🚴 Bike Ride Data Analysis using SQL

📌 Project Overview

This project focuses on analyzing a bike-sharing dataset using SQL to answer practical, business-oriented questions.

The goal is to move beyond basic query writing and demonstrate analytical thinking, data validation, and insight generation.

The analysis simulates how a data analyst would explore ride behavior, station usage, and user growth patterns in a real-world scenario.

📂 Dataset Description

The project uses three CSV files:

1. rides.csv

• Transactional ride-level data.
• ride_id – Unique ride identifier
• user_id – User who took the ride
• start_station_id, end_station_id – Station identifiers
• start_time, end_time – Ride timestamps
• distance_km – Distance traveled (in kilometers)

2. users.csv

• User-level information.
• user_id – Unique user identifier
• username
• age
• membership_level – Casual / Membership tier
• created_at – Account creation date

3. stations.csv

• Station master data.
• station_id – Unique station identifier
• station_name
• capacity
• latitude, longitude

🛠 Tools & Technologies

1. MySQL
2. SQL concepts used:

• JOIN
• CASE
• Aggregate functions
• Date & time functions
• Common Table Expressions (CTEs)
• Window functions (LAG)

🔍 Analysis Performed

1. Data Exploration

  Total records in each table

  Ride distance and duration summary statistics

2. Data Quality Checks

  Detection of very short rides (possible false starts)

  Zero-distance rides

3. User Segmentation Analysis

  Ride count, average distance, and duration by membership level

4. Time-Based Analysis

  Peak riding hours during the day

5. Station Performance

  Most popular starting stations

  Net bike flow per station (arrivals vs departures)

6. Ride Categorization

  Classification of rides into:

  Short (<10 minutes)

  Medium (11–30 minutes)

  Long (>30 minutes)

7. User Growth Analysis

  > Month-over-month user signup growth using window functions