information to this site?
- Does this article have spelling, stylistic, or factual errors?
- Are the topics driven by genuine interests of readers of the site, or does the site generate content by attempting to guess what might rank well in search engines?
- Does the article provide original content or information, original reporting, original research, or original analysis?
- Does the page provide substantial value when compared to other pages in search results?
- How much quality control is done on content?
- Does the article describe both sides of a story?
- Is the site a recognized authority on its topic?
- Is the content mass-produced by or outsourced to a large number of creators, or spread across a large network of sites, so that individual pages or sites don’t get as much attention or care?
- Was the article edited well, or does it appear sloppy or hastily produced?
- For a health related query, would you trust information from this site?
- Would you recognize this site as an authoritative source when mentioned by name?
- Does this article provide a complete or comprehensive description of the topic?
- Does this article contain insightful analysis or interesting information that is beyond obvious?
- Is this the sort of page you’d want to bookmark, share with a friend, or recommend?
- Does this article have an excessive amount of ads that distract from or interfere with the main content?
- Would you expect to see this article in a printed magazine, encyclopedia or book?
- Are the articles short, unsubstantial, or otherwise lacking in helpful specifics?
- Are the pages produced with great care and attention to detail vs. less attention to detail?
- Would users complain when they see pages from this site?
Wired also interviewed Amit Singhal and Matt Cutts regarding the Panda update which talks about how Google is trying to filter the low quality websites from the high quality websites.
How to know if you have been Pandalised
Quite simply really, keep a close eye on your rankings and organic traffic in Google Analytics. If you have been hit by Panda then you would expect to see a large steep drop in both rankings and traffic on one of the key Panda release dates listed further down in the post.
Below is an image of what you would see for a Pandalised site
Or if you see that you suddenly had a massive drop in traffic or rankings then suddenly recovered as shown in the image below, then you could deduce that you have managed to recover from panda. This may not always be the case as you may have accidentally noindexed your entire site, but we will assume you haven’t done that!
One thing to be aware of is that due to Panda being an iterative update, then any drops or subsequent recoveries will only happen on the date of an update. Not every update will likely be announced by Google, but most have been so far. So be wary of anyone who claims to have recovered from Panda not on one of the official Panda updates.
In addition, when a website is Pandalised or recovered then it would only be shown as a steep change in traffic/rankings. Any website which has slowly began to increase rankings and traffic will likely have nothing to do with Panda but more just generally improving their traffic / rankings based on the normal algorithm.
How the Panda algorithm works
Unlike the normal Google algorithm which is updated each time Google crawls your page, the Panda algorithm runs in iterative updates when the ‘data is refreshed’. Previously it wasn’t clear exactly how the algorithm worked, but it was recently clarified in a statement from Google;
“We’re continuing to iterate on our Panda algorithm as part of our commitment to returning high-quality sites to Google users.”
Within the blog post mentioned above Barry Schwartz mentions;
“It was a minor update, with no new ranking signals, but rather just a data refresh of the Panda update.”
So what does it matter how the Panda algorithm runs? Well it matters because for a website to recover from Panda it has to wait until the next update or data refresh. This can be a little annoying since the panda updates can be up to two months apart so and changes you make between now and the next update you will not know exactly what has made your site recover since you will have made lots of changes over that period.
Search Engine Round Table and Search Engine Land have been doing a fantastic job keeping up to date with the Panda iterations so feel free to have a look at their sources of information in their blog posts.
Below is a useful timeline which will be updated as further iterations are announced. There is no official numbering system with the updates and various sources contradict the numbering system, so best to focus on the key dates.
- 24th Feb 2011 – Panda 1.0 – Link
- 11th April 2011 – Panda 2.0 – Link
- 10th May 2011 – Panda 2.1 – Link
- 16th June 2011 – Panda 2.2 – Link
- 23rd July 2011 – Panda 2.3 – Link
- 12th August 2011 – Panda 2.4 – Link
- 30th September 2011 – Panda 2.5 – Link
- 14th October 2011 – Panda 2.5.1 – Link
- 18th November 2011 – Panda 2.5.2 – Link
- 19th January 2012 – Panda 3.2 – Link – Link
- 15th February 2012 – Panda 3.3 – Link - Link
- 23rd March 2012 – Panda 3.4 – Link
- 19th April 2012 – Panda 3.5 – Link
- 27th April 2012 – Panda 3.6 – Link
- 9th June 2012 – Panda 3.7 – Link
- 25th June 2012 – Panda 3.8 – Link
Panda Winners and Losers
The information below showcasing top winners and worst losers from Panda has been compiled from various sources, one of which is the data Searchmetrics released.
