A Supervised Learning Approach to Store Choice Behavior Modeling Using Consumer Panel Metrics

In today’s competitive atmosphere, consumers have vastly diverse expectations of the product, value, and environment of their shopping channels. Understanding store choice behavior guides the fast-moving consumer goods players to revise their strategy accordingly. The purpose of this study is to explore the determinants of store choice in fabric detergents sector using household consumer panel data. To do this, we first suggested a definition of store loyalty based on household consumption volumes in different fast-moving consumer goods channels. Then, we used supervised machine learning methods to understand the factors behind the store choice process. The case study was conducted based on 2020 calendar year data of fabric detergents sector in Turkey. We used consumer profiles and FMCG consumption data of 15858 households. Results show that total detergents consumption, total purchase of self-care products, food consumption and also customers’ geographical regions are among the most important factors behind the store choice.