GoFast Scooter Rental Data Analysis

Project Overview
With the rising popularity of electric scooter rental services, it's essential to understand user behavior and the key factors that influence customer choices. This project analyzes trip and user data from the GoFast scooter rental service across multiple cities.
Our goal is to uncover insights that will help optimize the business model, increase revenue, and improve the overall user experience. The analysis includes hypothesis testing related to the Ultra subscription and identifying factors that impact trip frequency and cost.

Research Goal
To analyze user behavior within the GoFast service and identify factors influencing the choice of the Ultra subscription and trip cost. This includes validating business-focused hypotheses aimed at increasing profitability and enhancing user satisfaction.

Project Workflow
Data Loading
Import trip and user datasets

Data Preprocessing
Clean and format data for analysis (handling missing values, duplicates, date formats, etc)

Exploratory Data Analysis (EDA)
Analyze user behavior, trip frequency, distribution of subscribers vs non-subscribers

Data Merging
Combine trip and user datasets for deeper analysis

Revenue Calculation
Estimate total revenue from different user groups

Hypothesis Testing
Evaluate statistical hypotheses to determine if the Ultra subscription significantly affects user behavior
Compare revenue between subscriber and non-subscriber groups

Final Summary
Key findings and visual insights
Impact of Ultra subscription
Business recommendations
How to increase subscription conversion
Pricing optimization
Suggested next steps

Expected Business Value
The results of this analysis will support data-driven decisions for GoFast's strategic growth and service improvement
Better understanding of user segments
Insights into what drives loyalty and repeat use
Concrete suggestions to optimize pricing and marketing

Tools and Libraries Used
pandas, numpy – data handling
matplotlib, seaborn – data visualization
scipy – statistical testing
jupyter notebook – interactive exploration