When customers or subscribers stop doing business with a company or service, it is known as customer churn. Customers in the telecom sector can choose from a wide range of service providers and actively swap between them. The telecommunications industry has a 15 to 25% yearly turnover rate in this fiercely competitive market. One of the biggest threats to revenue loss in the telecom sector is customer churn. Fostering customer loyalty is essential since the cost of recruiting new customers can be up to 25 times higher than the cost of keeping existing ones.
in our case a bank. Some studies show that acquiring new customers can cost 5 times more than that of satisfying and retaining existing customers. Thus tracking of bank customer churn rate through prediction will help in reducing marketing costs, lead to increase in capital ,expanding total customers and a lot more.
For the dataset we are going to work with it will enable us to predict a customer's churn. The dataset contains the following columns : RowNumber,CustomerId,Surname,CreditScore,Georaphy,Gender,Age,Tenure,Balance,NumOfProdcuts,HasCrCard,IsActiveMember,EstimatedSalary,Exited.
The dependent Variable is Exited which we will use to observe the outcome while the independent variables are the other columns which will enable us to get an outcome.
The Models were build using ANN and Logistic Regression to compare the outcomes after carrying out a predeiction for a customer details feed to the model.Logistic Regression was the model that we deployed on the web through flask this allowed a user to key in info that was feed to the model then a prediction would be carrried out and an outcome given