I am not going to comment on the winners and losers as it is clear just looking at the domain names why certain websites have benefited and why others have lost out, common sense can spot that straight away. But feel free to do some further analysis into these different websites to see if you come to the same conclusions as me, one group has fantastic content, the other doesn’t.
Factors that could be used by Panda
Well this section could span the whole of the Google’s 200 ranking algorithm signals although this would turn into a longer post than it already is, so here is some of the summary information instead. Below is a list of possible factors which could be causing a Panda penalty, note that there is no one simple fix, there is no silver bullet to a Panda recovery. The road to recovery is painful, long and slow but this post is designed to point you in the right direction with a clear Panda recovery action plan.
The factors listed below are ones that I have spotted lots of different people mention from various sources and I don’t necessarily agree with them all, but they are listed below for reference;
- Promoting brands higher
- Penalising keyword rich domain
- Over optimisation of keywords on a page
- Low quality / quantity back links
- Low quality content
- Low quality content on part of your site
- Poor user metrics (e.g. high bounce rate, low time spent on page)
- Site technical issues
- Websites which use templates
This is by no means an extensive list, to be honest it is a bit of a lame list but the reason behind that is that you will find someone, somewhere, saying anything is a factor in the Panda algorithm and honestly, no-one really knows what the ‘one-silver-bullet-factor’ actually is.
How to recover from Google Panda
This section is going to be quite long so it is going to be broken down into key areas about how to recover from Google Panda.
People who have recovered from Panda
There are lots of stories around the web and particularly on Webmaster World about people who have recovered from the update if you are interested in having a good read. Unfortunately there is no real data which can be analysed about what has happened from people who have recovered, which is why I can take on board what they are saying but with no hard evidence it is very difficult to take it as solid fact.
That said, what is common ground from people who have recovered is that they don’t really know specifically what they have done to recover. With Panda being an iterative update there were lots of changes made by each of the different people who recovered which may have helped them recover. All of these different items have been listed earlier in the post showing ‘factors that could be used by Panda‘.
On the other hand, it could simply be that the sites recovered because the Panda algorithm was tweaked so their site no longer fell into the category of ‘low quality’.
There is no way of knowing for sure if what people did actually made the difference or if it was a simply algorithm change which caused them to recover, but never the less I have listed a Panda recovery action plan which will be useful in helping your site on the road to recovery.
One of the most notable websites which claims to have recovered from the death grip of Panda is Hubpages and they did so by splitting their site between sub-domains for each author opposed to having these as directories. Their traffic originally dropped by around 50% according to the Wall street Journal. According to the article, Matt Cutts responded to an email from Hubpages advising them to try using sub-domains for some of their content.
“HubPages subdomain testing began in late June and already has shown positive results. Edmondson’s own articles on HubPages, which saw a 50% drop in page views after Google’s Panda updates, have returned to pre-Panda levels”
There has even been different people who have tested a lot of domains for Panda related fixes and concluded that it takes a lot more effort to get out of Panda than the fall into the traps. So best get to work and at a fast pace.
Action plan to recover from Panda
There are a lot of factors that could be causing your website to be deemed as low quality and honestly there is nothing magical about how to recover, just do more great SEO. Instead of trying to trick Google into ranking your site well based on the latest ‘technique’, instead create a fantastic website and instead of trying to emulate the signals that Google deems as quality then the quality signals may just be there for real. Simple!
If you are looking for the magic bullet on how to recover from Panda then I suggest you stop reading now as I am not going to be offering any magic beans here. Just gold old solid SEO advice to grow your website in the long run.
Make sure your site isn’t broke
Sounds obvious, but Google provides a massive list of errors to fix within Google Webmaster Tools so go and get these fixed to begin with. Make your site technically SEO sound. If this doesn’t fix Panda (which is unlikely on its own) then at least you have made your site better in the process.
Fix any 404 errors, 500 errors, 503 errors, redirect loops, broken back links etc. Get everything fixed you can find, but make sure at the same time you are focusing on the main issues behind Panda which is about providing high quality websites for searchers.
