Telecommunications

Data mining methods and approaches are widely used in the high-tech sector of telecommunications. The problems resolved are primarily related to loyalty programs, retention of the existing client base, and attraction of new customers.

The billing systems of telecommunication companies accumulate huge amounts of data. These first of all include subscribers’ information and statistics of services provided. It would be inefficient to analyze such information using manual and semi-automated methods. Loginom-based solutions allow creation of a client-oriented strategy by means of searching and interpreting the implicit patterns in the billing system data:

• Subscriber base monitoring. Loginom supports dozens of data visualization methods, including tables, charts, diagrams, multivariate maps, hierarchies, OLAP cubes etc. The user can generate any analytical report with just a few mouse-clicks and without the help of IT experts.
• Client profiling. The clustering technologies used in Loginom are indispensable for subscriber base segmentation. They work by identifying similar behavioral patterns including the frequency, duration and time of calls, as well as the monthly payments. This enables the user to plan marketing actions aimed at particular groups and identify the most and the least profitable segments.
• Targeted notification of subscribers. Companies often introduce new services, changes tariffs, and carry out seasonal actions. The efficient self-training methods implemented in Loginom can help to identify the segments of the existing client base to which new offers will be of the greatest interest.
• Client loss prevention. High client turnover is a problem for any company. Therefore it is important to identify the reasons that clients leave and estimate the probability of quitting of clients with specified characteristics. Based on the results of such research, users can develop new methods of customer servicing in order to improve loyalty. Loginom has various tools and algorithms which can be used for this purpose, including decision trees, neural networks, logistic regression etc.

Loginom is also capable of solving other problems typical of communication companies, namely loyalty programs, cross-sales of services, assessment of tariff plan efficiency and so on.