A few weeks after presenting your dashboard to the management, the Retention Manager from the telecom reaches out to you directly. He was impressed by your work and asked if you can put together a dashboard about customer retention.
In addition, to better understand the data, the telecom Retention Manager has scheduled a meeting with the engagement partner at PwC to cover these points:
- Customers in the telecom industry are hard-earned: we don’t want to lose them
- The retention department is here to get customers back in case of termination
- Currently, we get in touch after they have terminated the contract, but this is reactionary:
- it would be better to know in advance who is at risk
- We have done customer analysis with Excel: it has always ended in a dead-end
- We would like to know more about our customers: visualized clearly so that it’s self-explanatory for our management
- The Retention Manager has provided some information, have a look through the resource section.
This project will focus on analyzing the reasons for customers quitting and understanding where it actually has impact the most, then we look for Solutions that can drive the Retention Manager to a fruitful Operation. Particularly, we are going to be focusing on ..
Dragging out the reasons for the churning from the Majority of the dataset columns, running uni-variate and Bi-variate analysis between occasions to understand fully where people churned and where they didn't.
We can find the names of the columns on python using ChurnDataset.shape
or on R colnames(ChurnDataset)
.
Otherwise here are the Column Names from the Call Center Dataset with further details to go in with.
- CustomerID: Registered Identification of the customers in the Pwc Database, every ID is unique.
- gender: The gender of the Pwc customers in the dataset.
- Senior Citizen: The data shows if the customer is a senior citizen or not, 0 representing No and 1 representing Yes.
- Partner: This tends to relate the company with the people, the column shows if the Customer is a Partner or Not
- Dependents: Cant Drags this down to a particular meaning but we would say it shows the people who are relying on someone for earnings and those who are not.
- tenure: The number of Years the customers have been using our services.
- PhoneService: This shows of the Person uses phone services or not
- MultipleLines: Checks if the Customer has multiple lines or just one. if the customers have no Phone, It Shows no phone Services.
- InternetServices: Kind of Internet Service being used. e.g DSL
- OnlineSecurity: If the customer is using online security or not
- OnlineBackup: If the Online Security had Online Backup
- DeviceProtection: If the Customer Subscribed for device protection
- TechSupport: If the service hes wanted has tech support
- StreamingTV: If he/she do have streaming TV show or not.
- StreamingMovies: If the customer always streams movies or not
- Contract: The Kind of contract the Customer has with the Orgainsation.
- PaperlessBilling: If the Customer Does Paperless Billing or not
- PaymentMethod: What payment method the customer uses, Bank Transfer, Electronic, etc.
- Monthly Charges: The Monthly charges Reduced to the best average on a customer.
- TotalCharges: The total Charges collected from the customer
- numAdminTickets: Nil
- NumTechTickets: Nil
- Churn: where the analysis is based, The shows if the customer is an active one or not.
Your colleague, the engagement partner, asks you to do the following tasks:
Define proper KPI's
Create a dashboard for the retention manager reflecting the KPI's
Prepare a nice report to him (the engagement partner) explaining your findings, and include suggestions as to what needs to be changed
R Prog I | R Prog II |
---|---|
![]() |
![]() |
- A Python Script containing generated coding for showing insights on the Churn Dataset
- An interactive dashboard containing important KPI's and the rest of unanswered questions from the Churn dataset.
- An R Script containing generated coding for showing insights on the Churn Dataset
- A report containing Useful Findings on Churn Dataset.
Python I | Python II |
---|---|
![]() |
![]() |
- The data cannot be shared with unauthorized personnel.
- The data cannot be modified or altered in any way.
- The project must comply with data privacy regulations.
- The project must be completed within the allocated time and budget.
The data analytics project for Customer Retention is an important initiative that will help the Project Manager. The external consultant will work closely with the Project team to ensure that the project is successful and delivers the expected results.