BCGX PowerCo

Retaining customer clients vulnerable to churn¶

Part of the BCG Data Science Internship program

In this project, we work with another client who is a major gas and electricity utility that supplies to small and medium sized enterprises

The energy market has had a lot of change in recent years and there are more options than ever for customers to choose from
They are concerned about their customers leaving for better offers from other energy providers
We investigate whether price sensitivity is the most influential factor for a customer churning
Conduct feature engineering that is used to test hypotheses related to customer churn
And finally we utilise predictive modelling so that it can be used to highlight customers at risk of churn

Business Understanding & Hypothesis Framing¶

What you'll learn
- Meet your client PowerCo - a major gas and electricity utility who is concerned about losing customers
- How to interpret the business context
- How to break down the problem before you start your data analysis
What you'll do
- Determine the client data needed for analysis
- Outline the techniques you'll use to investigate your client's problem
- Write an email to your Associate Director summarizing your approach
Exploratory Data Analysis¶

What you'll learn
- How to investigate whether price sensitivity is the most influential factor for a customer churning
- How to use frameworks to conduct exploratory data analysis
What you'll do
- Use python to analyze client data
- Create data visualizations to help you interpret key trends
Feature Engineering & Modelling¶

What you'll learn
- How feature engineering can be used to test hypotheses
- How to build features to analyse the data for PowerCo
What you'll do
- Use Python to build a new feature for your analysis
Findings & Recommendations¶

What you'll learn
- How predictive modelling can be used to indicate churn risk
- How to communicate your insights with clients
What you'll do
- Build a predictive model for churn using a random forest technique
- Write an executive summary with your findings