Predictive analytics has been around for decades, but since the ubiquity of “big data,” the predictive modeling game has changed dramatically. Many data-brokerage companies aim to “predict” customer behavior based on a customer’s online clickstream data, while some specialize in the buying and selling of personal financial, household, geo-location and purchase history data in order to predict how lucrative a customer that person may be in the future. Using these “e-scores,” marketing and sales teams are able to aim messaging at a more targeted group of potential clients that are high on the purchase funnel.
As we get more advanced in behavior tracking and information aggregation, the predictions get more and more precise. But the info they gather is just what it says: predictive, not psychic. Consumers have no control over or access to this dossier of information, and therefore cannot augment, edit or change the information upon which marketers judge them. With this deluge of data, we are not able to predict, without a doubt, consumer behavior. We’re just better at guessing what they might do.