Hotel Booking Analysis: Understanding Guest Cancellations

Introduction

The hotel booking dataset consists of over 119,000 records capturing reservation details for two types of hotels — City Hotels and Resort Hotels — across a specific time period. This dataset provides insights into guest behavior, booking channels, lead times, and cancellation patterns. By analyzing these features, hoteliers and analysts can identify areas for optimizing revenue, improving customer retention, and enhancing overall operational efficiency.

Problem Statement

High cancellation rates and unpredictability in hotel bookings present a critical challenge to revenue management. A significant portion of bookings never materializes into actual stays, resulting in lost revenue opportunities, misallocation of room inventory, and operational inefficiencies. This project aims to:

1. Investigate the root causes of cancellations (e.g., lead time, market segment, deposit types).
2. Understand booking behavior patterns (e.g., group vs. individual, previous cancellations, total guests).
3. Develop actionable insights and strategies to reduce cancellation rates and improve overall hotel performance.

About the Dataset

The dataset includes 119,388 observations for bookings at two types of hotels: City Hotels and Resort Hotels. Each observation represents a booking made between July 2015 and August 2017, with key details captured across 33 columns. These columns include (but are not limited to):

• hotel: Specifies whether the booking was for a City Hotel or a Resort Hotel.
• is_canceled: Indicates whether the booking was canceled (1) or not (0).
• lead_time: Number of days between the booking date and the arrival date.
• arrival_date: Date the guest(s) arrived (or were meant to arrive) at the hotel.
• market_segment: The source of the booking, such as Online Travel Agencies (OTA), Direct, or Group bookings.
• deposit_type: Indicates whether the booking required a deposit (No Deposit, Refundable, or Non-Refundable).
• adr (Average Daily Rate): Revenue generated per room per night.

Some Terminologies in the Dataset

1. Lead Time: The number of days between the booking date and the arrival date.
2. Average Daily Rate (ADR): The average revenue earned per room per night, calculated by dividing the total lodging revenue by the number of occupied room nights.
3. Market Segment: The source of the booking, such as Direct, Online Travel Agencies (OTAs), or Group bookings.
4. Deposit Type: Indicates the type of payment or deposit made when booking (e.g., No Deposit, Refundable, Non-Refundable).
5. Special Requests: Specific requests made by the guest, such as a high floor, extra pillows, or late check-out.
6. Cancellation Rate: The percentage of total bookings that were canceled.
7. Previous Cancellations: The number of bookings a guest had canceled before their current booking.
8. Total Guests: The sum of adults, children, and babies in a single booking.