Multidimensional Analysis of Booking Data in Hospitality Industry Using Data Warehousing Techniques

Abstract

Understanding customer behaviors is essential for optimizing pricing strategies, enhancing guest experiences, and effectively meeting demand in the hospitality industry. This study presents the development of a data warehouse system designed to analyze hotel booking behaviors. Using ETL processes, reservation data from diverse sources is consolidated and standardized to enable comprehensive analysis. Then, multidimensional analyses of booking frequency and transaction value reveal key customer preferences and behavioral patterns. A real-world dataset comprising 119,390 records spanning from July 1, 2015, to August 31, 2017, was utilized to validate the system. Multidimensional analyses revealed that 70% of bookings occurred during peak seasons, with transaction values averaging 25% higher compared to off-peak periods. Additionally, customers who booked via direct channels displayed a 20% higher retention rate. The results validate the proposed system's capability to provide actionable intelligence, driving effective business strategies and supporting predictive modeling in the hospitality industry.