II.-Data-Science-Competition

Conducted Exploratory Data Analysis on a real-world but anonymized dataset from MORGEN’s hotel management system

1) Project Background

This project focuses on analyzing a comprehensive dataset from the hotel and accommodation industry, covering:

• Marketing effectiveness
• Website traffic and user behavior
• Booking engine activity
• Reservation trends and occupancy patterns

Objectives

The goal is to uncover actionable insights and develop data-driven strategies to optimize and enhance accommodation bookings by:

• Enhancing marketing efficiency to drive higher returns on investment.
• Improving website functionality and booking engine efficiency to increase user engagement.
• Identifying key booking behavior trends to capitalize on high-value opportunities.
• Providing a strategic framework for boosting hotel booking system performance and profitability in a competitive hospitality market.

2) Project Goals and Outcomes

Improving the marketing and booking efficiency of a hotel industry company, thereby contributing to increased profitability.

Key Contributions

• Data preparation and cleansing
• In-depth Exploratory Data Analysis (EDA):
  • Examination of search trends
  • Conversion rate analysis
  • Insights for revenue and yield management, as well as campaign optimization
  • Assessment of advertising spend and PPC performance
• Strategic recommendations based on insights to enhance business profitability

Datasets

• 1 CSV file – Website activity data
• 1 CSV file – Marketing channel data
• 1 CSV file – Occupancy data
• 8 CSV files – Search and booking data
• 3 Jupyter Notebook (.ipynb) files – Data preparation process
• 1 Power BI (.pbix) file – Data model
• 1 PDF file – Presentation

Goals

To process and analyze marketing and booking data across multiple hotel properties, aiming to optimize booking opportunities and enhance profitability.

3) Appendix

Methodologies

• Data Cleaning
• Data Preparation and Exploration
• Statistical Metrics Investigation
• Data Visualization
• Identifying Business Opportunities
• Advertising Cost and Medium Performance Analysis
• Presentation of Actionable Insights

IT Tools

• Python
• Power BI

Data Preprocessing and Model Building

• Data cleaning and preparation using Python
• Handling outliers and missing data using Power Query in Power BI
• Joining multiple data tables in Power BI
• Building a unified data model in Power BI
• Implementing a Star Schema by creating an independent date dimension table and linking it to fact tables
• Creating and optimizing DAX formulas, including complex functions in Power BI
• Developing interactive dashboards in Power BI

Current Capabilities & Key Features

This model is designed to automatically perform all data manipulations, analyses, and reports with a single click as soon as the next month's source tables are added to the source data folder.

This analysis was initially developed for an internal competition organized by Data36.com data science club, where it earned me the Medior Special Prize ranking.