How Machine Learning Can Help with Customer Retention -- by Eugenia Inzaugarat

Customer Churn, also known as customer attrition, or customer turnover, is the loss of customers and it is an important and challenging problem for ecomerce and online businesses.

One of the metrics to keep track of customer churn is Retention Rate, an indication of to what degree the products satisfie a strong market demand, known as product-market fit. If a product-market fit is not satisfactory, a company will likely experience customers churning.

A powerful tool to analyse and improve Retention Rate is Churn Prediction; a technique that helps to find out which customer is more likely to churn in the given period of time.

The aim of this project is to:

• Build a classification model to predict which customers will churn
• Analyze how the different features affect retention or more specifically, customer churn

The dataset used for this project was obtained from Kaggle. It contains data about customers who are withdrawing their account from a bank.

This repo contains the following files:

• A jupyter notebook named churn_analysis.ipynb containing the exploratory data analysis, feature engineering, search for the best model, evaluations of the best models found, as well as the analysis of the feature importance.

• A CSV file named Churn_Modelling.csv which contains the data from the bank customers