Churn Analysis, preventing more
1 hidden cell
Data:
The following project uses a dataset containing information on customers and how they related to the company, with a final detail that is very important; wether they churned at the end of the season or not.
Context:
The company of telecomunications gives services of internet, phone, cable; with different combinations and levels for them, but feels the drop of customers from their side each season, and wants to address the problem for the most likely root cause for it.
Process:
Using tools like python and multiple libraries (Pandas, Numpy, pyplot, seaborn, scipy) to import, clean, describe and then analyze the data. The process will explore relationship of the variables of interest, and find fields that may correlate by more than mere chance, finally addressing the most likely root cause for the problem.
Conclude:
At the end of the project a final conclusion will be given as well as a the point of attention, driven by data to focus on and make decisions.
Additional resources:
Externally there is a report in Power Bi to explore dinamically and will be available in my github.
Published GitHub: https://github.com/lrickso/project-Churn-Analysis
Demographics:
Gender:
Age:
Married status:
The root issue, Overall rate of churn:
Correlations of interest.
Contract type:
Tenure: