WILL THEY COME OR NOT? THE INVESTIGATION OF THE NON-ATTENDANCE BEHAVIOR OF THE HOTEL GUESTS

Purpose – This paper investigates cancellations and no-shows in hotel reservations using granular transaction data from a property management system. The aim is to identify typical behavioural patterns of guests who cancel their stays or fail to attend. Methodology/Design/Approach – A dataset of 33,263 reservations was analysed using the CHAID decision tree model. Classification included various nominal and continuous variables: reservation status, number of guests, length of stay, distribution channel, gross rate, rate type, lead time, and number of rooms. Findings – The analysis revealed that cancellations are more likely for reservations with long lead times and flexible cancellation policies. No-shows are more frequently associated with bookings at lower gross rates. Reservations made via intermediaries such as OTAs and wholesalers show higher cancellation and no-show rates compared to other channels. Originality of the research – The study’s originality stems from the use of detailed user transaction data, offering empirical evidence of customer behaviour in the hotel booking process. By analysing real operational data, the study contributes to a deeper understanding of cancellation and no-show behaviour, supporting more accurate forecasting and revenue management strategies in the hospitality industry.