Abstract

Every business has one overarching goal: to maximise sales and profits. When a company's regular clientele suddenly stops buying from it, it usually sees a precipitous decline in revenue. It has been established that keeping existing customers is less costly than finding new ones, making it a top priority in Customer Relationship Management, particularly in the retail sector. When a client quits patronising a store, they no longer provide an opportunity for more purchases or even cross-selling. As a result, businesses need to take preventative measures by identifying at-risk customers so they can be retained. This article demonstrates the utility of combining transaction data with machine learning for churn prediction in the retail sector. A total of 5,115,472 customer loyalty card records were pulled from a European retailer's data warehouse and used to train the machine learning models. According to the findings, machine learning models outperform their linear regression counterparts

Description