As we neared the end of training, we geared up for the much-anticipated Dashboard Week Challenge. In this challenge, we had a day to analyse data to create insightful dashboards with compelling stories with a blog. The first challenge was web scraping, where we had to extract data of our choice and build a dashboard. The challenge seemed simple, but I completely overlooked time-boxing.

I spent most of the day trying to look for data that was easily accessible but also needed to tell a compelling story. Finally, I chose data from Wikipedia that had a list of Game of Thrones episodes. The next step was to download this data using Alteryx. So, I created a workflow that downloaded the data from the URL and parsed all the information. One of the challenges which I faced was not being able to modify some incorrect data (such as the spelling of the director) that I put into separate containers so that I could return to it later.

Part 1 of workflow

Part 2 of workflow

Now that I had extracted the data from the website, I needed additional data to supplement my dashboard. So, I created a separate workflow to get all the episode ratings.

The final step was to build the dashboard. Since I had only a limited number of columns, it was challenging to find a good story. Nevertheless, I created the below dashboard that looks at the total viewership, the number of episodes directed by the directors and the impact they had on the ratings.

A few insights that I got from the analysis were:

  • The first seven seasons of the show were rated quite well. Episodes 9 and 10 had the highest ratings. The last season was the worst, but it had the highest number of viewers.
  • David Nutter directed the highest number of episodes but had an average rating of 8.5.
  • Neil Marshall directed only two episodes but had the highest average rating of 9.7.
  • Both D. B. Weiss and David Benioff previously directed episodes with high ratings, yet the last episode they directed together had the lowest review of all time.

In conclusion, my learning from day one was to manage time more effectively next time so that I could focus more on building a better story with good insights.


Icon Attribute: <a href="">Business illustrations by Storyset</a>
The Data School
Author: The Data School