data science

How We Contributed to Building Dynamic Pricing Recommendations and to Detecting Fraud

WHO

Our client, a telecommunications provider that offers wholesale carrier services to wireline and wireless operators and service providers globally, was looking to get a dynamic approach to its pricing and to detect fraud to tackle revenue erosion.

WHY

Companies in the telecommunications  sector have to manage complex product portfolios and multiple contracts. In  this ecosystem where competitive forces and new regulation keep on emerging,  pricing has to be well balanced. Our client needed a dynamic approach to pricing, taking into  account these forces while being adjusted in real time to make accurate,  optimal, real-time pricing decisions on a daily basis.

At  the same time, our client identified fraud - marked by the misuse of  telecommunications products or services with the intention of illegally  acquiring money from a communications service provider or its customers, as  one of the biggest sources of revenue erosion.

WHAT

To create dynamic pricing recommendations, we worked hand in hand with our client's team to adjust prices to a greater degree of detail and accuracy. We had to design many approaches taking into account price elasticity and market cannibalisation - test and combine them, before we could produce accurate, dynamic, real-time pricing recommendations for all customers. Going through these diverse approaches, we were able to ensure that a consistent methodology was applied to pricing decisions, bringing new efficiency to pricing approaches.

In the end, this solution has been integrated and easy to use for the many stakeholders involved in pricing decisions, providing decision-makers with an accurate data-based approach to inform the entire pricing process and to increase revenues in a sustainable way.

As for fraud, we contributed to building our clients’ competitive advantage by empowering them with detection tools. It includes automatic customer identification and authentication; and real-time fraud protection based on analytical models that can identify particular service uses.