Taming the Panda

Over the past couple of months I’ve been involved in a lot of discussions about the Google Panda Update at conferences and also as part of work I’ve done providing material for a couple of magazines.  Despite having been originally rolled out back in February in the US, and April in other territories, Panda remains one of the most important and highly discussed changes to Google’s algorithm for years.  Now, with the purported roll out of a further enhancement to Panda being revealed in  Search Engine Land, it seems that the discussion is set to continue over the next few months.

Google’s stated purpose with Panda was, as always, to improve the quality of their search results from a user perspective.  Larry Page has a much discussed obsession with speed and efficiency.  part of the reason why Google became so popular was because of the way in which it provided users with what they were looking for much more quickly than their competitors at the time.  Rather than having to refine their search multiple times, or use complex operators within the search box to track down content that met their needs, Google simply directed people to what they wanted quickly.

back in  the 1990s when Google launched, the web was much smaller than it is today – a couple of billion pages – and the barrier to getting new pages live was fairly high – you needed to have an understanding of HTML, and access to enough funds to put up a website.  Couple this with much less ability to monetise a web presence, and the ability to publish online was limited.  There was still an awful lot of crap on the net, and users still managed to find it, but it simply was not as profitable as it became.

Now, with simple publishing technology, hosting arrangements that cost pennies rather than thousands, and the portability of data, the amount of content on the web has exploded.  If you look at graphs of how many pages Google had in their index over time, you see a line that was and is exponential.  Layers of republished and spun content pound down on the “professional” commercial side of the web, and hide it from view.

With data storage and bandwidth dropping in price, and simple to implement advertising solutions such as AdSense, Chikita, and the myriad affiliate programmes, it becomes easy and cheap to build a profitable business online.    In a world where content was expensive, it did not make commercial sense to build a website targeting the long tail, but as it became easier to find content for a fraction of a penny per word, and arbitrage that effort against AdSense clicks, more and more publishers began to push out huge volumes of low quality content to target phrases.  This content sat on pages which did nothing to answer a user’s query but simply stood in the way of them getting to their final destination via a click on an advert.  Publishers who wanted to invest in quality content could not make an economic case for targeting the longer tail of search naturally, which left the playing field wide open.

Common signals within content farms make it very easy for Google to identify websites that provide an intermediary point in the user journey, and then ultimately block them from the results.  The more and more users who complained vocally about a perceived drop in the quality of Google’s results, the more the search engine needed to take action, and when they did, it led to an outpouring of emotion from people who saw their livelihoods stripped away over night.

Google were right to “do” Panda.  Users wanted it, and continued refinement means that they will maintain some kind of lead over content farms for a while.  Which is a good thing.  The question of course is what can publishers do to mitigate the impact of Panda on their business models, or adapt to a post panda world.

Taming the Panda

The key trends among sites that were penalised most heavily by Panda were that they were very large sites that published large amounts of very similar content very quickly, and then used page structures with mutliple advertising units on them to encourage as many clicks as possible.  Knowing this, and taking appropriate steps with your website to reduce the signature would help to avoid Panda penalties:

How Much Content Do You Need?

if you have 100 pages in your website each targeting slight variations on a phrase such as “Ways to save money on food”, then you probably have too much content on that subject.  I’d suggest having a single page that provides more detail, and using 301 redirects to point the limited value of the old network of pages back to the new authority page to aggregate the PageRank and relevance in a single location.

How Much Advertising Do You Need?

The fact that Google allow you to use 4 ad units on your page and 4 link units, and also allow you to use irritating Intellitext, and similar looking Chikita or other advert blocks does not mean that you need to.  The more adverts you have on a page, the greater the likelihood of links that have a low CPC.  Reduce the number of ads you have, and you ensure that only the ones with the higher CPM or CPC value will show.  You may get slightly fewer clicks, but they will pay out more.

Invest in Your Content

If the sole purpose of your website is to funnel users from the search results to clicks on adverts, then I’m sorry, but your business model is outmoded and ignorant.  If instead you think about what a user wants from your pages – information – and then provide it to them, you give them more.  If someone is searching for something like “How to I change the batteries on my Furbie”, provide them with actual information about how do perform that rather than providing some crappy content that skirts around the subject because you paid someone in a sweatshop $1 to write 500 words for you, and they couldn’t be bothered to learn English or do some research:

“One question that people are always asking is how do I change the batteries on my Furbie.  Well, the thing is that changing the batteries on your Furbie is actually really simple.  You need to have a Furbie and some batteries if you want to change the batteries on your Furbie, and then you should use new batteries, because if you change the Furbie batteries, and you have used old batteries, you will need to change the batteries on your furbie again very soon, which means that changing the Furbie Batteries becomes a regular occurence and takes more time than it needs to…”

How is that useful to someone?  If you’re writing a how to website, at least tell people what to do, rather than just creating thousands of useless pages cynically.

Recognise the Value of Users

You see those people whose adverts appear on your website?  They’ve invested in their content and their customers, and realise that there is more value available than just a single click.  They know that there is more potential to engage with a customer if you provide them with what they want rather than what you want.  They are willing to invest money to acquire customers.  If you are not, then you will soon find yourself out in the cold.

If another website is prepared to pay £50 to have a page of content written, and then spend £1 per click to get users to visit their website, you can be assured that they are able to run a business that makes a profit from that traffic.  they might sell more, or have the ability to capture user data that they can sell on.  Think about what you can do with your users, and invest in getting more of them.  There is no longer any point in simply trying to arbitrage traffic using your content farm, you need to think about users.  It might get harder, but it will also get more profitable.