Modern marketing, particularly with large data-sets, resembles nothing so much as the scientific method. It’s about formulating a certain hypothesis, testing it, and then presenting one’s findings to validate or modify the hypothesis. If one pursues the venture with rigor and diligence, the truth – whatever it may be – will be further illuminated. The truth, in the case of enterprise marketing, can be translated as “money.” Hypotheses tend to concern questions of customer response, retention and cross-selling/up-selling.

Consultant Eric Siegel does a nice job of outlining the process as it applies to customer intelligence. “A central capability of CRM analytics is to predict customer actions, that is, to forecast the individual behavior of each existing or prospective customer under certain conditions,” he writes. “Naturally, such customer predictions are key to allocating marketing and sales resources.”