RESEARCH ON METHODS OF DETERMINING CUSTOMER LOYALTY AND ASSESSING THEIR LEVEL OF SATISFACTION

Abstract

The subject of research in the article is methods of data collection and processing to assess the level of customer satisfaction and loyalty to the company, as well as the possibility of evaluating the collected data. The purpose of the work is the analysis of methods for determining customer loyalty and assessing their level of satisfaction, the development of a unified assessment algorithm based on various types of data. The article deals with the following tasks: analysis of information acquisition methods – questionnaires and reviews, evaluation methods definition and comparison questionnaires with closed answers, reviews and open answers’ text tonality evaluation methods analysis using an artificial intelligence, development of an algorithm for determining a unified evaluation and conducting an experiment. The following methods are used: theoretical research methods for determining existing data collection methods, as well as methods for assessing the level of customer loyalty and satisfaction using CSI, CSAT indexes, NLP methods for determining the text tonality, bringing the calculated values to one scale, determining the method of a unified assessment; empirical research methods for conducting an experiment, determining and proving the feasibility of applying the method. The following results were obtained: a method of assessing customer loyalty and their level of satisfaction based on the analysis of various types of information with further results unification was proposed. Various types of date are responses to questionnaires and user reviews. The questionnaires are analyzed using KPI, and reviews – using artificial intelligence methods. After normalizing the results (bringing them to one scale), the additive convolution method is used to unify the overall result. A prototype of the software system has been developed, which allows you to carry out a full cycle from collecting information to calculating both KPI-metrics and a unified assessment. Conclusions: Experimentally, it was determined that the method of assessing customer loyalty and their level of satisfaction based on the unification of a comprehensive assessment of various types of data is efficient and can be used to optimize business processes by reducing time and efforts spent on analyzing the gathered data. The use of this method is fully justified, since the measurement error is low, and the margin of error is acceptable.