Decoding hotel reviewers: insights from a decision tree analysis

Abstract

With the rise of user-generated content, understanding travelers’ propensity to share hotel experiences online has become a critical research area in tourism and hospitality management. This study examines factors influencing travelers’ likelihood of writing hotel reviews, with a particular focus on the role of different online platforms and sociodemographic characteristics. While previous research has explored general motivations for online review writing, little is known about how specific platforms influence review contributions. To address this gap, we conducted an online survey with 739 travelers and applied a machine learning (ML) technique – decision trees (DTs) – to classify customers based on their likelihood to write reviews. This approach allowed us to model non-linear relationships while ensuring interpretability for marketing practitioners. Our findings reveal that frequent use of Facebook for hotel searches, combined with being employed, is the strongest predictor of review-writing behavior, highlighting the role of social media engagement. However, limitations such as non-probability sampling and the omission of certain variables may impact generalizability. This study contributes to the literature by examining the link between platform usage and review-writing behavior, an area that has received limited attention. By employing DTs, a transparent ML approach, our findings offer actionable insights for hotel managers, helping them identify and engage potential reviewers more effectively. Understanding how platform engagement influences review behavior can enhance marketing strategies, encourage review generation, and ultimately shape consumer decision-making in the hospitality industry.