Churn prediction using machine learning: A coupon optimization technique
1 University of Nottingham, England.
2 University of Stirling, Scotland.
Research Article
World Journal of Advanced Engineering Technology and Sciences, 2024, 12(02), 332–354.
Article DOI: 10.30574/wjaets.2024.12.2.0310
Publication history:
Received on 09 June 2024; revised on 16 July 2024; accepted on 19 July 2024
Abstract:
Customer retention has been identified as one of the most crucial difficulties in every Business particularly in the grocery retail industry. In this context, an accurate forecast of whether a client will leave the organisation, also known as churn prediction, is critical for businesses to undertake successful retention strategies. High churn rates result in massive losses for corporations because keeping existing customers is more profitable than getting new ones and getting a new customer costs five times as much as keeping an old one. As a result, firms should be able to track churn rates to calculate client churn. Also, if we know which clients are going to quit before they do, we may devise preventative measures. Also, current marketing strategies such as giving coupons to customers, the majority of whom do not use them, incur significant costs for marketing and sending. Knowing which customers are not going to use that coupon will assist organisations in devising alternative strategies to retain that customer rather than sending coupons. This paper studies Dunnhumby data and proposes 2 different models one for predicting the churn and the other for coupon redemption model and both uses XGBoost Classifier Model. When both models are used together, one will predict if the customer is going to churn, and to prevent churn, we use marketing techniques such as sending coupons, so the coupon redemption model will target whether the customer will use the coupon or not, so we do not send them those coupons and propose different retention methods for these customers. This can help businesses save money by reducing churners and saving money on marketing staff and sending promotions.
Keywords:
Churn rate; Machine Learning; E-commerce; CRM; Data Analytics; Coupon Optimization
Full text article in PDF:
Copyright information:
Copyright © 2024 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0