Abstract
PurposeShort-term rentals (STRs) (like Airbnb) are reshaping social behaviour, notably in gastronomy, altering how people dine while travelling. This study delves into revenue management, examining the impact of seasonality and dining options near guests’ Airbnb. Machine Learning analysis of Airbnb data suggests owners enhance revenue strategies by adjusting prices seasonally, taking nearby food amenities into account.Design/methodology/approach This study analysed 220 Airbnb establishments from Madrid, Spain, using consistent monthly price data from Seetransparent and environment variables from MapInfo GIS. The Machine Learning algorithm calculated average prices, determined seasonal prices, applied factor analysis to categorise months and used cluster analysis to identify tourism-dwelling typologies with similar seasonal behaviour, considering nearby supermarkets/restaurants by factors such as proximity and availability of food options.FindingsThe findings reveal seasonal variations in three groups, using Machine Learning to improve revenue management: Group 1 has strong autumn-winter patterns and fewer restaurants; Group 2 shows higher spring seasonality, likely catering to tourists, and has more restaurants, while Group 3 has year-round stability, fewer supermarkets and active shops, potentially affecting local restaurant dynamics. Food establishments in these groups may need to adapt their strategies accordingly to capitalise on these seasonal trends.Originality/value Current literature lacks information on how seasonality, rental housing and proximity to amenities are interconnected. The originality of this study is to fill this gap by enhancing the STR price predictive model through a Machine Learning study. By examining seasonal trends, rental housing dynamics, and the proximity of supermarkets and restaurants to STR properties, the research enhances our understanding and predictions of STR price fluctuations, particularly in relation to the availability and demand for food options.