It’s day 4 of dashboard week already and this time, the challenge is to:
- Use API to extract data from IMBD
- Can find additional data to supplement our extracted data (optional)
- Build a visualisation on Tableau
One thing to note when using API is often you have a limit on the data you can download per day. So be careful and don’t forget to right-click on your download tool and click ‘Cache and run workflow’.
Figure 1 – Cache and Run Workflow in the menu box after right-clicking the download tool
I’m not really a movie buff and there are heaps of angles you can take to approach the dataset. At first, I wanted to analyse the top 250 movies’ music scores but in the end, I pivoted to a completely different topic – Disney movies.
I was mainly interested in the trend of how the movies performed and the reasons behind the flop movies (Disney also lose its magical touch at times). And from there, there are some interesting insights:
1) The trend of movies released and their rating:
Disney seems to ramp up the number of movies it released over time although there was a slump in the 1980s.
Although there are more movies, until the 1980s the IMBD rating trend was performing worse. Probably explained why Disney didn’t release as many movies suddenly – some introspection going on I suppose.
And skyrocketed back to the good Disney movies we love in the 1990s, started to partner with Pixar and created a whole bunch of great movies.
2) Flop movies:
Bigger budget = Bigger box office. No brainer. But how about movies that have a high budget but a low box office? Most people will usually think they are bad movies. But apparently, there are 4 movies whose IMBD ratings are above the average for all Disney movies and still ended up as flops.
They have their own reasons and stories why they’re performing poorly – will add the text explanations when I have the time to update the visualisation
Here’s the screenshot of the visualisation I created:
Figure 2 – Dashboard on ‘Exploring Disney Movies’
My approach this time is to focus on the insights and story. There are some updates that I plan to do when I have the time:
- Add text descriptions to explain the insights with some user interactions and
- Design it with a dark background and a more Disney nuanced theme (a full-blown design consumes lots of time)
There’s a trade off depending on the approach you took when you only have a limited amount of time to complete a dashboard. On day 2, I went for a more stylistic approach and went for one screen desktop layout so just focusing on one insight (you can refer to my previous blog here). But this time, I sacrificed on the stylistic design for more insights and story.
1) Going with dark background usually takes a longer time as you need to format all your text, axes and labels to a lighter colour
2) There are some design styles I usually lean to if I don’t have the time to create a full-blown design using other softwares (may it be Figma, Canva or even Powerpoint). I will probably cover this in more detail in my future blog post – some small easy things which can make your design still look pretty good and clean