About the Client
Telecom Service Provider/Retailer
Case Study
Construction of a segmentation model based on interactions with the network in order to locate and target customers more likely to churn
Problem Overview
Customers perceive fluctuations in network performance differently, and reactions are varied. The client needed a way to properly identify groupings of customers and their propensity to churn in order to proactively assist those customers.
- Based on all data towers used in a month, a central location was calculated for each customer.
- Distances between the central location and towers were used to calculate the customer's traveling spread in that month.
- Cell tower usage and studies on travel in the United States were utilized in order to determine distance groupings for the above spreads.
- The exercise was repeated over a three-month period of time and each customer was categorized based on the distance and variability of their travel, giving them their network footprint group.
- Each network footprint was then clustered to determine natural groupings within each network footprint.
- The clusters showed distinct patterns in churn and device purchasing behaviors, giving the client manageable groups to target to reduce customer churn.
TeraCrunch Solution
Impact On the Business
This solution is used by the client to understand why customers may be experiencing the same network but perceive it differently. By better understanding the customers, the client can target potential high-risk customers and reduce overall churn and increase customer satisfaction.
