Video Game Dashboard

In this project, I analyzed video game sales data from Kaggle using JupyterLab again. Only this time I showed the SQL equivalent to my python code to test my knowledge with both languages. The end result of this analysis is presented in a Tableau dashboard.

Step 1: Importing

Using pandas to bring the dataset into JupyterLabs and prepare it for analysis.

Screen Shot 2022-07-19 at 9.32.30 AM.png

Step 2: SQL vs. Python

As you will see below, the green code followed by a "# " is SQL. The code below that is its Python equivalent and then the output.

I started by rounding and selecting all of the sales columns, which are in the millions.

Screen Shot 2022-07-19 at 9.39.57 AM.png

Then I multiplied them all by 1,000,000 to get the full numbers. Please note that these are unit sales, not currency.

Screen Shot 2022-07-19 at 9.42.28 AM.png

I then selected Wii games where EU sales were greater than NA sales.

Screen Shot 2022-07-19 at 9.48.14 AM.png

From there, selecting 2010 games ordered by global sales in descending order.

Screen Shot 2022-07-19 at 9.51.14 AM.png

Calculating monthly sales in a new column

Screen Shot 2022-07-19 at 9.53.51 AM.png

Selecting all columns, grouping by platform, having monthly sales greater than 1.

Screen Shot 2022-07-19 at 9.57.22 AM.png

Using a window function, creating a running total.

Screen Shot 2022-07-19 at 9.58.49 AM.png

Step 3: Visualization

After exporting my dataset to Tableau, I created a dashboard that will aid a business in drilling down deeper into the sales insights.

The dashboard starts by outlining the difference in sales amongst title genre. It also shows the top 5 games sold, the highest selling publishers and unit sales comparison between North America, the EU and Japan.

Screen Shot 2022-07-20 at 10.44.13 AM.png

I also included a tooltip that pops up when you hover over each genre. This shows the total global sales, top 3 games and global sales over time for each highlighted genre.

Screen Shot 2022-07-20 at 10.44.36 AM.png

As you can see, you can click each genre and the top publishers, as well as the line graphs at the bottom will change to reflect the category that you click.

Screen Shot 2022-07-20 at 10.45.26 AM.png

Finally, you'll notice that you can use the slider on the top right of the dashboard to have the data reflect any range of years that you choose.

Screen Shot 2022-07-20 at 10.45.57 AM.png

There were 3 goals that I had for this project, that I believe were executed well:

1. To showcase my knowledge of SQL and use my obstacle of not having an SQL test environment, as an opportunity to demonstrate my understanding of of Python equivalents to the SQL code.

2. To make it easy for a business to gain insights from this data, while making sure the design stayed fun, with a classic video game feel to it.

Hopefully you can also see the technical usefulness and feel the nostalgia that I did while doing this project.