Improve user metrics on your site
This should be being done as part of your conversion strategy anyway, but if you aren’t already, then look at ways to improve. There are various tools out there to help with feedback about your website including ClickTale, Kampyle, Kiss Metrics, CrazyEgg and likely many more.
Find out how customers are behaving on your website then find ways to improve. A happier customer is more likely to tell their friends about your website and build those natural links.
Why not set up a free survey with Survey Monkey? Ask your newsletter subscribers what they really think about the site and how it could be improved.
Remove low quality content
Panda doesn’t like low quality content, so get rid of it. Some people say you only need to ‘noindex’ the content where as others say that you need to fully get rid of the content by 301′ing the content or 404′ing it. There is no one simple fix here, personally I am in favor of simply noindexing low quality content or better still just improve the content. If the content doesn’t really serve any purpose then just bin it.
One thing I would say here is that if you have content which you honestly believe is good and valuable then don’t go binning half of your website. Understand how Google is trying to judge websites and see how it can be improved. I have seen pages getting traffic when the Google Keyword Tool was telling me there was zero people searching for the phrases each month, so should these pages get binned? If they are driving quality traffic to your website then I would keep them and put them on the ‘improvement list’.
Create high quality content
Where does your content come from at the minute? Scraped? Duplicated? XML Feed? Affiliate sites? Or is it just low quality human created content or poor quality user generated information? Still surprised you got hit by Panda?
Re-read the list of websites who have benefited from Panda then re-read again. Study those websites. Why are they successful when others aren’t? It is blindingly obvious that the people who have benefited from Panda are those who have fantastic content. If you provide the best information (or anything for that matter) then you fall into the category of a ‘brand‘ by default.
Think if you and a handful of competitors were selling cheap nasty products in a specific niche where there is demand for this. Who is seen to be the best in that market (and the best for a certain search query)? The company who is doing better than everyone else.
If you don’t provide the highest quality content and information on your website for the specific keyword(s) you are trying to rank for, then do you really deserve to rank number one? Like wise, if you do provide the highest quality content and information on your website for that specific keyword(s) you are trying to rank for, then do you really deserve to rank anywhere but number one?
Instead of having low quality pages which are very similar, for example ‘how to skin a cat’ and ’10 ways to skin a cat’ why not create one great piece of content on the subject instead? Aim to remove / combine your lower quality pages of your website into something much better. If there is a valid need to keep things separately then do so and look at improving the quality of those pages in the process.
Have a watch of this quick 2 minutes video about how good websites that don’t know about SEO can still rank well, it gives some insight into what Google are looking for (if you really need any further evidence)
What is high quality content?
Lets take a few examples of what is good content VS what is bad content. Firstly lets take the search term ‘how to tie your shoe laces’ and here it what is obviously a very poor website for that query
Now I think anyone can see why this page is totally unsuitable for the searchers query. The rest of the page above is just full of adverts and useless information and navigation, the above is the only real content on the page, I have just cut out the rest of the jump from the page so it is clear what is on the page. The person responding to the question is struggling to speak in full words and sentences. Would I really trust this person providing the information? I think I would rather have a go myself than follow this persons advice! (and yes, I can already tie my shoe laces before anyone asks! )
If I couldn’t tie my shoelaces then here is what I would look for in high quality content (click on the image to open the full size view to read the comments)
Here is what one of the ‘knot’ pages looks like when you click through (click on the image to open the full size view to read the comments)
A couple of things that would make this page even better for content would be including some more videos and then add some more social sections to the page such as user comments etc. Whilst the user can feedback in terms of voting it would be useful for people to leave some real textual (or even images and videos their self) feedback.
Can you clearly see the difference here? Now is it clear why Ian’s website ranks #1 for how to tie your shoe laces?
Great content brings long tail traffic which gets users involved in your site, and because you have great content (as you are now on with creating this to recover from Panda, aren’t you!) they will share it socially and build some natural links for you which ultimately helps with your rankings. It all starts with great content!
Great content requires both quantitative and qualitative information, research and a knowledgeable person to knit it all together. Employ a content manager for your site and give them a budget and resources to put the content live as quickly as possible. Then promote this via social media to increase the impact.
Make your website the source of information people quote. Stop spinning content from other sources into your own unique content for search engines. Instead go and get some information that no-one else has. This is the true way to beat Panda.
