DASHBOARD WEEK! This blog is the first of the series, as I will be providing a recap on our challenge, one for every (work-)day of the week.

What is Dashboard Week? During this week, we are given a challenge in the morning (mostly). Then, working individually, we have to process the data and produce a dashboard before the end of the day.

Let’s jump straight into my recap of our first challenge!

The Challenge

We had to use the IGDB API to access the wealth of gaming-related information in their database. There is a lot of information, so we really needed to find something to focus on, analyse that, and produce a Tableau dashboard by the end of the day.

First Thoughts

When I saw the documentation for the character endpoint, one field immediately caught my attention – gender. I have always wondered about the distribution of gender for characters in games, and this was a good opportunity to explore that thought.

A quick exploration of the API (using Postman) revealed two issues: (1) genders are only coded as male, female, and unspecified; (2) only 500 out of the 14000+ characters have gender information. Despite the above, I felt like there was enough data to give an indication of the trends, so I decided to proceed with my idea.

The Data

The API is really well documented, but it is not the most straightforward to work with. Firstly, the database seems like it’s fully normalised which makes sense, but it also means that there are A LOT OF endpoints to work with, in order to fetch different pieces of information and then stich them together. To further complicate things, some pieces of information are coded as integers, and the definition of those are only available on the website, so there’s a need to work that into the data.

Unsurprisingly, considering the amount of data they had (e.g., there are 161,387 game titles), there is pagination to deal with. As such, I went about creating an iterative Alteryx macro that would work with the API:

and then used that to hit up all the endpoints that I’m interested in:

After that, it was a matter of cleaning up the data and putting them into the format that I want:

Visualisation

My design philosophy for this dashboard is, “Make It Colourful!”. Obviously not an insane number of colours, but I just wanted to have a more playful sort of design that we don’t usually get to do. It is also apt since we are working with data on games (ok, fine, I just wanted to make a cute dashboard).
(edited to add: I created the cute background in Microsoft Powerpoint! It’s really easy to do. I might write a blog about it someday.)

Considering the amount of time that we had and the amount of gender information that was available, a detailed analysis is out of the question. I decided to go with a few BANs to highlight the important numbers, a line chart that clearly shows the trend over the years, and an interactive “character database” section where users could explore the characters in detail.

Here is my final dashboard:

If you have the appetite for more, check out my Day 02 blog for Dashboard Week!

 

 

J Tay
Author: J Tay