A leading mobile carrier used Qualitest’s solution to predict which clients had a high likelihood of purchasing roaming packages. By leveraging Qualitest’s solution, the company was able to identify clients planning to travel abroad and target them with roaming plan offers before they made a buying decision.
Direct calls and emails promoting roaming plans were hit-or-miss, yielding low accuracy, as it had been challenging for the company to identify which clients were interested in the service.
The company leveraged Qualitest’s machine learning technology to predict which clients were highly likely to travel abroad one month in advance. The goal was to increase roaming plan sales to existing clients through targeted direct emails.
Qualitest was guided by Data Citizen vision, empowering data analysts with no statistical or mathematical expertise to easily and efficiently create accurate prediction models. Qualitest provided the technology to apply 40 machine learning algorithms to targeted clients’ online browsing data, with an automatic fine-tuning mechanism. The software tool identified the best combination of explanatory variables along with the most suitable algorithm for the specific task.
The initial database included over 220 variables, but our experts determined that only 63 were influencing variables. Our software tool was used for data exploration and prediction model creation, eliminating the need for a Data Scientist. The solution’s ability to process data and build prediction models automatically—without writing a single line of code—proved highly efficient. In less than two weeks, a stable and transparent prediction model, including all necessary data preparation, was ready for deployment.
Qualitest’s solution provided a variety of methods for implementing the chosen model in a production environment. The implementation process included deploying the model based on a web service component.
The best model was implemented in the production environment. Based on clients’ online browsing data, the company was able to identify, with 70% accuracy, which clients would travel abroad one month before their flight. The company then targeted these clients with direct sales calls for roaming plans to increase sales.
Qualitest’s machine learning software tool for non-data scientists delivered effective results and achieved the company’s goals. The solution proved to be accurate, efficient, and powerful, increasing sales by enhancing the effectiveness of sales calls.