Just add loads and loads of great high quality content!
Engage with your websites customers
Google announced an algorithm update back in November 2011 which was all about providing fresh content to searchers. How can you provide fresh content to your users? Well engaging with customers on your website can be a great way, include a commenting section on your product pages, encourage reviews of your products or service, and how about going the whole length and creating a nice community and high quality user generated content platform for your customers. People love to engage so let them!
Build high quality links to your site
Some people claim that Panda is a links based algorithm, some people disagree. Until Google officially announces either way then no-one will know. So carry on building some great high quality links to your website. Look at building links for people and not for search engines.
With no-one knowing if Panda is a links based algorithm or not then it would be fair to say that if you have a lot of low quality links from poor quality websites (such as press release websites or article websites) then if their rankings have suffered (which they will have done as they are ‘low quality’) then this is bound to have a knock on effect to your website.
If you aim to drive traffic to your website from the links you are building then you are on to a winner. High quality and relevant links.
Add real value!
The video below was released by Google on 19th March 2012 and is an absolute must watch if you want to recover from Google Panda.
As you can see from the video, Google recommends the following…
Avoid these common SEO mistakes;
- Having no value proposition – If you don’t offer anything unique as a business then why do you exist?
- Segmented approach to SEO within large organisations – Make sure the left hand knows what the right is doing and everyone within the company is working together towards a common goal. Results will be much greater.
- Time-consuming workarounds – This is basically talking about stop trying to ‘trick’ the search engines and instead focus on best practices.
- Don’t get caught in SEO trends – I.e. Same point as above really, focus on what is going to bring long lasting visitors to your website.
- Slow iterations – In other words, speed up your development process and be the first to get new technologies out of the door. If you can speed up your ‘flight time’ from the initial idea to launching then you will stay ahead of your competitors and gain a competitive advantage.
Google also list a further six fundamental SEO tips of things to do;
- Do something cool! – Says it all really!!
- Include relevant keywords in your copy – Whilst they do say not to focus on keyword density, you still need to include relevant keywords within your content for which you want to rank.
- Be smart about your tags and site architecture – Optimise your title tags and meta descriptions to increase clicks to your site.
- Sign up for email forwarding in Google Webmaster Tools – In other words, keep up to date with the messages Google is sending you.
- Attract buzz – Natural links, +1s, likes, follows and any other social signals. Be interesting, surprising or entertaining as a business and this will happen naturally.
- Stay fresh and relevant – Are you present on social media? Do you have a mobile website? Are you keeping your content up to date?
How not to recover from Panda
I am going to be as polite as possible here, but Panda is all about providing high quality websites to users. Some websites have unfortunately got caught up in the low quality trap which makes their recovery a little more challenging than others, but either way the advice outlined above still stands.
For websites who have seriously suffered from the Panda update, well it’s quite simple, your website has suffered because Google deems your websites as shit (for whatever reason). Cold hard fact.
You are not going to recover from Panda by re-organising your website into a different shape, you still have a shit website. Tweaking a H1, meta title, meta description, restructuring your website or lots of other small pointless changes aren’t going to make a difference. Don’t get me wrong here, if all those things need improving or fixing anyway then get them improved, but don’t expect that if these aren’t too bad in the first place that by doing all those changes that Google is going to automatically bring your site back.
Above is an example of what you are ultimately trying to do, trying to make a square from the same round pieces. It doesn’t work like that! Instead think of the bigger picture, if you want to build a square then you need to add some bits to the edges to make it bigger (and in the right shape).
Another blindingly obvious thing on what not to focus on when fixing Panda issues is trying to increase user metrics without actually changing much on the page. Instead make big bold changes to see some significant results, add much better content and generally make your website far better and easier to use. Don’t just try and manipulate ‘user metrics’ to game Google’s Panda algorithm (if user metrics are being used, which no-one knows for certain).
If you actually make your website better then your user metrics are going to increase by default.
Affiliate websites in a post Panda world
Some affiliate websites have been hit quite hard in the Panda update, mainly because the affected sites were set up purely to generate affiliate commissions and not actually add any value for the end user. The action plan for affiliates is really the same as any website that has been hit by Panda and that is to ask yourself why you, more importantly your website, actually exists.
