Too few, too many, just right: Creating the necessary number of segments for large online customer populations

Abstract

We develop a framework to reduce the number of customer segments to the smallest quantity without losing essential information of the underlying population in the electronic marketplace. As a use case of this approach, we create personas for these segments to enhance customer understanding. We use (a) matrix factorization to identify customer behaviors and construct customer segments, (b) statistical heuristics to collapse into meaningful segments, and (c) automation to enrich by generating a persona profile for each segment. We evaluate our approach in a case study using more than 21 million online flight bookings of a major airline company resulting in a 57.5% decrease from 1194 to 507 segments, thereby reducing segment noise. Three customer mega-segments emerge: Behaviorally same – Demographically different, Behaviorally different - Demographically same, and Behaviorally and Demographically different. As one of the first efforts at the essential task of customer segmentation reduction for large customer populations, findings have implications for organizations desiring to employ segmentation and/or personas for enhanced customer understanding.