Airbnb Booking & Review Trend Analysis

This project presents an end-to-end data warehousing and analytics solution to understand booking behavior, pricing trends, and guest engagement patterns in the short-term rental market, using Airbnb data for Denver, Colorado.

Leveraging data sourced from InsideAirbnb, the project involved extensive preprocessing, dimensional modeling, SQL-based analytics, and dynamic dashboard creation to uncover actionable insights that benefit hosts, guests, and platform stakeholders.

The core of the architecture was a star schema data warehouse built using Oracle SQL, which organized over 100,000 listings and millions of daily availability and review records into optimized fact and dimension tables. This enabled fast aggregation and trend analysis across time, location, room type, and guest interaction.

Once data was cleaned and modeled, advanced SQL analytics (including LAG, NTILE, and PERCENT_RANK) were applied to uncover value zones, review trends, availability fluctuations, and pricing dynamics. These results were visualized through a series of interactive Tableau dashboards, including heatmaps, time-series charts, and comparative bar plots.

Key insights revealed:

• Entire homes command the highest prices, while private rooms deliver better value with strong guest engagement

• Capitol Hill and Downtown are the most expensive and reviewed neighborhoods

• Availability peaks in summer months, aligning with travel seasonality

• Listings with mid-range pricing and high availability tend to receive the most reviews

This project demonstrates a full-stack analytics workflow — from ETL and data modeling to SQL-driven insight generation and visual storytelling — delivering a strategic decision-support system that could scale for real-world Airbnb hosts, analysts, or data product teams.  Tools & Technologies Used:

• Oracle SQL – ETL, dimensional modeling, analytic queries

• Python (Pandas, NumPy) – Data cleaning and feature engineering

• Tableau – Interactive dashboards and storytelling visualizations

• Excel – Early-stage data profiling and validation