Google doesn’t hate affiliates, it hates affiliates who don’t add any value for the user. Think about where you can add value for your customers and focus on that. If you don’t offer anything extra over the people who you are partnering with then why would Google rank your website high in the search results when they could just rank the main website at the top instead?
Affiliate marketing used to be as simple as getting a keyword rich domain within your niche, finding all of the related keywords in your niche from Google Keyword Tool and creating a bunch of pages with low quality keyword rich content of approximately 500 words then build a few back links in your niche and get reciprocal links or one way links if you were better. Not any longer though. More about adding real value.
Quotes from other people in the SEO community
All of this post so far is my conclusion/opinion based on what I have read about Panda over the past 12 months. Lets see what other SEO experts in the community have to say on the matter. I asked a few others in the SEO about how they would beat Panda and here is what they said….
Question: What are the three key areas you would focus on to beat Panda?
Quote from Eric Enge
Stone Temple – Holistic Internet Marketing Optimization
Identify weak pages and NoIndex them or kill them entirely. This is one of the factors that nailed a lot of sites. Those low value pages that you put up just to garner search traffic? Time for them to go. We led one site out of the Panda morass through this change all by itself.
Check your user engagement vs. competition. You can do this using services like Compete.com. It won’t show you bounce rate, but it will show you other metrics such as time on site, page views per visitor, and the percentage of your visitors that are there for the first time. These metrics can help you see if users are more engaged with your competitor’s sites than yours.
Work on making yours better than that of the competition. Your pocketbook will thank you for it even if it is not your Panda issue.
Cut back on the advertising. Some sites that got his were heavy on the advertising. While I don’t think this was a direct signal in Panda (that was something that Google made a ranking factor much later), it can still drive reduced engagement, and we have seen sites where this was the issue.
You asked for 3? So much for my listening skills. Look at the balance of your inbound link signals. Have all your links from one type of source? Bad idea. Don’t limit this to links either. Have 10,000 great links and no one is talking about you on the web? Also a bad idea. A site with 6,000 similar quality links and active social conversations will kick your site to the curb every time.“
Quote from Steve Overton
The Google Panda update principally targeted those websites that in Google’s algorithms opinion offered little or no value… in other words those that had thin or duplicate content, were perceived as spammy with too many adverts above the fold and had non authoritative links.
Quote from Distilled
Paddy Moogan and Dave Sottimano at Distilled
Get a second opinion on what people think of your site. Panda was based on human quality rater questions that you can easily replicate using Mechanical Turk surveys. If enough respondents say they wouldn’t buy from your site, you might be looking at a complete re-design. Will Critchlow pioneered this method and shows how to do it here: http://www.seomoz.org/blog/replicate-googles-panda-questionnaire-whiteboard-friday ”
Quote from Felix Lueneberger
Felix Lueneberger from LateRooms.com
“Escaping Panda is like transforming a fallen apart ruin into a shiny new property. Tear down, rebuild, expand:
Quote from Caroline Bell
Content: Remove, consolidate or no index thin content pages created purely to chase the long tail. Ensure that the content on remaining pages is totally unique and relevant. Focus on engagement. Pages with high bounce rates can be an indicator of low quality.
Review your content syndication tactics. If you are distributing every piece of content produced for your site you are asking for trouble. Much better to keep the majority of content on your site unique and produce different content to syndicate to third parties.
Links: Conduct a review of your link building campaign starting with an analysis of competitors in your space that rank top 3. Build links that emulate their link landscape. This does not mean that you should get links from every source that they have, more that you should aim to get those ‘type’ of links. This includes site type, anchor text mix and PR spread.
Focus on quality over quantity by acquiring links on sites with an engaged audience and evidence of social media sharing. In addition, you could look to remove any poor quality links that you may have built in the past. Hundreds or thousands of site wide links from inferior blogs can affect the overall quality of your back link profile.
Social: Ensure that your site has brand pages on Facebook, Twitter, Linked in and Google +. This may seem obvious but it’s amazing how many sites don’t have. To be seen as a brand you need to act like a brand! Make sure you are making it as easy as possible for visitors to your site to share your content by providing prominent sharing buttons.
