SEO Experiments Now Come In A New Flavor: Google’s Messing With Your SERPs
SEO experimentation is a science. Hypothesis: If I increase the quantity of inbound links pointing to a page (independent variable), then I will see my ranking on the search engine results page (SERP) for the page’s targeted keywords increase (dependent variable). By running enough scientific experiments that were in similar, carefully-controlled environments, SEO’s can make very precise conclusions, i.e. 10 targeted exact-match anchor text links in a guest post author bio no longer increase SERP rankings for a competitive keyword after the latest Penguin algorithm update.
At least, they could make precise conclusions. This causality mindset is being phased-out as a trend of the past as we speak.
We’ll release this hit single:
I pinged you on the web back in the dot com blues
Eye fixed on my screen oh what have I got to lose
I had things tested so I knew what tactics to choose
You create this patent with your new technology
Fight web spamming techniques that users all agree
and now I don’t understand the tests I used to see
Oh oh – I did some experiments
Oh oh – Why did you kill them
Google Killed the SEO star
Google Killed the SEO star
SEO Science Called into Question
On August 14, 2012, Google was granted a harmless sounding patent called “Ranking documents ,” which describes how Google will randomly move the search engine ranking of a website when “rank-modifying spamming techniques” are being used. Basically, if Google suspects your site uses rank-modifying techniques (because, for example, some change you have made to a page has led to an increase in rank), they may intentionally change the ranking of your page for up to 70 days. Based on the adjustments that happen on said page, Google gains clues as to whether you’re intentionally trying to game its search algorithm. From my understanding of the patent’s wording, anyone who is intentionally doing stuff other than “creating great content” for higher SERPs is a spammer. (Don’t tell me you never thought about asking your friends to help you like your website!)
If you need help on likes, Matthew Inman can help teach you a thing or two about being awesome.
Remember the example on guest post author bio links? Well those links may or may not have increased SERP rankings for their targeted keywords, because links that were built may or may not have been flagged as spam. Only Google would know. Aside from the 500+ algorithm updates Google performs yearly, you now have an even bigger worry when you run your experiments. Is Google directly tampering with my SERPS and playing mind games with me?! You’ll never know for sure, so now your experiments have lost even more credibility.
“Link-based manipulation may include the creation or the manipulation of a first document or a set of first documents to include a link or a number of links to a second document in an attempt to increase the rank of the second document… Such manipulation of search results degrades the quality of search results provided by existing search engines.” (Emphasis mine) – Ross Koningstein, inventor of Ranking documents patent
Welcome to the age of transparent nontransparency.
SEO Experimentation IS NOT DEAD
SEO isn’t just about rankings. Let’s not forget that there are also plenty of A/B tests that impact conversion and not SERPs. While traffic is good, qualified traffic is better, and engaged qualified traffic is the best. Conversion optimization and user experience are two other areas SEO’s test to improve visitor engagement. Not all SEO experiments will cease to exist – correlational studies are actually getting more and more popular.
SEO correlational studies have never been so widely shared. (Have you read SEOMoz’s Search Engine Ranking Factors ?) Even though I personally don’t find correlational studies actionable, you can often spot the ones that are usually directionally correct, and the ones that are just misusing statistics.
e.g. Do SEO’s create more webspam or does the increased amount of webspam cause more people to want to become SEO’s? That’s actually the wrong question. The correlation only shows that webspam and the number of SEO’s on LinkedIn are both increasing in a similar trend.
Is This Another Case of Google Being Evil?
In my personal opinion, the patent’s aims are noble – to combat web spam for end users. Sure, some publishers could lose their evil link-building prowess, but others will gain simply from shifting the focus from tracking SERPs and tactics to creating great content. Sometimes having perfect data can hinder you, as the effort spent to ensure its perfection is unnecessary – trends are perfectly viable forms of data analysis to make sound business judgments.
Remember, we’re absolutely talking in a theoretical sense here because nobody has found proof of this patent in action yet. I wouldn’t call this patent the Higgs Boson of Google’s increasingly complex algorithm, but we all need to be wary of its potential impact on SERPs because of all the SEO snake oil salesmen out there with their new #1 SERP case studies. Can you imagine meta tag stuffing temporarily improving a web page’s SERP?
Do spammy SEO tactics work? We’ll never know for sure…
Bill Slawski  is always the first to react on any sort of Google patent. He does a great job going more in-depth and breaking down key points of the patent.
Aaron Wall  believes this patent is another example of Google being evil.
My colleague and favorite blogstress, Elisa Gabbert, pointed me towards a very recent article written by Ryan DeShazer  who earnestly asks, “Are SEO Spammers?”
Victor Pan  is WordStream's resident search samurai. When he's not busy gathering and analyzing web data, he's legitimately practicing the way of the sword, kendo.