Use case

Customer Profiling

Problem definition

When a company becomes aware that knowing customer’s preferences is the key to success, then it usually becomes a customer-centric organization. Understanding customer behavior provides it with better insights, as well with as a possibility to redefine communication in key market segments, which in turn increases customer satisfaction and the number of purchases.


Companies can maximize profit and improve the response to campaigns, all by acknowledging customer needs and preferences in the right moment. Creating different customer behavioral profiles enables organizations to design a tailor-made offering for each and every profile.


The solution lies in clustering machine learning algorithm, based on the most important customer characteristics – such as frequency, profitability, activity, cost, gain and much more. Customers are divided into segments, grouping similar profiles together. It is possible to create specific profiles, in line with specific business goals.

In the telecommunication industry, different segmentation techniques are used for consumer postpaid segment, prepaid customers, small business customers and bundled offers. Customer profiling often acknowledges dynamic activity patterns and combines this technique with personalized targeting engine. Customized dashboard is used for tracking user’s movement between the profiles over time, as well as to display main profile characteristics.


  • Defined customer clusters and main characteristics in the database.
  • Simple overview of customer’s initial cluster and cluster migrations.
  • Complete integration with the campaign management system.
  • Customized dashboard for all relevant statistics.