Behavioral Analytics

DEFINITION of 'Behavioral Analytics'

An area of data analytics that focuses on providing insight into the actions of people. Behavioral analytics is used in ecommerce, gaming, social media, and other applications to identify opportunities to optimize in order to realize specific business outcomes. Behavioral analytics includes demographic and geographic data, but it also goes deeper in profiling a user’s past activity and pulling in any additional data that is available.  

BREAKING DOWN 'Behavioral Analytics'

Behavioral analytics is based off hard data, but there is a lot of ambiguity when it comes to the nature of the insights that it yields. For example, online advertisers use behavioral analytics to help them tailor the right offer at the right time. This is often done using the user’s demographic data, any past search or social information, and a locational market to put the user into a bigger group, sometimes called a cohort. The user is then served with ads or offers that match the ads and offers that have the highest success rate with that group.

Behavioral Analytics and Experimentation

The ambiguity in behavioral analytics enters when it comes to establishing the cause of an action. For example, say you have data that shows people reading an article about a product tend to click on the article link and then 30% click-through to view the product page and 5% add the product to their shopping cart. But another data set shows that people coming directly into the product page only buy the product 2% of the time.

You now have the behavioral analytics you need to build a few hypotheses to test. Perhaps the difference maker is the content in the article and you can increase the incoming search traffic conversions by integrating that content into the product page descriptions. Perhaps it is a look and feel problem that can be addressed by making the product page layout closer to that of the article page. Perhaps completely different ideas present themselves by adding in more demographic data - is there a difference based on the age or location of the eventual purchaser? Do older buyers read the article before making a purchase while younger buyers go straight to the product page to make a purchase? If so, where are the younger buyers getting their pre-purchase information from and can you influence that?

The behavioral analytics can support a number of different hypotheses, so the process of elimination comes from experimentation and evaluation. Businesses usually are looking to increase conversions, so if the change makes it worse, that hypothesis can be thrown out in favor of a different one or no change at all. Behavioral analytics are most often used to inform A/B testing where one variable is changed at a time. As the behavioral analytics have deepened and the technology to test multiple changes in real time evolves, companies are getting much better at targeting customers.

The Future of Behavioral Analytics

Amazon offers a personalized homepage based on demographics, past purchases, search queries, and products viewed using behavioral analytics - and each product page shows you what people just like you did after viewing that page. This data trove is the real power behind Amazon. Starting in 2015, Amazon was among tech companies like Google in releasing in-home voice products that will become a wealth of behavioral analytics on your off-line life just as your actions on their sites are a source of data for your online life.