Anna Andreeva – Postgraduate Student, Department of Business Analytics, Faculty of Business Informatics, National Research University Higher School of Economics.  
Address: 20, Myasnitskaya str., Moscow, 101000, Russian Federation.
E-mail: ann.v.andreeva@gmail.com

In the current economic situation with its high level of competition and high volatility of consumer preferences, methods of management based on impersonal mass production are giving way to client-oriented ones, or CRM - Customer Relationship Management.

The article focuses on the problem of improving the efficiency of a company’s customer management by using models predicting the number of clients on the basis of Markov’s chains.

The research results in suggesting a new approach to estimating parameters of the model, predicting the number of customers by using a matrix evaluating the ratio of new customers to those who dropped out. A model has been developed, which predicts the number of customers by taking into account the intensity of transitions between groups of customers, both for the case of constant and time-varying intensity. Factors, affecting the intensity of the transition of clients, have been identified. On this basis, an integrated model of customer relationship management has been suggested, allowing taking into account specifics of consumers’ behavior and socio-demographic differences between groups of customers, both with and without budget limitations. Also, a mathematical programming problem has been formulated which when applied delivers an optimal, in the terms of the accepted quality criteria, solution to the CRM tasks. At the same time, as the parameter evaluating the management efficiency, we chose the increase of net income from a customer, but not the probability of the customer's purchase.

The practical significance of the paper is based on the fact that it offered a model of optimal customer relationship management for addressing one of the most important tasks of tactical management of a company, which takes into account a wider set of features of customer base and impact of marketing activities on different groups of customers. The model can be applied in organizations working in the sector of consumer goods, when the impact of the previous customer-company interaction has not been diagnosed yet.