Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/panagiotischaviaropoulos/google-data-analytics-case-study


https://github.com/panagiotischaviaropoulos/google-data-analytics-case-study

bigquery data-visualization sql

Last synced: 9 days ago
JSON representation

Awesome Lists containing this project

README 

        
# Google-data-analytics-case-study

### Summary

In this project I am working for a fictional company named Cyclistic along with some other key team members, and I will be performing several junior data analyst real life tasks to adress the business problem.

In order to answer the business questions, I will utilize the six steps of the data analysis process: Ask, Prepare, Process, Analyze, Share and Act.

### Company information

In 2016, Cyclistic launched a successful bike-share offering. Since then, the program has grown to a fleet of 5,824 bicycles that are geotracked and locked into a network of 692 stations across Chicago. 

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. The majority of 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 the bikes to commute to work each day.



Customers who purchase single-ride or full-day passes are referred to as casual riders. Customers who purchase annual memberships are Cyclistic members. Cyclistic’s finance analysts have concluded that annual members are much more profitable than casual riders. The director of marketing (Moreno) has set a clear goal: Design marketing strategies aimed at converting casual riders into annual members. In order to do that, however, the team needs to better understand how annual members and casual riders differ.



### Scenario

Let's assume I am a junior data analyst working on the marketing analyst team at Cyclistic. The director of marketing believes the company’s future success depends on maximizing the number of annual memberships. Therefore, our team wants to understand how casual riders and annual members use Cyclistic bikes differently. From these insights, the team will design a new marketing strategy to convert casual riders into annual members. But first, Cyclistic executives must approve the recommendations, so they must be backed up with compelling data insights and professional data visualizations.



## Ask

The director of marketing, Moreno has assigned me to answer the question: How do annual members and casual riders use Cyclistic bikes differently?

### Business task

Determine how to maximize annual memberships by converting casual riders to members. 



## Prepare

The data I will use are the historical trip data from January 2023 to December 2023 and it can be found and downloaded from [divvy_tripdata](https://divvy-tripdata.s3.amazonaws.com/index.html).

There are 12 files one for each month of 2023 and each file contains 13 columns named ride_id, rideable_type, started_at, ended_at, start_station_name, start_station_id, end_station_name, end_station_id, start_lat, start_lng, end_lat, end_lng and member_casual.

The data has been made available by Motivate International Inc. under this [licence](https://divvybikes.com/data-license-agreement). However due to data-privacy issues I can't use riders’ personally identifiable information. This means that I won’t be able to connect pass purchases to credit card numbers to determine if casual riders live in the Cyclistic service area or if they have purchased multiple single passes.



## Process

The relevant code in SQL:[SQL Data combination](https://github.com/PanagiotisChaviaropoulos/Google-data-analytics-case-study/blob/main/SQL%20Data%20combination).



In order to combine, clean and analyze the data I will use BigQuery because the size of the files is too big for spreadsheets to handle and even if it could, it would take a substantial amount of time.

### Data combination

First I created a dataset named 2023_biketrips. In that dataset I created 12 tables, one for each month, and I uploaded all the csv files. Some of those files exceeded the limited amount of 100Mb in BigQuery so I uploaded them with Google cloud. Then I combined all the tables in a unified one containing all the data from 2023. From the details in the new table we see that there are 5.719.877 rows.

### Data cleaning

The relevant code in SQL:[SQL Data cleaning](https://github.com/PanagiotisChaviaropoulos/Google-data-analytics-case-study/blob/main/SQL%20Data%20cleaning).



Now I need to clean my new table in order to be ready for my analysis. The steps I use to determine how to clean my table are the following.

1. I run a Query to determine the number of null values in each column. The following table shows the results.

   

   ![null_values](https://github.com/user-attachments/assets/180c8d26-750c-4cd9-8f19-14fd5b8f2656)

2. I want to check if there are any duplicate rows in order to remove them. After I run the appropriate Query I determine that there are not any duplicate rows.

3. I notice that all the values in the ride_id column have a length of 16 as shown in the table below. So there are no mistakes there.



   ![length_ride_id](https://github.com/user-attachments/assets/92b28384-f90b-40de-adfd-5633edb4d4c1)

4. The different types of bikes are three as shown below.



  ![types_of_bikes](https://github.com/user-attachments/assets/38e77784-7d0b-4832-8163-c8ea90759cc0)

  

5. From the columns started_at and ended_at I can create two new helpful columns ride_length and day_of_week that will help in the analysis.

6. In the new column ride_length I want to remove too short trips (under a minute) to get rid of irrelvant to the analysis trips, and negative trips (trips where the started_at column is later than the ended_at column) which is a mistake in the entry.

7. The columns start_station_id, end_station_id, start_lat, start_lng, end_lat, end_lng will not add anything to the analysis, so I need to remove them.

8. There are 875.716 missing values in the start_station_name and 929.202 missing values in the end_station_name, so the respective rows need to be removed.



## Analyze

The relevant code in SQL:[SQL analysis](https://github.com/PanagiotisChaviaropoulos/Google-data-analytics-case-study/blob/main/SQL%20analysis).



Now it is time to perform my analysis on the clean table. In order to identify trends and patterns I will query the following.

1. The minimum, maximum and average ride_length for members and casuals.

2. The number of rides by each type of bike for members and casuals.

3. The number of trips per day of week.

4. The number of trips per hour.

5. The average ride_length per day.



## Share

After I queried the abovementioned I used Tableau to visualize most of them.

First the results of the minimum, maximum and average ride_length are shown in the table below.



 ![min_max_avg](https://github.com/user-attachments/assets/fa57c5d8-6dd2-4a25-a8a4-0e89f30a7379)



We notice that members tend to use the bikes for a shorter amount of time than casuals do.



 ![Rides_for_each_bike_type](https://github.com/user-attachments/assets/bd4f203f-eac8-4609-810d-7645c5a5d40c)



It is clear that the most popular bike is the classic bike and that members tend to do more trips than casuals. Another observation is that members do not use docked bikes at all.



![number_of_trips_per_day](https://github.com/user-attachments/assets/5b31e983-b6a6-4968-96ab-e763daad42db)



We see that casuals use the bikes more during the weekends for leisure, while members tend to use the bikes more during weekdays to commute to work.



![no_of_trips_per_hour](https://github.com/user-attachments/assets/62710fe7-8a41-4365-ba84-403f734371b9)



From the graph it is clear that members during the day use the bikes most at 8 in the morning and again at 6 in the afternoon to go to work and back home. Casual riders on the other hand start using the bikes more during noon and reach the highest level during the afternoon. This implies again on touristic and leisure activities.



![avg_ride_length_per_day](https://github.com/user-attachments/assets/1a6f5a05-edfc-40ed-95ce-6eafeb6f532f)



We notice in this graph that casual riders use the bikes longer than members on average and especially during weekends. Moreover combining this graph with the previous we reach to a conclusion that casual riders ride longer than members, however members ride more frequently.



## Act

After determining the differences between casual riders and members we can develop marketing strategies to target casual riders and convert them to members.

1. As we noticed casual riders use the bikes more during the weekends, thus the company can provide weekend passes for members only.

2. Another observation is that casual riders ride significantly longer than members. Offering discounts for longer rides as a membership bonus may be a good way to convert them to members.

3. Since the afternoon is the most popular time to ride for casuals, and the classic bike is the most popular out of the three, the company can combine those two and create offers for classic bikes during the afternoon for members only.