To be honest, every site should be employing the above tactics regardless of whether they have been pandalised or not!“
Quote from David Addison
David Addison from Dirigo Design & Development
Begin to fixate about the end-user. This is your North Star. It will guide you to recovery. Google wants to provide the most value to the end user. At the end of the day, Google only wins by serving the most useful content. Google’s survival is at stake here. Panda is a game changer. The old PR search patent went nonexclusive last year. Google is searching for a new/better model. One that cannot be gamed. The painting Emerald Sea by Albert Bierstadt sums it up in a visual.This is Google unless they can break into the social space and provide the best search results.
Poor quality sites are going to be expunged from the engines. Panda is updated periodically and behaves like a penalty. Get outside help to identify blind spots. I am no FARMER. But I certainly farmed (with small letters). I learned how to make big money by publishing really good content. Over an eight year period I let analytics drive my content schedule.
Without even noticing it, I generated a huge mass of semi-duplicative content. Okay, it was duplicative! Not what I would have termed duplicative content a year ago. And it was not low-end content. This stuff cost several thousand dollars a page. It was journal researched and written from scratch. Nonetheless, duplicative by the new definition. Kill it. Noindex it. Just get rid of the duplicate stuff. You have nothing to lose. I unlinked hundreds of pages with a production cost of $500k. Dump it. You can always add it back later after retooling it later.
Align internal anchor text to ensure that all anchor text points one direction (e.g. all xyz links should point to page ABC. My linking had become digital spaghetti.). I aligned 70K internal site links – a painful process. Take a hard look at your inbound links. Link building and blog seeding should appear natural. Stop any outright manipulation.
In case you missed step one, delete duplicative content. Saying the same thing in different words no longer works. Use a keyword tool like SEOQuake. Do you have lots of pages with the same exact top kw phrases? Add user generated content with FB or Disqus. Meaningful engagement from authentic users helps.
Ensure that your authors/site has authority. Be credible. Build authority. Would you give your site a credit card? Would you trust it for life and death medical advice? Fixate on bounce rate and time on site. These are indicators of value. Survey your users. Focus group them if you have the budget.
Like many, I believe that human quality scores are a factor with Panda. Matt Cutts said to make it look like Apple packaging – ”Think about something like an Apple product, when you buy an Apple product you open it up, the box is beautiful, the packaging is beautiful, the entire experience is really wonderful.” So, make your site beautiful and wonderful.
Settle in for the long haul. Adjust your business model and livelihood for less revenue. I lost seven figures and eight fine employees. It took me six months and 10,000+ tech hours to escape Panda. The entire event was a black box. Keep moving. Don’t wait for signals or clear direction. Follow the North Star in Setp One. No stone should be left unturned.
We edited 10% of our pages. We reskinned several thousand pages, upgraded the www and db servers for better speed, deployed a CDN, move to sprites, paralyzed downloads, externalized all js, html validated, compressed images, built a new UI, completely replaced all navigation, added FB, Twitter, G+, AddThis/ShareThis, FB comments, disqus comments, open graph tags, rel author tags, replaced most graphics…
We tackled cascading 301 redirect problems, filed hundreds of DMCA notices, threatened three large inbound linkers with legal action. We chewed up 20 people working overtime for 6 months in a living hell to fix every know technical issue – no matter how small. It would take 150 bullets to list all that we did. And to this day, we don’t know which was the magic bullet. We do know that we have a stronger better site.
Escaping Panda is going to take time and cost lots. Gaming the system is short sighted. The algo is too complicated with too many signals that cannot all be manipulated in unison. Focus on adding true value. Low-end SEO is dead. Despite having a good ROI, low-end SEO has always been short sighted. Spend your money building something that is the very best. When and if you recover, change your business model. Google is not your friend. We’ve moving into traditional media (radio, print and TV) after almost ten years of pure-play internet.”
I hope this post is useful for anyone reading and has got across the point that to recover from Google Panda then there needs to be a serious change of thinking for the business / websites as a whole. SEO is no longer about tricking Google into ranking a website higher, but about genuinely providing the best / highest quality website possible.
Doing so will mean that your site will survive future algorithm changes since Google is aiming to find high quality sites, so instead of trying to blend in with what are good websites why not actually become a good website that Google tries to base its algorithms on instead. Provide real value to your users.
If all this hasn’t convinced you how to ‘fix Panda‘ then have a good read of the interview of Vanessa Fox by Eric Enge over at Stone Temple about a holistic look at Panda, fantastic interview. Ask yourself what is your raison d’être?
Here is a list of all the websites quoted in this blog post for reference and easy of referring to if needed.