Identifying the attributes of consumer experience in Michelin-starred restaurants: a text-mining analysis of online customer reviews

Abstract

Purpose: The main goal of this paper is to identify the attributes of consumer experience in Michelin-starred restaurants and to estimate their effects on restaurant ratings. Design/methodology/approach: A sample of 70,233 online reviews of 224 Spanish Michelin-starred restaurants were analysed with the latent Dirichlet allocation algorithm. A sentiment analysis and a logistic regression analysis were also employed to estimate the effect of attributes on restaurant ratings. Findings: Customer attention, food quality, decor and ambience and value for money are frequently used to define restaurant experience. However, it is shown in this study that the experience in a Michelin-starred restaurant goes beyond the evaluation of those four attributes. Furthermore, the effect of the factors that were identified on customer satisfaction differed depending on the restaurant ratings. Research limitations/implications: The findings are linked to the context of Spanish Michelin-starred restaurants. It is also assumed in this study that online reviews are based on truthful opinions. Practical implications: Restaurant managers should primarily focus on customer attention and food quality to achieve customer satisfaction. In addition, those restaurants with an error-free service and a highly appreciated wine list among diners are more likely to achieve the culinary excellence that deserves a 5-star rating on TripAdvisor. Originality/value: The attributes of the restaurant experience are frequently identified in literature reviews. Research based on text-mining analyses of customer reviews to discover a posteriori the factors that define a restaurant experience is scarce, and particularly difficult to find in the context of Michelin-starred restaurants.