HOW DOES A BIKE-SHARE NAVIGATE SPEEDY SUCCESS?

Marketing campaign strategy for a bike-sharing company.

It’s been a month since I started working on a case study which I got through the Google Data Analytics Course. Through this case study, I got the chance to showcase my skills and learnings from this course.

I chose the case study: Cyclistic. In this case study, I am working as a junior data analyst in the marketing team of a fictional company Cyclistic which is based in Chicago. Their goal is to start a campaign through which they can convert casual customers to members. My job was to find the relations and trends of both types of customers and help the company to design the marketing campaign. So I started with understanding the clear requirement of the project. Then I mined the data and cleaned it. After cleaning the data I analyzed the data to find the possible trends and ultimately converted my readings into visual form.

Data Mining:

The data has been made available by Motivate International Inc. under this license(https://www.divvybikes.com/data-license-agreement.) This is public data that can be used to explore how different customer types are using Cyclistic bikes.

Data cleaning:

The company provided the12 month data (April 2019- March 2020) quarterly. The data consists of 13 fields which consisted of various records of the ride of each month. I started cleaning the data through excel. There I used different tools and filtering techniques to find the bland spaces and uneven data. Since the data was too big for excel, I used SQL to remove and join columns. I performed the SQL cleaning in Microsoft Sequel Server Management Studio.

Analyzing the data:

After cleaning the data, I transferred it back to excel to analyze. I used the pivot function to do calculations and make summaries. In the calculation section, I found the total travel time of each ride, the mean of total travel time, found the weekdays and months, and many more.

After doing the majority of calculations, I moved to R. why R? I want my whole process to be in one place where I can perform every process smoothly. Also visualizing and doing the calculation in the big data sets is easier in R. I have attached the snapshot of my codes in R.

Findings:

After my deep analysis, I found out different conclusions.

  1. Casual users are more than the yearly member.
  2. The casual users use the cycles /service mostly on weekends i.e. Saturday and Sunday.
  3. The yearly members use the service mainly on weekdays. This shows that most members use bikes to commute to work.

In pic1 we can see how the yearly member's graph is varying on week days. We can see that the value rose on weekdays and declined during the weekends.

In Pic2. we can clearly see that the casual members tend to use the service mostly on weekends.

In pic3. we can see that most casual users of the service belonged to Street er Dr & Grand Ave.

Conclusion:

From the above findings, suggestions I would like to give are :

  1. The company should form a new membership plan where they cover the weekends as casual riders are mostly on weekends.
  2. The campaign should be more around the Street er Dr & Grand Ave as there are more casual riders in that area.