“I eat bamboo and crap sites”
Contents
The First Signs of Recovery Came Early
After being struck down by Panda on April 11th 2011 I got to work and made a whole bunch of on-site changes. I made a small recovery quickly, however, soon learned that there was still a lot of work to do.
Apr 19, 2011
Glad to report that this morning it seems that the work I have been doing over the last week is paying off.
One of the sites that I manage saw a drop in traffic after the UK Google Panda update last Monday (11th April, 2011). Since then I have been working relentlessly, pretty much around the clock, to try to learn what went wrong and then to resolve the problems.
Google Panda is essentially a website quality update. It is the first step towards search engines returning quality sites that are not based on webmaster votes (i.e. links).”
I have been working about 12 hours a day, probably more some days, and have a change log with over 500 entries.
I spent a lot of time reading through the various Google webmaster forums and speaking to people that were trying to recover. The general concensus was always that “nobody recovers from Google Panda”, however, it seems that this is not true.
Labnol reported that they recovered after making a few changes to their site, it took them 2 weeks. My recovery is starting just 1 week after the UK Panda Update.
So what did I do? Simple – I followed Google’s Quality Guidelines. To the T. Admittedly the recovery is not a full one at the moment, but I am clawing my way back up, each day gets a little better, touch wood!
I will be writing more, in fact there may be a special report written shortly for those that want it. In short – build the best website possible for your users, and do not live in denial! Google Panda is essentially a website quality update. It is the first step in towards search engines returning quality sites that are not based on webmaster votes (i.e. links).
A Graph From A Slightly Recovered Site That I Manage
Still not fully recovered but moving in the right direction:
On the final day of this graph traffic finally exceeded the post Panda Slump, but not fully recovered yet. The green line is the traffic in the week running up to the Panda update, end of the line being the Tuesday 12th, and the blue line is the traffic for the week following the update.
Pagerank, Links and SEO
What Google did years ago was revolutionary – they developed a search engine that ranked pages based on how popular those pages were in the web community. When people linked to a page, in the days before Pagerank, it was because they liked a page, wanted to share it, or referenced it. So it was logical to use that data to determine which pages were best.
The alternative was human edited directories. Search engines before Pagerank were dire, using META keywords and descriptions to determine what a site was about and then listing all websites in order of relevance. This was spammed to death.
For a while Pagerank stopped spammy sites appearing in the search engine as simple META keyword spamming no longer worked. However, this lead to a whole new industry of SEO (search engine optimisation) which saw people swapping, trading and stealing links to enhance their website pagerank. The most aggressive SEO’s generally got their sites to the top of the search engines.
Also, in its early day the only way a link would be added to a site was by the webmaster himself (or herself). When “Web 2″ came along, comments, forums and wikis all suddenly provided a way for people to add their own links to another persons website. Spam was invented. Once again, poorer quality sites featured at the top while well researched and written sites were left buried.
Along Came A Panda
So over the last year or so (by all accounts) Google started working on a major update to its search engine. First called Google Farmer update by the professional SEO community, and then Google Panda by Google (apparently named after one of the brains behind it all), the new update was rolled out in the USA last month, and then hot all other English language sites on April 11th.
“This is an algorithmic change and it doesn’t have any manual exceptions applied to it” Michael Wyszomierski, a Google Employee.
The result was that many websites saw a major drop in Google search traffic, from between 30% to up to 99%. The site I manage dropped by around 68%, which was pretty major, especially as the pages that many advertisers targeted were hit hardest.
So How Did I Start My Recovery a from the Panda Penalty?
Without going into too much detail now, this was my approach (bear in mind that this is really a condensed version of about 80+ hours of work).
- I first searched for information on the sites that lost and those that gained. There are many websites and blogs talking about Panda, although many have data that is a little dubious (some websites report to have been listed as the biggest losers and yet say that they are doing just fine).
- I then read the Google Guidelines all over again, looking for any updates they were sharing
- I read the Google Webmaster forum threads to see more about they types of sites suffering.
- I ignored everyone that said “it is impossible to recover from Google Panda”
- I instead believed Google when they said that it was a totally algorithmic update and no sites are manually targeted.
So what did all this teach me? It taught me one simple (although I am sure Google would not like to hear me call it simple) fact – quality counts more than anything.
