Airbnb listings’ performance: determinants and predictive models

Abstract

The present study analyzes Airbnb listings’ performance in terms of occupancy rate, number of bookings and revenue, by employing data mining methodologies. The research objective is twofold, to highlight the strongest determinants that influence customers’ purchase intentions and to propose reliable models capable of predicting listings’ performance. The data set refers to the Airbnb market of Thessaloniki, Greece and contains explanatory variables related to the hosts, lodgings, rules, and guests’ ratings. Elaborated classification methods derived from Machine Learning are used as analytical tools. The findings highlight the central role of the host. Superhost badge, rich host presentation and quick response to customers’ requests are factors that boost performance. Other determinants are the provision of amenities and high overall rating. In terms of predictive models, Random Forest outperforms its competitors and is proposed as the most suitable classifier for the specific domain. The paper contributes to the existing literature in several ways. It adopts a data-driven research approach, employs machine learning techniques, proposes reliable models capable of predicting listings’ performance and highlights the most influential determinants. The results and conclusions can be useful to individual hosts, professional listings’ managers, as well as legislative and taxation authorities.