MODELLING TOURISM DEMAND USING GOOGLE ANALYTICS: A CASE STUDY OF PORTUGAL’S ALENTEJO REGION

Abstract

The development of information and communication technologies, specifically the Internet, has changed the way tourists plan their trips, being one of the most important information sources for tourism decision-making. However, a limited number of studies has been developed to analyse the causal relationships between the web interaction and tourism demand. Therefore, this paper intends to shed light on the usefulness of big data analytics to understand the tourism demand of a destination. More specifically,
it aims to examine the causal relationship between website’s visitor interactions and the tourism demand of a destination and verify whether there are differences in this relationship according to the visitors' country of origin. In order to achieve the research objectives, the Alentejo region in Portugal was selected as a case study. Monthly data for the period between 2007 and 2017 was used to examine the long-run causal
relationship between the sessions of the users to the official website of the Destination Management Organization of Alentejo (measured through Google Analytics) and tourism demand of this region (measured trough the number of guests in tourism accommodation establishments). To analyse if there are differences in this relationship according to the country of origin of visitors, the most important tourism markets for this destination were selected. Cointegration (Johansen´s maximum-likelihood method), Granger causality test, Vector Autoregression Model, and Vector Error Correction Model were used to examine the relationship. The results reveal a causal relationship between Internet search and the tourism demand. However, this relationship varies among the tourism market analysed.