There were (or still are) many theories and rumours that I think are false about Panda:
- Google Panda does not downgrade a site based on its CMS. WordPress sites are only affected so much because so many people use them.
- Google Panda does not mind if you have a reasonable amount of advertising on a page. Many people stripped out all their ads and then noted that traffic did not recover. Advertising may be an issue, but generally it is not, in my opinion.
- Collateral Damage – this may still be partly true. Many SEOs think that one problem is that many sites dropped in ranking because their links (the Pagerank votes) from other sites have been reduced in value as a result of those sites being penalised by Google Panda. There may be some truth in this, although the only way to know is to test it by creating 2 identical and excellent websites and then creating the same number of links from equally pageranked sites, but one set considered “low quality” and the other set considered “high quality”. Of course, nobody in their right mind will ever test this. And first, you need to know what “quality” is.
So, where does that leave us in finding the solution? Simple – quality content.
When Google Panda first rolled out in the USA I started investigating a little, and asked over in the Google forums for suggestions on how we find low quality pages on our sites. No answers came. Just these words:
“If you believe you’ve been impacted by this change you should evaluate all the content on your site and do your best to improve the overall quality of the pages on your domain. Removing low quality pages or moving them to a different domain could help your rankings for the higher quality content.” said Michael Wyszomierski (a.k.a. Wysz) here (warning, there are already 1800 replies to this thread, only start reading if you have a week to spare!).
So I was left to try to determine how I could seek low quality pages. I created some rules that I felt could be followed by a program (search engine) and then went about manually finding pages on my site and adjusting them. It seems to have worked.
On Site SEO Alive and Well
For so long many people have relied too heavily on off-site SEO (link building) and not done enough work on their on site SEO (site architecture, content, copy, structure, navigation, usability, design etc.). All of this is now a factor, and all of it needs to be changed for the better to make your site rank well.
Soon I will reveal all, however, I have not washed for about 24 hours and so to celebrate the return of my page into the higher search positions in Google, I will have a shower and maybe also a nice cup of tea. Also I do not want to shout too much about how I recovered until I have at least a few days of data in the bank, and some steady positions. So off to the shower I go, for an early bath.
Wednesday 13th and Wednesday 20th Traffic Comparison
Traffic Change from Wednesday 13th to 20th April
(Many thanks to J. Patrick Fischer for sharing the photo of the panda).
How To Optimise A Website Post Google Panda / Farmer
Another positive update, but again, things did get worse! A real roller-coaster ride!
Apr 19, 2011
Finally some promising news post-Panda – we have managed to regain much of the traffic that was lost after the UK Google Panda Update, on one of the sites we manage. Not all traffic has been recovered, but the changes that we are making are having a positive effect.
We are now in the process of putting together a document that will guide other webmasters through the steps that we took to aid our recovery from the Google Panda update.
Please note that Webologist is not the site that we have been working on. The site in question is a medium sized information website, with between 1000-1500 pages and a global audience. It has several revenue streams from advertising and has a daily reach in the region of 20K pageviews. Target audience is the USA, UK, Canada, Australia and it also receives readers from various other countries to a lesser extent. However, as we do not wish to reveal our updates to our direct competitors we can only provide the information in a premium eBook.
The eBook should hopefully be available for purchase by next week. If you would like to know as soon as the book is available then simply leave a comment below and tick the “subscribe to comments” box (you will only get notifications for this thread – this is a standard WordPress feature and you can unsubscribe to comments at anytime). Once the book is ready I will update the comments list and you will be notified.
I will not be charging the earth for the book, but will offer a 25% discount to anyone that “signs up” below before it is published. I estimate the retail price to be just £30 (discounted to £22.50 for all of you!).
What Will The Panda Book Cover?
It will cover all the information that I have obtained from speaking with numerous SEO’s and Google employees (on their public forums) over the last month. It will cover the best practices that need to be followed to not only lift your Panda penalty but also to ensure that going forward any website that you develop will be optimised best for Google.
The book will cover all the steps that I took to repair some of the damage on a website that I manage. It will show the proof (in the form of Google Analytics screenshots – one will be posted on this blog entry shortly once all the data is in).
It will cover some essential site architecture improvements that have helped me. Although the work that I carried out was done a WordPress blog, the rules should apply to all websites.
