Customers today are becoming increasingly demanding, with their needs and desires growing more complex. Consequently, for businesses it is important to adapt to this new dynamic to maintain their competitiveness. By adjusting to new market conditions businesses can tailor their offer to meet specific customer needs and desires, which leads to establishment of long-term relationships. Once such a relationship is established, it is essential to manage it effectively to ensure long-term customer satisfaction and loyalty, which ultimately leads to increased revenue. In this long-term relationship businesses collect significant amount of information which through advanced analytical tools such as machine learning, can be used to predict customer behavior. At some point, customers may decide not to repurchase a companies product or service which directly impacts revenue. The objective of this paper was to use an analytical approach of customer relationship management to develop a classification model for predicting customer churn and to identify the variables that most significantly contribute to the models predictive power. A decision tree algorithm was used to create the model, following cross industry standard procedure for data mining (CRISP-DM). The model was developed using Python programming language. The model achieved an accuracy of 96% and an area under the curve (AUC) of 93%. The high accuracy indicates that the model correctly predicts customer churn, while the AUC confirms the models ability to distinguish between customers who churn and those who don't. The analysis revealed that the most significant variables for predicting churn where the total number of purchased products, the number of free smart devices and the total number of days between purchase and delivery. Other variables also contribute to the prediction but to a lesser extent. These results enable better targeting of customers and better optimization of marketing strategies which ultimately leads to increased revenue for the company.