Cyclistic Bike-Share

1.0 INTRODUCTION

The Cyclistic Bike-Share project serves as my capstone project for the recently completed Google Data Analytics Professional Certification course. The course comprises of 8 modules and 180+ hours of instruction designed by tech giant Google and hosted on Coursera to prepare data analytics aspirants for entry-level roles.

The course is one of the best recognised programs that introduces beginners to data analytics. The course content covers foundational concepts such as the data analysis (DA) lifecycle, which includes data collection, cleaning, preparation, processing, and visualization. The Cyclistic Bike-Share project allows me to apply my acquired knowledge to solve a business problem.

Cyclistic is a fictional company that offers bike-share services to two categories of customers — members and causal riders. Cyclistic offers a range of bike types to make bike-share more inclusive to people with disabilities and riders who cannot use a standard two-wheeled bike.

Cyclistic aims to optimise its customer base and increase profitability. The project task is to understand how casual riders and annual members use Cyclistic bikes differently and use the insight to support the marketing strategy formulation and decision-making process at Cyclistic in leading casual riders to become annual members and optimise their customer journey. This project therefore aims to answer the following questions:

· How do annual members and casual riders use Cyclistic bikes differently?

· Why would casual riders buy Cyclistic annual memberships?

· How can Cyclistic use digital media to influence casual riders to become members?

2.0 DATA SCRAPPING

The company’s historical trip data is made public by Motivate International Inc. under this license. The dataset is stored in Comma-separated Values (CSV) format. 12 CSV files representing each month within the scope of the project were downloaded from the provided web platform titled Index of Bucket. It is ROCCC i.e., reliable, original, comprehensive, current, and cited.

The available data covers the period between April 2020 — December 2022. This project however focuses on the preceding one-year span i.e., December 2021 — November 2022, as instructed in the brief. The structured data includes 13 fields that includes RideID, Rideable_type, Day_of_week, Started_at, Ended_at, Ride_length, Start_station_name, End_station_name, Start_lat, Start_lng, End_lat, End_lng, Membership_status.

3.0 DATA PREPARATION

Microsoft Excel was used as the data preparation tool. The 12 downloaded CSV files were merged into a master workbook titled Master Data. Each CSV file was transformed into a sheet in the aggregated dataset. Thus, the Master Data contains 12 sheets as a whole.

4.0 DATA PROCESSING

Microsoft PowerBI was used as the data processing tool. The Master Data was fed into the software as a data source and transformed using Power Query.

The data was checked for:

• Duplicates
• Null values
• Inconsistent formatting
• Outliers.

The data was also modified for:

• Refined field titles
• Calculated fields.

The Advanced Editor query feature, using Power Query M Code, is used to maintain consistency across the 12 sheets in the PowerQuery. These sheets are combined using the Append Queries feature into one table titled Annual Data.

5.0 DATA ANALYSIS

The analysis was performed using functions of Microsoft PowerBI including DAX code and the in-built visualization options. These functions are used to calculate:

• The average ride duration
• The maximum ride duration
• The ride volume by time of the day
• The ride volume by day of the week
• The ride volume by month
• The ride volume by bike type
• The ride volume by membership status
• The ride volume by membership status by day of the week
• The ride volume by membership status by season

6.0 DATA VISUALIZATION

The data visualization choices were hinged on the degree of importance of the metric, legibility, and minimalism. The visualization choices are detailed as follows:

• Card is used to visualize the aggregates: average ride time and the maximum ride time:

• 100% Stacked bar chart is used to visualize the ride volume by month:

• Clustered bar chart is used to visualize the ride volume by day of the week:

• Donut chart is used to visualize the rides by membership status, rides by bike type:

• Line chart is used to visualize the ride volume by the time of the day:

• Pie chart is used to visualize the casual ride volume by days of the week, member ride volume by days of the week, casual ride volume by season, member ride volume by season:

• Slicer is used to categorize the data based on membership status:

These tiles were combined to build two dashboards as seen below:

7.0 SUMMARY

The exploratory analysis highlights the differences in the patterns of patronage of Cyclistic bikes by the two customer categories. These differences relate to the ride time aggregates, the ride volume per periods, and bike preferences. The differences are captured in a 2-page report and culminates into 6 key insights:

• Extreme weather significantly and negatively affects the patronage of Cyclistic bikes.
• Casual riders’ patronage is highest in the summer and spring seasons.
• Members’ patronage is highest in the summer and winter seasons.
• Casual riders use Cyclistic bikes more than members during the weekends, especially on Saturday.
• Members use Cyclistic bikes more during the week, especially at 8AM and 5PM.
• Electric bike and classic bike are the preferred bike types in that order by both user categories.

8.0 CUSTOMER PREFERENCE & THEMES

• Cyclistic customers prefer the membership model to casual rides.
• Cyclistic member ride volume exceeds casual rides by 18%.
• Members demand for Cyclistic bikes more during the week (75%).
• Casual riders demand for Cyclistic bikes more during the week (63%).
• Extreme weather significantly and negatively affects the patronage of Cyclistic bikes:
• Casual rides are more popular in the summer (48%), then in the spring (21%).
• Member rides are more popular in the summer (37%), then in the winter (28%).
• Cyclistic members and casual riders show similar demand volume in both spring and autumn.
• Members use Cyclistic bikes to commute to and from work, whereas casual riders have alternative use for the bikes.
• The peak periods of members’ demand for Cyclistic bikes are 8AM and 5PM during the week.
• The peak period of casual riders’ demand for Cyclistic bikes is 5PM during the weekend.
• Electric bikes and classic bikes are the preferred by the bike types:
• Casual riders are responsible for the demand for docked bikes, although the docked bikes is the least preferred of the three bike types.

9.0 CONCLUSIONS

· Members rely on Cyclistic bikes to commute to and from their offices.

· Causal riders use Cyclistic bikes for alternative mobility purposes.

10.0 RECOMMENDATIONS

· Market discounted offers for members during the weekends to encourage casual riders to subscribe.

· Market discounted offers for members especially during the summer season to encourage casual riders to subscribe.

· Conduct further analysis with expansive data on Cyclistic customers to better understand their social, behavioural, and cultural preferences.

‍