Introduction

Cyclistic: A bike-share program that features more than 5,800 bicycles and 600 docking stations. Cyclistic sets itself apart by also offering reclining bikes, hand tricycles, and cargo bikes, making bike-share more inclusive to people with disabilities and riders who can’t use a standard two-wheeled bike. Most riders opt for traditional bikes; about 8% of riders use the assistive options. Cyclistic users are more likely to ride for leisure, but about 30% use them to commute to work each day.

This project seeks to understand how casual riders and annual members use Cyclistic bikes differently. Insights from the data analysis will help the marketing team design marketing strategies to convert casual riders into annual members.

Problem statement

This project analyzes Cyclistic’s bike-trip historical data to identify how casual riders and annual members use Cyclistic bikes differently. The marketing team believes that maximizing annual members is key to future growth. Converting casual riders into annual members would help achieve that since casual riders are already aware of the Cyclistic program and have chosen Cyclistic for their mobility needs.

Business Questions

Key Business Question: How do Cyclistic’s casual riders and annual members use Cyclistic’s bikes differently?

The following questions will guide our analysis and the insights will help us answer the business question.

1. What are the differences in ride duration between casual riders and annual members?

2. How do the usage patterns vary by day of the week between casual riders and annual members?

3. How do the usage patterns vary by time of day between casual riders and annual members?

4. What are the seasonal trends in bike usage for casual riders compared to annual members?

5. What types of bikes (rideable type) are preferred by casual riders compared to annual members?

Data Source

Cyclistic’s historical trip data was used to analyze and identify trends. The previous 12 months (March 2022 – February 2023) of trip data were downloaded from Index of bucket "divvy-tripdata". The data was made available by Motivate International Inc. under this Data License Agreement | Divvy Bikes.

The datasets were downloaded and stored in a local folder called Cyclistic Data.

Skills demonstrated.

• Data cleaning

• Data Visualization

• Python

• Power BI

• Data Analysis

• Data Management

• Source Control (Git)

• Problem-solving

• Critical Thinking

Data Transformation/Cleaning

The dataset was duplicated and stored in a folder. A copy was loaded into Python’s Pandas package for pre-cleaning and transformation.

The following steps were taken to clean and transform the data for analysis.

• The monthly datasets were combined into a single dataset.

• The appropriate data types for each column were specified.

• Year, month, weekday, and hour columns were extracted from the date columns.

• A calculated column (ride length) was created.

• Columns were renamed.

• Text columns were properly formatted by using the trim and clean functions.

• Duplicate rows were removed.

Data Analysis

Following the data cleaning and transformation using Python, the dataset was exported as CSV and loaded into Power BI for analysis via Power Query. Power Query allows users to extract, transform, and load (ETL) data from different sources into Power BI or Excel.

In Power BI, measures were created to aid data analysis and faster charting techniques.

Exploratory Data Analysis

Combined, the data contained 5.8 million total rides from March 2022 to February 2023. 3.5 million rides from annual members and 2.4 million rides from casual rides. The maximum ride length was 0 minutes, while the maximum ride length was 1068 minutes.

Answering the Business Questions

What are the differences in ride duration between casual riders and annual members?

Insight: Despite having more annual members (40.6%), Cyclistic casual riders had a longer average ride length (28.9 minutes) than annual members.

How do the usage patterns vary by day of the week between casual riders and annual members?

Insight: Casual riders tend to use Cyclistic’s bikes during weekends (Friday, Saturday, Sunday) compared to annual members whose usage are higher during weekdays.

This might be because annual members use bikes to commute to work, while casual members use bikes more for leisure during weekends.

How do the usage patterns vary by time of day between casual riders and annual members?

Insight: While both types of rides peak by 5 pm, the annual members have two peak usage times of day – 8 am and 5 pm. This further establishes the fact that annual members use Cyclistic bikes for commuting to and from work.

However, as seen from the line chart above, the same usage pattern seems to be the case for some casual users. This could be casual users who also use Cyclistic bikes to commute to and from work but are not annual members.

What are the seasonal trends in bike usage for casual riders compared to annual members?

Insights: Overall, both casual riders and annual members use the bikes more in summer than other seasons of the year. As presented in the chart below, both ride-types tend to have longer ride lengths in summer, closely followed by spring.

What types of bikes (rideable type) are preferred by casual riders compared to annual members?

Insight: Cyclistic’s electric bike is the most preferred bike type among all riders. However, annual members have a slightly higher preference for classic bikes over electric bikes and do not even use the docked bike at all.

Casual riders, on the other hand, prefer the electric bike to the classic bike. This preference accounts for the longer ride length in casual riders. It also suggests that casual riders travel longer distances compared to annual members. They also use the docked bike, unlike annual members. Further analysis

Visualization

Summary of Findings

Members who pay annually tend to utilize Cyclistic bikes more during the weekdays, whereas those who are casual riders prefer these bikes for the weekends, from Friday to Sunday. This indicates that annual members likely use Cyclistic bikes for their daily commute, whereas casual riders probably use them for recreational activities or for specific needs.

Casual riders tend to go on longer rides than annual members. They are also the only ones who have their bikes docked. Given the distance of their commutes, they tend to favor electric bikes. The shorter commute times of annual members likely lead them to prefer Cyclistic’s traditional, bike-friendly models.

The time of day when most people use Cyclistic bikes is 5 pm for both groups. However, annual members also experience a second peak in usage at 8 am. This supports our hypothesis that annual members might use Cyclistic bikes for their daily commute or for other short trips throughout the day.

Casual riders also exhibit a similar trend in terms of the times of day when they use Cyclistic bikes the most. This suggests that there are casual riders who use these bikes for their daily commutes.

To sum up, annual members use Cyclistic bikes more often but for shorter distances during the week, whereas casual riders take longer rides but do so less frequently than annual members.

Recommendation

Cyclistic should target their ads to casual members who use their bikes for commuting to or from work. This category of riders are frequent users and have a need for using the bikes throughout the year. These users won’t mind paying for an annual membership at a relatively discounted amount.

Cyclistic can offer Casual members who use their bikes for a longer time during rides but less frequently a monthly or quarterly membership option. The usage pattern observed by these riders suggests that they use Cyclistic bikes for specific purposes which could be for leisure since the location is a coastal area known for its tourist attractions.

Finally, the marketing team could organize summer campaigns as this is the peak usage time of the year. Users can use their membership status to access perks and special privileges during peak periods and bike shortages.

Limitation

The data given is insufficient to draw specific conclusions. Therefore, additional details are required to verify our assertions. Information such as the reason for the ride, details regarding the cost structure of the rides, and the unique count of customers by type of rider will aid in revealing further patterns of usage among riders.

Appendix

Cover Photo by Leonardo Zorzi on Unsplash