More technical websites with a wide range of media (photographs and videos) may need to take slightly different action to what is taken for more information rich websites.
What I must warn is that the changes that are needed to recover from Panda are hard changes to make and also require a lot of hours work. There are no easy fixes. Some people have tried and failed to recover from the Panda update and believe that for some websites recovery is simply not possible.
However, in the last week I have proved that a partial recovery at least is possible within a week when you are willing to make some major changes to your site. This will of course be explained in detail, with some examples in the text. I am still making more changes and hoping to make a full recovery, the greatest task is finding the problems to then be able to correct them.
The Aim Of the Panda Recovery Manual
The aim of this eBook is simple: to provide the average webmaster the information and task list required to “clean up” their site and meet Google’s new quality guidelines.
You will probably need to put in a lot of hours to get your site fixed, but £30 for an eBook as a lot cheaper than hiring a pro SEO for a week. If you do wish to hire a pro SEO for a week, I do know some in the UK that may have some availability, use the contact form in that instance. Unfortunately I am not for hire at the moment.
So far results look promising, although we have not recovered all traffic. We are still experiencing a “Post-Panda Google Dance” which indicates that Google are making new changes each day and also that there are ranking updates resulting as the changes we have made are factored in as more of the pages are crawled.
It is possible that for any website there will be some issues that go beyond the scope of this document. We cannot guarantee that following the advice in the book will lead to your site bouncing back, all we can say is that it worked for our information website – a website that is dominated by content and works on an advertising business model – precisely the type of website that has generally been hammered by The Panda.
Fighting Pandas With Platinum SEO Pack 1.3.7
N.B. Since writing this post in April 2011 I moved from Platinum SEO Pack to the Yoast WordPress SEO. They are very similar but I felt that Yoast has a few further advantages.
Apr 13, 2011
OK, so one of the sites that I manage has been badly hit by the Google Panda update (previously known as Farmer Update). I have been busy implementing various changes which mostly deal with removing poor content.
It is generally thought that the latest Google search algo update penalizes websites that have some poor quality content. How poor quality content is defined is a mystery, but anything duplicated or abbreviated is likely to be top of the list of poor. So WordPress tag pages, category pages, archives and any other pages that just repeat content from the main page are likely suspects to bring a site down.
Is WordPress The Main Problem Regarding Panda?
Some people have commented how there are many WordPress sites that have suffered from the Panda update and believe that there is something specific about WordPress that has caused this. However, I do not believe that Google will penalize a website just because it is build on WordPress. It is more likely to be that WordPress sites have suffered most because it is such a popular platform. Google loves open source and people love WordPress because it is free.
So, what have I been doing today?
My first task was to delete all WordPress tags that had been used less than 5 times. Some people delete all tags and forget about them, but I do use them for navigation, they are very useful. For example, I have categorized this post as SEO Advice and so far tagged it as Panda. If I was to write more on Panda, I would tag the new articles Panda too. I do not plan on doing so though, as there are many heavyweight SEOs talking Panda to death already.
So, I deleted tags. I also tackled all the errors that I could from the Google Webmaster Tools, such as the soft and hard 404′s, duplicate titles and descriptions etc.
Next task is something that I picked up on Amit Agarwal’s blog about recovering from Panda. In it he explained that as well as trashing loads of tags and working through the error reports he also noindex sub-pages of tags and categories.
Now, this sounds like a good idea. On busy categories I may have 50 pages, and each of these is containing duplicate content. But, how to keep the main category page? I have until today been using the All in One SEO Pack, however, Tam, a regular poster over at Cre8asiteforums.com suggested that I give Platinum SEO Pack. I just tested it, and by jove it works! It does precisely what Amit Agarwal explains on his blog.
Other Good Things About Platinum SEO Pack
I was a little concerned about using it at first, as I was worried about the effect it may have on changing loads of descriptions. But it has a nice little import from All in One SEO Pack function.
The Procedure I Used To Migrate from All In One To Platinum
- Deactivate All in One SEO Pack
- Activate Platinum SEO Pack
- Press “Migrate from All in one SEO”
- Ignore the recommendation to backup the database (best backup just in case)
- Check that “Use noindex for sub pages” is ticked
Job done.
