1. The Ask: Business Task

The central goal of this analysis is to answer a key question for the Cyclistic marketing team: How do annual members and casual riders use Cyclistic bikes differently?

The director of marketing, Lily Moreno, believes that future success depends on maximizing annual memberships, as finance analysts have determined members are more profitable than casual riders. By understanding the behavioral differences between these two user groups, the marketing team can design a targeted strategy to convert casual riders into paying annual members. The final recommendations will be presented to the Cyclistic executive team for approval.

2. Prepare & Process: Data Cleaning and Manipulation

To answer the business question, we used historical trip data for Q1 2019 and Q1 2020, as recommended for use in RStudio. The following steps were taken to prepare the raw data for analysis.

Data Aggregation and Standardization

The data was initially spread across two separate files with inconsistent column naming and data formats.

  1. Loading Data: Both Divvy_Trips_2019_Q1.csv and Divvy_Trips_2020_Q1.csv were loaded into R as separate data frames.
  2. Standardizing Column Names: The 2019 dataset used different column names than the 2020 data. Key columns were renamed for consistency (e.g., trip_id to ride_id, start_time to started_at, and end_time to ended_at).
  3. Standardizing Categorical Data: The user type column contained different values across the datasets (“Customer” and “Subscriber”). These were recoded into the more consistent categories of “casual” and “member,” respectively.
  4. Merging DataFrames: After standardization, the two data frames (trip_19_new and trip_20) were merged into a single comprehensive data frame named all_q1_trips to allow for unified analysis.

Feature Engineering

Two new columns were created to facilitate the analysis:

  • tripduration: Calculated the duration of each ride in minutes by subtracting the started_at timestamp from the ended_at timestamp. This is essential for comparing ride lengths.
  • day_of_week: Extracted the day of the week from the started_at timestamp to analyze usage patterns across the week.

This documented cleaning process ensures the data is accurate, consistent, and ready for analysis.

3. Analyze & Share: Key Findings and Visualizations

The primary analysis focused on comparing the average trip duration for casual riders and members for each day of the week. The results reveal a clear and significant difference in behavior between the two groups.

Average Trip Duration by Day of the Week

Average Trip Duration by Day of the Week

Average trip duration for casual riders vs. members, broken down by day of the week.

Average trip duration for casual riders vs. members, broken down by day of the week.

Key Findings from the Visualization:

  • Casual Riders Take Longer Trips: Across every single day of the week, casual riders have a significantly longer average trip duration than members. This suggests that members primarily use the service for shorter, more functional trips like commuting, while casual riders engage in longer, likely recreational, rides.
  • A Massive Peak on Thursday: There is a dramatic spike in the average ride duration for casual riders on Thursdays, which far exceeds any other day for either user group.
  • Weekend Usage: While members’ usage remains stable, casual riders also show elevated average ride times during the weekend (Saturday and Sunday), further supporting the leisure-use hypothesis.

4. Act: Conclusion & Recommendations

The data tells a clear story: casual riders and members use the Cyclistic service in fundamentally different ways. Members are consistent and efficient, while casual riders are situational and take much longer trips, especially on Thursdays and weekends. This insight directly informs a targeted marketing strategy.

Results from Pilot Campaign

Based on the insight that casual rider engagement peaked on Thursdays, the marketing team decided to spend heavily on a pilot campaign engaging this audience. The results were outstanding: the team achieved 60% more subscriber conversions from this targeted Thursday audience compared with the overall casual customers engaged during all other days of the week.

Top 3 Recommendations

Based on the analysis and the successful pilot, here are the top three recommendations to convert casual riders into members:

  1. Launch a “Thursday Explorer Pass” Campaign: Capitalize on the massive Thursday spike by creating targeted digital media ads and email promotions. Offer casual riders who rent on a Thursday a special introductory discount on an annual membership. This leverages their peak day of engagement.
  2. Create Leisure-Focused Membership Tiers: Since casual riders take longer trips, especially on weekends, a standard commuter-focused membership may not be appealing. Introduce a “Weekend Warrior” or “Explorer” membership that offers benefits for longer rides or a set number of recreational hours per month.
  3. Implement In-App Targeted Promotions: Use the Cyclistic app to send push notifications to casual riders during a long ride. A message like, “Enjoying your trip? Rides like this are cheaper with a membership. Tap here to learn more!” can effectively target users at their highest moment of engagement and demonstrate the value of a membership.

By implementing these data-driven strategies, Cyclistic can effectively convert high-value casual riders into profitable annual members, securing future growth for the company.