This project involves a comprehensive analysis and prediction of customer churn for a leading online E-commerce company. The goal is to identify customers who are likely to churn, enabling the company to take proactive measures, such as offering promotional deals to enhance customer retention.
The dataset comprises two tables:
E Comm - 5630 rows, 20 columns Dataset Details
The primary table, E Comm, contains information about individual customers, with 20 different attributes for each customer. The columns in this table are as follows:
E Comm CustomerID - Unique customer ID.
E Comm Churn - Churn flag (indicator of whether a customer has churned).
E Comm Tenure - Duration of a customer's relationship with the organization.
E Comm PreferredLoginDevice - The device that a customer most often uses to log in.
E Comm CityTier - The tier of the city in which the customer lives.
E Comm WarehouseToHome - The distance from the warehouse to the customer's home.
E Comm PreferredPaymentMode - The customer's preferred payment method.
E Comm Gender - The gender of the customer.
E Comm HourSpendOnApp - The number of hours the customer spends on the mobile application or website.
E Comm NumberOfDeviceRegistered - The total number of devices registered for a particular customer.
E Comm PreferedOrderCat - The category of items that a customer most often ordered last month.
E Comm SatisfactionScore - The customer's satisfaction score for the service.
E Comm MaritalStatus - The marital status of the customer.
E Comm NumberOfAddress - The total number of addresses registered for a particular customer.
E Comm Complain - An indicator of whether the customer raised any complaints last month.
E Comm OrderAmountHikeFromlastYear - The percentage increase in the order amount from last year.
E Comm CouponUsed - The total number of coupons used by a customer last month.
E Comm OrderCount - The total number of orders placed by a customer last month.
E Comm DaySinceLastOrder - The number of days since the customer's last order.
E Comm CashbackAmount - The average cashback amount the customer received last month.
By using these data points, we aim to build a predictive model that can accurately identify potential customer churn and assist the company in retaining valuable customers.
Stay tuned for updates on our progress and findings!