This repository contains the final assessment project for the Data Analysis Program at MasterSchool. The project aimed to analyse the customer funnel of Metrocar, a ride-sharing app similar to Uber/Lyft, using SQL for data querying and Tableau for data visualization.
The project focused on conducting a comprehensive funnel analysis to identify areas for improvement in Metrocar's customer journey. It aimed to address specific business questions, providing insights for optimization and recommendations based on data-driven findings.
For all details about the project, I kindly invite you to go through the report available here
The project aimed to conduct funnel analysis on Metrocar's customer journey to uncover drop-off points and recommend strategies for improvement. Funnel analysis helps in identifying where users exit the funnel, aiding in enhancing desired outcomes like sales and conversions.
Metrocar serves as a ride-sharing intermediary, connecting riders with drivers through a user-friendly mobile application. The customer funnel includes stages such as app download, signup, ride request, driver acceptance, ride, payment, and review. Drop-offs at each stage are analysed to identify optimization opportunities.
The analysis addressed the following key business questions:
- Optimizing the Funnel Stages
- Platform-Based Marketing Insights
- Age Group Performance Analysis
- Surge Pricing Strategy
- Lowest Conversion Rate Identification
This analysis has identified the needs for the company to balance supply and demand during peak hours, the shortage in those hours currently results in a high cancellation rate because many customers don’t want to wait a long time. The suggestions are to implement a surge price technique and ride-sharing option. The last option will also help reduce CO2 emissions by picking up more users during the same journey. The analysis has also identified that the vast majority of the users have an iOS device and fall into the age group between 35 and 44 years old.
During the analysis, I have also identified the needs to understand the point of view of our customer and I have tried to gain an understanding of the customer reviews by assessing the sentiment score and identify the most common adjectives and noun used in the reviews. I have only analysed the reviews with a 1 start score because with my resources I couldn’t extend the resources to hundreds thousands of reviews to avoid system crashes and also because I didn’t want to get out of the main goal of the project. My point is that without understanding the customer opinion it is really difficult to improve the service efficiently and from the quick insight I gain from this extra analysis I have been able to identify things that may slow down the company growth. For this reason, an in-depth sentiment analysis will definitely help to take better decision in the future and allocate the budget even more efficiently.