Case Study: How Does Cyclistic Bike-Share Navigate Speedy Success?

Background Information

My Role: Junior data analyst at the marketing analyst team

Situation: Cyclistic financial analysts concluded that annual members are more profitable than casual riders

Objective: Analyze datasets and share findings and recommendations

Ask Phase

(Define the business problem and confirm stakeholders expectations)

Business Task: design marketing strategies aimed at converting casual riders into annual members

Key Question: How do annual members and causal riders use Cyclistic bikes differently?

Key Stakeholders:

• Lily Moreno (Director of Marketing & My Manager)
• Cyclistic marketing analytics team
• Executive team

Prepare Phase

(Collect and store data for analysis)

Data Source: Motivate International Inc.

Data Integrity: https://ride.divvybikes.com/data-license-agreement

Data Type: CSV file and Public data

Data Description:

• Total of 12 datasets (View on Kaggle: link)
• Time period: April 2021 to March 2022 (12 months)

Data Variables:

• ride_id: a unique ID for each bike rider
• rideable_type: type of bike used (docked, classic, electric bike)
• started_at: date and time the bike is checked out
• end_at: date and time the bike is checked in
• start_station_name: name of station the start of the ride
• start_station_id: a unique ID for each start station
• end_station_name: name of station the end of the ride
• end_station_id: a unique ID for each end station
• start_lag: latitude of the starting point
• start_lng: longitude of the starting point
• end_lag: latitude of the ending point
• end_lng: longitude of the ending point
• member_causal: type of bike user (casual / member)

Process Phase

(Clean and transform data to ensure integrity)

• Data Processing (line 7 - 47 in R Markdown)
• Data Cleaning (line 49 - 83 in R Markdown)

Analyze Phase

(Use data analysis tools to draw conclusions)

• Data Aggregation (e.g. MIN / MAX / AVG / MEDIAIN --> line 85 - 98 in R Markdown)

Share Phase

(Interpret and communicate results to others to make data-driven decisions)

• Look at the graphs here: data visualizations
• Total Number of Rides for Each User Type
• Bike Preference by User Type
• Number of Rides per Month for Each User Type
• Number of Rides per Day of Week for Each User Type
• Total Bike Ride Duration in Hours for Each User Type
• Average Bike Ride Duration by Month for Each User Type
• Average Bike Ride Duration by Day of Week for Each User Type

R Code --> line 100 to 234 in R Markdown

Act Phase

(Put your insights to work to solve the original problem)

Conclusions from the data and visualizations

1. Casual riders and annual riders account for 44% and 66% of the total number of rides respectively

2. Classic bikes are most prefered for both user types, and only docked bikes are rode by casual riders

3. Both causal and annual riders rode the most trips between July 2021 to September 2021

4. Both casual and annual riders rode most on weekends.

5. Casual riders rode 1.95 hours more than annual riders

6. Both casual and annual riders rode the longest duration between April 2021 to July 2021

7. Casual riders rode the longest on Monday, Saturday, and Sunday while annual riders ride the longest from Friday to Sunday

Recommendations

1. Advertise the marketing campaign in warmer months (July to September)

2. Introduce reward program for more and longer rides (1.5x points for annaul riders)

3. Increase the bike rental fees on weekends for both riders

4. Provide more membership benefits to annual riders

5. Improve the design and quality of bikes towards annual riders

Additional Data for Further Analysis

1. Usage of the bikes (e.g. transportation / leisure )

2. Inclusion of demographics data

3. Churn rate for regression analysis