This project makes use of a fictional churn dataset from a Telecom provider called Databel to analyze customer churn. The dataset consists of 29 different columns and has one row per customer.

According to Investopedia, The Churn Rate, also known as the rate of attrition or customer churn, is the rate at which customers stop doing business with an entity. Since it is considered easier to keeping customers than getting the new ones, reducing churn is a priority for many companies. The simplified formula for churn is to divide customers lost by the total number of customers. There are multiple ways to calculate churn. It varies by industry and revenue model. For example, an e-commerce platform could define a churner as customers who have not made a purchase in the last 12 months.

After doing the analysis, here are some insights gained from the data:

1. General Insights The churn rate for Databel is around 27%. 45% of the reasons why customers churn is related to competitors. Comparing the rate in the whole states, the churn rate in the state of California (CA) is significantly high (>60%).

2. Churn Rate by Age Considering the age of the churners, the churn rate for senior citizens is 10% higher than the average (38.22%). By binning the age of the customers, the churn rate has in general an increasing trend through the age brackets.

3. Churn Rate by Contract Databel offers group contracts to customers from the same household. The advantage for the customer is a discounted rate, while it’s a great way for Databel to grow its customer base. From the analysis, the lowest churn rate is found in the contracts consisting of 6 customers in a group (23%). Furthermore, those who have monthly-based contracts churn more than the customers who have yearly-based contract.

4. Churn Rate by Data Plan Use Databel has a hypothesis that people who are not on an unlimited data plan are more likely to churn. From the data, it appears that customers who are on an unlimited plan are more likely to churn. To see if it is related to certain amount of mobile data(GB) being used, we further classify group on the basis of average monthly GB download. The highest rate is found in the group with unlimited plan and consume less than 5GB of data.

5. Churn Rate by International Activity The analysis requirement given by Databel includes a request to analyze the international activity of customers and its relationship to churn. They are curious about the behavior of customers who call internationally, and if paying for an international plan influences their loyalty. It seems that the state of California (CA) has a particularly high percentage of churners. When considering only this state, 72% of customers actively make international calls but do not have an international plan.

6. Churn Rate by Customer Service Databel also wants to improve its customer service since there have been some reported issues. Investigating customer service calls (the ones who call customer service often are also the ones who churn more than average) and payment method, it is found an average 1.47 calls per customer for those who are on a monthly contract and pay by direct debit. The churn rate for those customers are around 54%.

Here are some business plan we might offer to Databel regarding our findings:

1. Contacting customers who are on an interanational plan but have not called internationally and propose they downgrade their plan. By proposing a cheaper plan and explaining the rationale, it is expected that we can increase customer satisfaction and stop customers from churning. In this case, the previously mentioned 72% of customers without a plan can be potential clients for the new promotion of the international plan.
2. Since the churn rate is considered for customerswith direct debit payment method, it makes sense to advice Databel investigating what is going on with the payment. It is possible that there was a problem with the payment method that needs to be looked into.