Improving guest satisfaction by identifying hotel service micro-elements failures through Deep Learning of online reviews

Cuadernos de Gestion. 2025. Vol. 25. No. 1. P. 71–88.

Kazakov S., Cuesta-Valiño P., Butkovskaya V., Muravsky D.

This study provides an in-depth examination of often-overlooked hotel service micro-elements within the broader spectrum of hospitality services, with the aim of improving service delivery and enhancing guest satisfaction. To achieve this, we develop a methodological framework that integrates: (a) VADER text-based sentiment analysis, (b) a robust logistic regression procedure to identify the specific hotel service components that trigger guest frustration, and (c) semantic network analysis to generate nuanced guest insights in the context of underperforming hotel service micro-elements.
The findings highlight fifty specific service micro-elements associated with negative sentiment and a subsequent decline in guest satisfaction. The study further focuses on the ten worst-performing micro-elements, using semantic network analysis to uncover the root causes behind common guest frustrations in hotel experiences. Although identified through hotel reviews, some service failures also have relevance for the broader field of destination management.
Overall, the results offer a valuable resource for managers to detect and correct malfunctioning hotel service micro-elements that are critical for increasing guest satisfaction in their properties. The findings also encourage hotel and destination managers to implement tailored strategies aimed at improving guest satisfaction across hotels and destinations.

Keywords: big data, online reviews, Electronic word-of-mouth, natural language processing, hotel service elements, tourist satisfaction

Publication based on the results of: Применение омниканального маркетинга для построения успешного взаимодействия с потребителями в условиях цифровой экономики (2025)