Mastery Project of the Data Analysis Bootcamp course @ Masterschool.

Metrocar's business model is based on a platform that connects riders with drivers through a mobile application. Metrocar acts as an intermediary between riders and drivers, providing a user-friendly platform to connect them and facilitate the ride-hailing process.

I aimed to analyse the data and make recommendations based on the following business questions to uncover valuable insights for improving specific areas of the customer funnel:

What steps of the funnel should we research and improve? Are there any specific drop-off points preventing users from completing their first ride?

Metrocar currently supports 3 different platforms: iOS, Android and web. To recommend where to focus our marketing budget for the upcoming year, what insights can we make based on the platform?

What age groups perform best at each stage of our funnel? Which age group(s) likely contain our target customers?

Surge pricing is the practice of increasing the price of goods or services when there is the greatest demand for them. If we want to adopt a price-surging strategy, what does the distribution of ride requests look like throughout the day?

What part of our funnel has the lowest conversion rate? What can we do to improve this part of the funnel?

I conducted a funnel analysis and addressed these business questions to effectively provide meaningful insights and reasonable recommendations based on evidence extracted from the data. Funnel analysis allows businesses and organizations to identify where users drop off or convert, helping them ultimately increase desired outcomes, such as sales, sign-ups, or conversions.

The project is divided into two essential parts:

SQL queries

The main SQL query summarises the count of users and the count of rides for each step of the funnel (download, sign-up, ride requested, ride accepted, ride completed, payment and review), including platform, age range and download date.

The other SQL query summarises all the rides requested at each hour of the day. 

Tableau

I imported these aggregated datasets into Tableau to create visualisations that would adequately represent the different stages of the customer funnel with labels that show the absolute number of customers at each step and the conversion rate. This helps me with addressing the business questions and getting insights across different segments such as platform and age group. 

I also created a Tableau dashboard with more granular control, empowering stakeholders to explore and interact with the data on their own.

The dashboard contains the following functionalities:

It enables users to view the user-level and the ride-level funnel. It enables users to visualise the percent of previous and percent of top conversion metrics. It enables filtering on platform, age range, and date range.

I compiled the visualisations into a Tableau story where each story point has a clear message and addresses a business question, guiding the viewer through the data journey by conveying the information in an easy and understandable way.