Research


PenguinAnalysis and the Penguin Vulnerability Score is a new class of scientifically-derived SEO tools. Using the same types of machine-learning techniques used to create the Penguin algorithm, we have built a model for assessing Penguin risk based on leading industry data sets. We sourced data from providers across the web including…

and our own internal tools analyzing on-page and on-site factors that contribute to Penguin vulnerability. With over 70 variables ranging from CitationFlow Standard Deviations above the Mean to mozTrust Differentials, we applied these to a massive data set of known Penguin 2.0 victims and survivors. Applying Gradient Boosting machine learning, we built a prediction model that can categorize increased Penguin risk with over 90% accuracy.

plot-graph

More importantly, we can segment link strategies and site features that are the likely culprits behind increased Penguin vulnerability and help you inoculate your site against future Penguin updates.

Update: Penguin 2.1

Many of you may wonder how the Penguin Vulnerability Score held up against the latest data refresh of Penguin. We are proud to say that it came out with flying colors. Of the sample data set of sites hit by Penguin on the weekend of October 5th, 2013, sites hit by Penguin 2.1 had an average Penguin Vulnerability Score of over 280%. The average score of commercial websites that outranked those sites and did not get penalized was 110%. Those hit by Penguin 2.1 on average had Penguin Vulnerability Scores more than twice as high as those that did not.

So what are you waiting for? Go ahead and Get Started today!