Hotels struggle with high cancellation rates and fluctuating booking trends.
Pricing strategies and market segmentation are often reactive, not data-driven.
Project Objective: Leverage data insights to reduce cancellation rates, optimize pricing strategies, and enhance market segmentation for improved hotel performance.
Dataset: Hotel Booking Demand Dataset
Source: Hotel Booking Demand Datasets – Published in Data in Brief (2019) by Nuno Antonio, Ana Almeida, and Luis Nunes.
Data Cleaning & Prep: Handling missing values, duplicate records, and inconsistent date formats.
Exploratory Data Analysis (EDA):
Total Bookings & Cancellations (Resort vs. City Hotels)
Revenue Analysis by Market Segment (Corporate, Groups, Online, etc.)
Monthly Trends in Guest Arrivals & Cancellations
Tools Used:
Excel (Power Pivot, Aggregate Functions, Lookup Functions, Data Visualization, and Dashboard Creation)
Insight 1: Resort Hotels Have Higher Conversion Rates but Lower Revenue Per Booking
Problem: Booking conversion rate is 72% for Resort Hotels vs. 58% for City Hotels.
Recommendation: Adjust pricing models to optimize pricing models to increase revenue per booking.
Insight 2: Cancellations Are a Major Issue for City Hotels
Problem: 42% of City Hotel bookings get canceled, compared to 28% for Resort Hotels.
Recommendation: Implement flexible yet strategic pricing models (e.g., non-refundable rates, discounts for early commitments) or loyalty-based discounts to reduce cancellations.
Insight 3: Online Travel Agencies (OTA) Drive the Majority of Revenue
Problem: Over 65% of bookings come from OTAs, leading to high commission costs.
Recommendation: Strengthen direct booking incentives to reduce reliance on third-party platforms.
What I learned:
How to analyze and compare hotel performance metrics.
How seasonal trends impact cancellations and pricing strategies.
How different booking channels affect overall revenue.
Next Steps:
Add SQL queries to analyze customer booking behavior.
Enhance dashboard with interactive Power BI visuals.
Explore time series forecasting models for demand prediction.
Take a look at the presentation below!
Click the link to view it on Google Slides for more details and a better viewing experience. 👇