Data Analysis workflow
Step 1 - Data Check
This step can be divided into 2 sub steps a.) Check for duplicate or missing values b.) Do a sense check with other internal data sources.
Step 2 - Data Exploration
This step involves asking right questions and explore data accordingly. Also in this step develop simple visualization to understand the look and feel of the data.
Step 3- Analyze & Visualize Data
In this step the essential thing is to choose right set of visualizations to answer business question and also perform advanced analysis to make sure not to miss on any important insights.
Step 4 - Dashboarding
Integrate multiple visualizations to create a dashboard which can provide insights to the end customers in a streamline and self-serve manner.
Step5 - Communicating Insights
This is the most essential part of the overall data analysis workflow as it requires to communicate key insights to the stakeholders based on the analysis conducted.
Problem Statement
Discover why customers are churning i.e. the rate at which customers stop doing business with an entity. Churn rate is the rate of attrition of a customer or the rate at which customers stop doing business with an entity.
Simplified Formula of Churn Rate
Churn Rate = customers lost / total number of customers within a certain time period
Data Used for Analysis
- Data of a fictitious Telecom provider
- There are 29 different attributes
- One row per customer
- Snapshot of the database at a specific point in time
Insights Discovered
- The churn rate for the company is ~27%
- ~45% of the reasons why customers churn is related to competitors
- The churn rate in California is abnormally high (>60%)
- Senior Citizens churn more often then average with churn rate of 38.46%
Data Visualization using Power BI