Case Study: How Does a Bike-Share Navigate Speedy Success?

Capstone project example of the Google Data Analytics Certificate.

Scenario

You are a junior data analyst working in the marketing analyst team at Cyclist, a bike-share company in Chicago. The director of marketing believes the company’s future success depends on maximizing the number of annual memberships.Therefore, your team wants to understand how casual riders and annual members use Cyclist bikes differently. From these insights, your team will design a new marketing strategy to convert casual riders into annual members. But first, Cyclist executives must approve your recommendations, so they must be backed up with compelling data insights and professional data visualizations.

Business task

• Question: How do annual members and casual riders use Cyclistic bikes differently?
• Key stakeholders: Cyclist executive team
• How can your insights drive business decisions? By getting insights on the different user behaviors of annual members and casual riders, marketing campaigns or features of the subscription or bike offer could be adapted to the needs of casual riders to motivate them to subscribe to an annual membership

Preparation and processing of data

For details see file prepare-process.

Description of data sources used

The data used is usage data from the fictional company Cyclistic from the last 12 months. There is one file for each month, from May 2021 to April 2022. Each file contains 13 columns (ride id, rideable type, member type, start and end time, start and end station and their location). The monthly average of bike rides is 479,795, ranging from 103,770 to 822,410 rides per month.

Licensing

The data comes with a license, that states that the data is available to the public, and that it can be included as source material in analyses, reports or studies published for non-commercial purposes. This is applies to the case of this analysis.

Data cleaning

I merged all 12 files to one and saved it as a new file in the folder “output-data”. The original files are in “raw-data”. I backed up the original data, chekced for missing values, irrelevant data, checked if strings are consistent and meaningful, misfielded values, mismatched data types, irregularities, and created new columns that could be useful for analysis. For details, please see here.

Analysis and visualizations

For details see file analyze-share.

Insights and recommendations with regards to the business task

The question was “How do annual members and casual riders use Cyclistic bikes differently?”. Here are some insights:

• There are more rides from annual members (3,221,055), than from casual users (2,536,267). What we don’t know is the number of rides by person.
• Annual members use only classic and electric bikes, not docked bikes. Casual riders use all three types. This might be due to restrictions in docked bike usage for members, but I would have to ask the stakeholder to find this out.
• The most popular bike type in general are classic bikes, but the percentage is greater in annual members. Maybe annual members are more sportive, maybe they rather choose the classic bike because it is cheaper, or they pick classic bikes because they go on shorter rides.

There could be many different reasons.

• Casual riders tend to go on longer rides than annual members, and they rather ride in the weekend, compared to annual members who ride more on weekdays.
• So, how to make the annual membership more attractive for casual riders?
  • Maybe allow docked bikes for annual members, some casual members might prefer them (if my theory is true that they are not available for annual members)
  • Make the annual membership cheaper compared to the renting bikes casually.
  • Put advertisements in workplaces, because annual membership seems to be more popular amongst people who use it to go to work. But that would be for new members, not casual riders.

Reflection on the case study

This case study felt like a really good practice and was fun! It was helpful that there were lots of guiding questions and suggestions for steps, because just setting everything up and getting it running is hard enough if one is not used to doing it. I feel no i actually internalized what I learned in theory, by putting it into practice, and I feel more confident. It was a lot harder (of course) than in the classes, because there was much more liberty and no ‘one correct answer’.

I could have spent a lot more time on this, and I think my markdown files are too full and messy to be nice and readable for stakeholders, and don’t have enough interpretational text. They are rather a help / documentation for analysts at this point.

But overall I’m happy with the learning experience!