As an example, the source code for webologist.co.uk/category/google reads:
<!– platinum seo pack 1.3.7 –> <meta name=”robots” content=”index,follow,noodp,noydir” /> <link rel=”canonical” href=”/category/google” /> <!– /platinum one seo pack –>
and the source code for webologist.co.uk/category/google/page/2 reads:
<!– platinum seo pack 1.3.7 –> <meta name=”robots” content=”noindex,follow,noodp,noydir” /> <link rel=”canonical” href=”/category/google/page/2″ /> <!– /platinum one seo pack –>
Note that Google’s search bots can still spider their way through the sub-pages, they just will not be indexed. So links back to the main blog, i.e. posts, will be followed and then ranked accordingly.
Where have my descriptions gone?
Noticed 2 things:
- I seem to have lost descriptions. This is a bad thing.
- I have no idea what noodp and noydir are.
So, fix descriptions time.
“By checking this option, your META descriptions will get auto-generated, if there’s no excerpt. i.e. the first 160 characters of your post/page content will be automatically assigned to the meta description tag.”
Ok, it seems that as I was running the All In One SEO Pack I had deleted the WordPress META description code from the templates. Need to pop that back in.
Will test then update…….
Ah, no, descriptions are there on the post pages, just not the cat pages. That is OK, if Google decides to list the first category page then it can make up its own description.
So for this post, the META is:
<!– platinum seo pack 1.3.7 –>
<meta name=”robots” content=”index,follow,noodp,noydir” />
<meta name=”description” content=”OK, so one of the sites that I manage has been badly hit by the Google Panda update (previously known as Farmer Update). I have been busy implementing various” />
<link rel=”canonical” href=”/search-engine-optimisation/fighting-pandas-with-platinum-seo-pack-1-3-7″ />
<!– /platinum one seo pack –>
I do seem to have two descriptions though. It was on another blog that I deleted the default WordPress one. Above this description I also have:
<meta name=”description” content=”OK, so one of the sites that I manage has been badly hit by the Google Panda update (previously known as Farmer Update). I have been busy implementing vari…” />
<meta name=”keywords” content=”Panda,” />
<title>Fighting Pandas With Platinum SEO Pack 1.3.7 | Webologist</title>
Guess I can get shot of the first description that WordPress generates.
Anyway, if you want the Platy All In Oner then you can find it via your Plugin search in WordPress or get it direct from wordpress.org/extend/plugins/platinum-seo-pack
Will The Save Me From Pandas?
No idea.
What Is Your Google Panda Story?
While we are all working hard to improve our sites, maybe we could share some Panda stories. There has been a lot in the press about how the Google Panda update has affected some major companies, but as we all know, it is the smaller businesses that are really suffering.
OK, nobody wants to lose traffic, but if a company is seeing its search referrals drop from 1,000,000 a day down to 100,000 a day, they are still getting a lot of good business from Google and probably have a high exposure in social media and a loyal readership. Some people have mentioned how they have received thousands of emails from clients and fans since Panda, asking them where the site is – they still manage to find it though!
My Panda experience has been unpleasant to say the least, but I am seeing it as a new learning curve. Although I have lost a lot of good business from Panda, so much that my wife has had to go back to work (and I will look after my kids 1 day a week – that will be fun!) I am still confident that it will work out better in the end. Why?
I think that Google decided that enough was enough with poor quality PAGES, and Panda is there to stamp them out. Note, not websites of poor quality, but pages.
Many websites that have suffered talk about huge communities, forums that are 10 years old, millions of comments, videos, images, hundred of thousands of pages. This is all very well – but how many of those pages are actually read by customers and web surfers today? Maybe 10% at most. The rest sit dormant.
But think about poor old Googlebot, it still has to go and check those pages. Thousands of pages across millions of websites. Maybe the Panda update is a message to webmaster from Google, that is really just saying:
“Tidy up your mess, I am fed up with having to walk around the rubbish to get what I want!”
Other ideas is that it is designed to stamp out the scrapers and content thieves. Yes, many well established sites have suffered because of Panda. But think about it. If a professional scraper has a network of say, 1000 websites, that scrape 10,000 different domains, that is going to cost them a lot of money to run. If they too have lost a lot of traffic, they may be forced to close down their operations.
Yes, you may be suffering now, but maybe this will recover in time. The scrapers may still appear top of search for some terms, but not all.
Anyway, I digress. What is your Panda story? Share away.
