Analyzed 5+ million bike-share trips to identify behavioral differences between casual riders and annual members, developing data-driven marketing strategies to increase conversions.
- Casual riders are leisure-focused: 41% weekend usage, 28-minute average rides, peak on Saturdays
- Members are commute-focused: 73% weekday usage, 14-minute average rides, peak on Wednesdays
- Seasonal impact: Summer has 8.5x more rides than winter; casual riders comprise 52% of summer rides
- Weekend-to-Weekday Membership: Target high-volume weekend casual riders with a transitional membership tier
- Summer Conversion Campaign: Capitalize on peak season engagement (July = 54% casual riders)
- Commuter Trial Program: Convert regular casual commuters with targeted cost-savings messaging
- Data Processing: Combined and cleaned 12 months of trip data (5M+ records)
- Tools: Python (pandas, matplotlib, seaborn), Jupyter Notebooks
- Analysis: Temporal patterns, geographic clustering, user behavior segmentation
cyclistic-analysis/
├── data/ # Data files (not included in repo)
├── notebooks/ # Analysis notebooks
│ ├── 01_data_exploration.ipynb
│ ├── 02_exploratory_analysis.ipynb
│ └── 03_insights_and_recommendations.ipynb
├── visualizations/ # Charts and dashboard
└── README.md
Expected conversion rate of 10-15% could yield 230,000-350,000 new annual members.
Due to file size limitations, the raw data files are not included in this repository.
To reproduce this analysis:
- Download the original Divvy trip data from [link to data source]
- Place the CSV files in a
data/raw/folder - Run the notebooks in order (01, 02, 03)
The dataset used includes 12 months of trip data from October 2020 to September 2021.