Cyclistic Bike-Share Analysis

Google Data Analytics Capstone Project

1. Business Problem

Cyclistic aims to increase the number of annual members by converting casual riders. The objective of this analysis is to understand how casual riders and annual members use the bike-share service differently, in order to support targeted marketing strategies (Cyclistic Case Study Brief, 2023).

2. Data & Preparation

The dataset consists of Cyclistic (Divvy) trip-level data, including ride duration, station information, user type, and limited demographic variables.

Data considerations:

• Each row represents a single trip

• Key variables: tripduration, usertype, station names, birthyear

• Data contains inconsistencies such as unrealistic birth years and very short trip durations

Cleaning approach:

To ensure meaningful analysis, the following filters were applied:

• Excluded birth years earlier than 1931

• Excluded trips shorter than 60 seconds

These steps removed unrealistic or non-representative records while preserving overall usage patterns.

3. Method

The cleaned dataset was analysed using a pivot table.

Structure:

• Rows: usertype

• Values:

  • Average trip duration

  • Ride count

This approach allowed for a direct comparison of behavioural patterns between casual riders and annual members.

4. Key Findings

1. Casual riders take longer trips

Casual riders have a higher average trip duration than annual members. This suggests that casual users are more likely to use the service for leisure or occasional trips.

2. Members take shorter, more consistent trips

Annual members show shorter and more consistent ride durations, indicating routine usage such as commuting.

3. High frequency of very short trips

The dataset contains many trips of approximately one minute. These likely represent aborted rides or system-related recordings and were not treated as meaningful behaviour.

4. Demographic data is unreliable

Birth year inconsistencies limit the ability to draw strong conclusions about age-based usage patterns. Behavioural data was therefore prioritised.

5. Visual Evidence

Figure 1: Pivot Table – Ride Duration and Usage by User Type

The pivot table summarises ride count and average trip duration by user type, providing clear evidence of behavioural differences.