A Machine Learning Approach to Predict Customer Churn of a Delivery Platform

The use of delivery platforms has become widespread due to the impact of the Covid-19 and the O2O industry. However, the ELEME delivery platform, a subsidiary of Alibaba Group, which represents China, has recently been losing market share. This means that companies need to constantly look at strategies to attract new customers and maintain existing ones. In general, it costs at least five times more to attract new customers than it does to manage existing customers. This paper attempts to predict customer churn using the ELEME customer dataset to develop strategies to identify and prevent churn in advance. The results of the analysis using machine learning approach found that the most influential feature that can predict churn is the number of clicks made by the user. This paper presents the process and explanation of applying various algorithms for predicting customer churn on a distribution platform. It also proposes strategies for dealing with customer churn.