Big data can be leveraged in various ways using analytics. In this chapter, we elaborate on some popular examples of applications without claiming to be exhaustive. More specifically, we zoom in on marketing analytics, fraud analytics, credit risk analytics, and HR analytics. For each of these areas, we define the overall modeling problem, the target to be optimized (if any), and the business implications. We also relate back to the key characteristics of successful analytical models in each of these settings as discussed in Chapter 1: accuracy, interpretability, operational efficiency, regulatory compliance, and economical cost. We give recommendations concerning which of the analytical techniques discussed in the previous chapters can be used to tackle each of the applications. Subsequent chapters then further elaborate by introducing the profit perspective.
Various types of marketing analytics can be distinguished. A key characteristic is that they all center on the customer as the basic entity of analytical modeling. Therefore, marketing analytics is sometimes also referred to as customer analytics. By carefully analyzing historical data and transactions, analytics can help us to understand customer behavior from an acquisition, retention, selling, segmentation, lifetime value, journey, or recommendation perspective as we discuss next.