This case study analyzes hotel booking data to identify the key drivers of cancellations and deliver actionable recommendations for revenue management.
Across both hotel types, approximately 37% of bookings were canceled—a significant challenge for forecasting and operations.
The analysis revealed that lead time is the strongest predictor of cancellations. Bookings made far in advance are nearly four times more likely to be canceled than last-minute reservations.
Deposit type also plays a critical role. Non-refundable bookings show unexpectedly high cancellation rates, while refundable and no-deposit bookings are more stable.
Customer type and seasonality further influence reliability. Transient travelers and peak summer months drive the highest volatility.
Finally, higher ADR bookings are more sensitive to cancellation risk, especially when combined with long lead times.
Based on these insights, the study recommends targeted policy adjustments—including partial deposits for long lead-time bookings, seasonal policy changes, and loyalty incentives for transient customers—to reduce cancellations and improve revenue predictability.
This case study is part of my Google Data Analytics Capstone. I analyzed hotel booking cancellations using real-world data from 2015–2017 to uncover the main drivers behind cancellations. The findings are turned into clear, actionable recommendations that hotels can use to reduce cancellations, improve revenue forecasting, and optimize customer experience.
Scope: The dataset includes both Resort and City hotels, offering a broad view of booking behaviors in the hospitality sector.