Overview

A leading telecom service provider spread globally, it aimed to expand its customer base while retaining existing customers and optimizing the company’s profit. The company wanted to understand customer behavioral patterns for the execution of effective campaigns.

Solution

Understanding customer behavior included studying the usage, spend, and attributes like region, subscription and so on. With the help of this data, the telecom service provider wanted to optimize its templates across channels to increase customer engagement.

Process

We at Vinmar, proposed machine learning-based solutions to fulfill the client’s needs. We opted for a two-dimensional approach. In the first, we used a multi ARM bandit algorithm to identify the right template for the customer. In the second, we used a decision tree algorithm to identify the right set of customers for the campaign, ensuring a higher conversion rate.

Result

Once this two-dimensional approach was executed, the client realized a higher return on investment with better customer engagements. Our machine learning based solution also incorporated automated messages to reduce the communication gap, manual spending, and helped in increasing operational efficiency.