Cyclistic: Turning Casual Rides into Loyal Memberships
Viola Aliwarga
Data Modelling Analyst
Data Scientist
Data Analyst
Data Analysis
PostgreSQL
Python
Overview
Cyclistic operates a bike-share program in Chicago with 5,824 bikes across 692 stations, offering single-ride passes, full-day passes, and annual memberships. The objective is to convert casual riders into annual members by analyzing usage patterns and recommending targeted marketing strategies.
Step 1. Ask
The business objectives are to understand the differences in bike usage between annual members and casual riders, identify factors that could persuade casual riders to purchase annual memberships, and develop digital media strategies to encourage membership conversions.
Key stakeholders include the Cyclistic Executive Team, Marketing Director, and the Cyclistic Marketing Analytics Team.
Deliverables for this analysis include a statement of the business task, a description of data sources, documentation of data cleaning, a summary of analysis, visualizations and key findings, and recommendations based on the analysis.
Step 2. Prepare
The business task is to analyze bike usage patterns of casual riders versus annual members and recommend strategies for increasing annual memberships.
The data source consists of 52 .csv files, including trip start and end times, start and end stations, and rider type. The dataset contains 22,929,303 rows and 13 columns.
The dataset is reliable, original, comprehensive, current, and properly cited, and is released under a public license.
Step 3. Process
Tools used for this analysis include Python (Jupyter Notebook) for data consolidation and cleaning, PostgreSQL for analysis, and Tableau Public for visualizations.
The data consolidation process involved combining 52 .csv files into a single dataset using Python’s pandas library.
Data cleaning addressed missing data with KNN imputation for key columns and removed rows with minimal missing data. Additional columns were created from datetime information for further insights.
Step 4. Analysis & Visualization
Key findings from the analysis include:
Annual members use the service more frequently than casual riders.
Casual riders generally take longer rides compared to annual members.
Casual riders are more active on weekends, while annual members use the service more during weekdays.
Casual riders peak in the late afternoon, while annual members peak in the morning.
Casual riders frequently visit popular city destinations, while annual members follow regular commuting routes.
Casual riders prefer electric bikes, while annual members favor classic bikes.
Popular start times for casual riders are between 13:00 and 15:00 on weekends.
Most frequented stations by casual riders include Streeter Dr & Grand Ave, Michigan Ave & Oak St, and DuSable Lake Shore Dr & Monroe St.
Step 5. Recommendations
Focus marketing campaigns on peak times and popular routes for casual riders. Target late afternoons to early evenings when casual riders are most active.
Highlight the convenience and savings of memberships at popular stations and during peak commuting hours. Emphasize the benefits of membership for high-demand times and stations.
Develop personalized promotions based on casual riders' preferences, such as frequent use of electric bikes and specific start times. Use social media, email campaigns, and local partnerships to effectively reach target audiences.
See here for more technical aspects of the analysis.