FINAL REPORT:-

2022-09-16

1.Introduction 

Established in 2016, Cyclistic is a bike-share offering service located in Chicago, USA. Cyclistic currently offers more than 5,800 bicycles that are geo-tracked and locked into a network of over 690 stations across Chicago. The bikes can be unlocked from one station and returned to another station in the system anytime.

The Cyclistic team have recently concluded that annual memberships are more profitable than casual riders. Furthermore, the team have noted that while 30% of users use the bikes for their work commute, the majority of Cyclistic users ride for leisure. This report will assess how existing Cyclistic causal riders can be encouraged to convert to annual memberships.

2.process

Cyclistic have provided historical trip data to be analysed. For the purpose of this analysis, only data between April 2019 and April 2020 will be assessed. 

There are around 100,000 - 500,000 entries for each month saved under their own MS Excel CSV. Due to the large file sizes, R has been used to clean and process the large datasets. There is minimal human error and data bias since the primary, structured, historical data is taken from the bikes themselves. However, due to data privacy rules, there is no data relating to the type of user.

The data has been cleaned by way of merging all 4 datasets of twelve months into one, deleting incomplete data elements, removing test station results, removing negative ride lengths and summarizing the dataset by date and time variables and changing of names of variables or headings of tables according to our understanding. The full data cleaning process has been documented in “Data Cleaning Process”.

The cleaned dataset has been saved under the file name “all_trips_V2”.

3.Analysis

Our business task is to find out what could motivate casual riders to change to an annual subscription based on their behavior. Hence, we will look at the behaviour patterns of both casual riders and members, will find similarities and differences in how they use the service, and based on our findings will come up with hypotheses on how we can convert casual riders into members, as well as some metrics to measure the success of our hypotheses. 

The SOFTWARE used for analysis are R software and Tableau.

my analysis and visualization are as follows.

Here we can easily say that members are more active in using bikes as compare to casual riders.

These patterns may indicate a more leisure-oriented usage of the bikes by casual riders in contrast to a more “commute” usage by members. 

Looking at the hourly bike usage, it’s better to separate weekdays and weekends traffic as there are some drastic differences for both members and casual riders.

4.Sharing

Now we’re ready to use these insights to make recommendations for the marketing team.

1. Use cases

• Casual riders mostly use bike-sharing for leisure and tourism purposes and are highly active on weekends;

• Members use bike-sharing to commute to work during the week and are more active on weekdays rather than weekends.

Based on this finding, it’s worth considering to offer new types of membership focused on weekend rides, family membership (families tend to spend their weekends together), or offers created in collaboration with museums/theatres and other institutions to where casual riders travel the most.

2. Usage time

• Casual riders use bikes for much longer trips than members.

Based on this insight, we can think about offering bonuses for longer rides.