Overview

This project was done for the DataDNA - Dataset Challenge, conducted by Onyx Data for the month of January 2024 to showcase my data analytical and visualisation skills.

The dataset was Udemy Courses from 2011 to 2017. The data consists of the following:

The dataset was loaded and transformed in PowerBI. The data was was fairy clean, only data type correction was needed and the duplicates were removed.

Measures were created to calculate the total courses, revenue and subscribers

Objective

To create a report:

• to show an overall trend in Udemy Courses subscribed by users.
• to show the Top Reviewed Course.
• to display individual course details.

Findings

• Total Courses is 3672 and 310 courses were free courses during this period.
• Wed Development and Business Finance accounts for almost 65% of the published subjects over the years.
• Web Development was the most subscribed subject at 68%.
• Courses in the price range of $1 to $50 was the most published, with 2029 courses.
• 310 courses were free courses during this period.
• Web Development generated the most revenue($627.6M), with 'The Web Developer Bootcamp' course generating the most.
• There was a small negative corelation between subscribers and price. As price of the course increased the subscribers dropped.

DAX

KPIs:

Total Courses = 
DISTINCTCOUNT(
    'Onyx Data -DataDNA Dataset Chal'[course_id]
)

Total Subscribers = 
CALCULATE(
   SUM(
    'Onyx Data -DataDNA Dataset Chal'[num_subscribers]
   )
)

Total Price = 
SUMX(
    'Onyx Data -DataDNA Dataset Chal',
    'Onyx Data -DataDNA Dataset Chal'[price] * 'Onyx Data -DataDNA Dataset Chal'[num_subscribers]
)

For the course details section:

Subject = 
IF(
    HASONEVALUE(
        'Onyx Data -DataDNA Dataset Chal'[course_title]
    ),
    VALUES('Onyx Data -DataDNA Dataset Chal'[subject]),
    "-"
)

I used similar code to display :

• Level
• Duration(Hours)
• Paid/Free
• URL
• Number of Lectures
• Price per course
• Subscribers per course
• Reviews per course

Visuals

Visuals were made using ZoomCharts, and I used three of them

• Column Chart:
  • Drill Down Combo Pro (Filter) - To drill down the number of courses published by year & month.
• Pie Chart:
  • Drill Down Donut Pro (Pin) - To drill down the number of course, subscribers and revenue generated by subject , level and if it is paid / free course.
• Bar Chart
  • Drill Down Combo Bar Pro (Filter) - This was for the price distribution of the course. Thought drill down option is available for this, I did not provide the second parameter for the it as I just wanted to show the distribution of price.

The background for the report was created in power point and saved as an svg file (Scalable Vector Graphics).