Boutique Hotel Service Digitalization: A Business Owner Study

Abstract

The COVID-19 pandemic has generated negative, economic impacts on the tourism and leisure sector in Thailand, especially small boutique hotels. These hotels have had to develop more efficient and innovative approaches to meet new normal expectations, for example, contactless service. Digital technologies, such as Machine Learning and Artificial Intelligence, can open new possibilities and opportunities for hotels to digitize their customers’ services. A review of the literature indicated that data important to the management of hotel products and services include Customer Segmentation, Customer Profiling, Menu Engineering, Productivity Indexing, Customer Associations, Forecasting, Energy Consumption, and Room Rates. These characteristics can be examined by machine learning. This study used a mixed qualitative and quantitative research method. The data were gathered by interviewing two boutique hotel owners in Bangkok and collecting the hotels’ data, including online travel booking agents and direct booking logs, for the period April 2016 – September 2021. The analysis was conducted using the booking data from the two hotels: 3946 records from Hotel A and 3948 from Hotel B. In this research, k-means clustering was used to segment hotel guests. Two-class logistic regression and a two-class boosted decision tree were used to predict the prospective customer, while linear regression and decision forest regression were used to forecast the market demand. The findings reveal a model of hotel business owners’ requirements to innovate new service solutions, such as the contactless software solution, that guests can employ for check-in, check-out, order services, and talk to the hotel through the mobile application. This would help hotel owners to manage costs, employees, and customers. The solution also means that hotel managers would no longer need to be involved in the manual implementation of revenue management tasks. This data analytics approach can effectively sift through the signals detected from market variables, discover patterns and anomalies, make predictions for guest arrivals, and calculate optimum prices in real-time, as the market changes.