Contact us: (800) 732-3440

Cable & Telecom Case Study – Wireless Broadband

Cable & Telecom Case Study – Wireless Broadband

AccuData uses predictive modeling to help top wireless communications provider up-sell and cross-sell their way to new cross
adds per month, while reducing churn.

Challenge

A leading U.S. wireless communications provider was experiencing high churn rates and a steady decline of customers. In the highly competitive wireless telecommunications market, intense competition was forcing companies to constantly find innovative ways to attract new and keep existing customers. They used everything from handsets and pricing, to services and network coverage in their special offers.

Solution

The company realized that if they were going to maintain, let alone grow their business, they had to first reduce churn. As a result, they began to evaluate new business opportunities and strategies that could improve both their cross-sell and up-sell success rates. A known lever to drive up gross adds and driving down churn. After hiring AccuData, they identified a promising opportunity that would target existing customers with an "add-a-line" direct mail promotion. The goal was also to ensure that the solution would seamlessly integrate with their existing systems - including a call center and Web site, as well as POS scripts that would prompt an add-a-line offer through all customer-initiated touch points. As a first step, the company focused on identifying existing customers who were more likely to add lines to their plan within the next few months. They then performed an exploratory data analysis (EDA) on thousands of customer mobile phone data points including usage, payment, contract, service and handset type, as well as external variables appended to their file. Next, they wanted to gain insight into specific customer group usage patterns. By performing a decision tree technique, based upon adjusted significance testing (CHAID analysis), they were able to capture manufactured, customized and predictive data points. This data was then used to conduct a regression analysis and develop a model that could rank customers by their likelihood of adding a new line within a specified timeframe. The modeling solution included geographic variables that accounted for differences across the various operating regions and helped the company.

Result

By using the predictive tool, the company was able to successfully cross-sell and up-sell customers with their new "add-a-line" campaign. The campaign helped the client "right size" accounts, while significantly reducing churn rates and improving profitability. Gross adds increased by almost 10,000/month while the client's overall churn was reduced by 10 basis points.