Focused customer retention programs
Reducing churn is a key focus for many businesses, as retaining existing customers is often more cost-effective than acquiring new ones. Strategies for reducing churn include improving customer service, enhancing product or service offerings, implementing loyalty programs, and addressing issues that may be causing dissatisfaction among customers. Our goal through this exploratory analysis is to gain deeper insights into strategies that are more effective in keeping customers. For instance: what incentives should the company implement or what customer profiles should be targeted in marketing campaigns.
- Drop unnecessary columns: CustomerID
- Handling Missing values
- Filter Outliers: rows with tenture
- Standardized Data Types: categorical vs numerical variables
- Encoded categorical variables
- Targeted Variable Encoding: converted Churn to factor for classification analysis
- Logistic Regression
- Classification Tree
- Random Forest
- Support Vector Machine
- K-Nearest Neighbors
Considering the need for both high accuracy and interpretability in the telecommunication business, Logistic Regression seems like a strong candidate. It provides a transparent way to understand why customers might leave and allows for easy communication of the results to non-technical stakeholders, which is valuable for implementing strategic business decisions.
However, if we value predictive power more and have the capacity to handle a more complex model, Random Forest or Tree Classification might be better, as they offer slightly higher accuracy and can capture more complex relationships in the data.
- Samantha Vaga (@samanthav416)
- Zainab Sunny
- Kate Pferdner