Hotel Reservation Cancellation Risk Data

Columns

• hotel: Type of hotel chain, identifying properties as either Resort or City Hotel.
• is_canceled: Binary field indicating if the booking was canceled (1) or confirmed (0). The mean cancellation rate is 37%.
• lead_time: The duration, measured in days, between the reservation date and the scheduled arrival date.
• arrival_date_year, arrival_date_month, arrival_date_week_number, arrival_date_day_of_month: Components detailing the intended arrival date.
• stays_in_weekend_nights: The count of weekend nights (Saturday or Sunday) booked by the guest.
• stays_in_week_nights: The count of weekday nights (Monday to Friday) booked by the guest.
• adults, children, babies: Numerical counts of guests across different age categories for the booking.
• meal: Specifies the type of meal package reserved (e.g., BB, HB).
• country: Guest's country of origin, presented using ISO 3155-3:2013 codes. Portugal (PRT) accounts for 41% of records.
• market_segment: Designation of the market segment (e.g., Online Travel Agency, Offline TA/TO).
• distribution_channel: The channel through which the reservation was distributed.
• is_repeated_guest: Indicates if the guest is a recurring customer (1) or a new customer (0).
• previous_cancellations: The number of prior reservations canceled by the customer.
• previous_bookings_not_canceled: The number of prior reservations confirmed by the customer.
• reserved_room_type, assigned_room_type: Codes representing the type of room reserved and the type of room actually assigned.
• booking_changes: The total number of modifications made to the reservation since its creation.
• agent, company: Identification numbers for the travel agent or the company responsible for payment.
• days_in_waiting_list: The duration, in days, the booking spent on a waiting list before being confirmed.
• customer_type: Category of the reservation (e.g., Transient, Group, Contract).
• adr: Average Daily Rate, calculated by dividing the sum of all accommodation transactions by the total number of nights stayed.
• required_car_parking_spaces: The number of parking spaces requested by the client.
• total_of_special_requests: The total number of special requests made (e.g., specific bed type or floor).
• reservation_status: The final status of the booking, categorized as Canceled, Check-Out, or No-Show.
• reservation_status_date: The date when the reservation status was last updated.

Distribution

Usage

• Financial Modelling: Calculating estimated annual savings if business contracts were renegotiated to link payments only to materialized (non-canceled) reservations.
• Risk Assessment: Testing strategic hypotheses, such as whether longer lead times, lower Average Daily Rates (adr), or the absence of special requests correlate with a higher probability of cancellation.
• Data Development: Practicing fundamental SQL commands (SELECT, FROM, WHERE, GROUP BY, COUNT, AVG) using database environments like Google BigQuery.
• Business Intelligence: Creating visual reports and dashboards using tools like Power BI to communicate actionable findings to hotel management.

Coverage

License

Who Can Use It

• Data Analysts and Scientists: For building predictive models focused on guest churn and booking reliability.
• Business Executives and Financial Managers: To understand revenue risk, evaluate contract profitability, and derive actionable business decisions regarding operational strategy.
• Students and Trainees: For learning core data management skills, including data cleaning, filtering, summarizing, and visualization using data analysis and BI tools.

Dataset Name Suggestions

• Hotel Reservation Cancellation Risk Data
• Global Hospitality Booking Analysis (2015–2017)
• Predicting Reservation Reliability
• Hotel Contract and Cancellation Metrics

Attributes