Discovering marketing rules for the tourist sector in visitor service quality surveys

Abstract

Advances in information technology nowadays offer to marketers intelligent tools which are able to transform all sorts of data into useful knowledge. Well known methods for discovering knowledge in primary data and applying them to solve marketing problems can be roughly categorized in two large families, data mining and statistical data analysis. In this paper, methods from both categories were applied on primary survey data in order to reveal the strengths and weaknesses of each approach for marketing decision support. The problem considered was to improve the positioning and the perceived quality of hotels, by relating tourist characteristics with their perceptions and satisfaction attributes. The aim was to extract hidden knowledge from standard service quality surveys and to provide it in the form of rules to support the marketer in his decision making. Initially, data from a questionnaire-based survey on expectations and satisfaction of tourists from their hotels were analyzed using multidimensional factor analysis, followed by a knowledge modeling process, which resulted in a rule-set suitable for decisions in hotel positioning. The next step was to apply association rule mining and decision tree methods to the same dataset, resulting in a new set of rules. It was found that the two approaches produced complementary knowledge, which revealed different aspects of the underlying trends that could be consolidated and increase the effectiveness of marketing decisions.