The MetroCar project focuses on analyzing user interactions within the MetroCar ride-hailing platform, evaluating critical areas for improvement, including user signups, ride requests, cancellations, and driver behaviors. The aim is to identify bottlenecks in the user funnel and enhance both driver and rider satisfaction through data-driven insights.
- Analyze key metrics from MetroCar, including user ride requests, cancellations, and ratings.
- Evaluate the efficiency of the platform's service, identifying critical drop-off points.
- Provide actionable insights to improve service performance and user retention.
This project utilizes SQL and Python to extract and analyze data from various MetroCar datasets, including ride requests, transactions, and reviews. Key steps include:
- Extracting data from a PostgreSQL database.
- Performing data cleaning and processing with Pandas.
- Conducting visualizations to highlight trends in user behavior.
- Analyzing relationships between peak hours and cancellation rates.
- A 50.2% drop-off between ride requests and completed rides highlights an area for service improvement.
- Ride cancellations were highly correlated with peak hours, where almost 42% of demands went unfulfilled, primarily due to driver availability.
- Average ratings suggest room for improving the user experience, with an average rating of 3.06.
Access the Google Colab notebook for the MetroCar project here.
Watch the project presentation video here.
- Improve Peak Hour Coverage: Increase driver availability during peak hours to reduce ride cancellations.
- Optimize the Request Process: Reduce average waiting time by optimizing the algorithm for matching riders and drivers.
- User Experience Improvements: Implement features such as real-time updates and better communication to enhance user satisfaction.