Day 4 of dashboard week and Andy surprised us with a project which has been covered already by two previous cohorts: DS11 and DS22. The challenge was to download some data about the top boybands from 1980 onwards, according tot the Billboard Hot 100 list!
After downloading the data I loaded it up into Alteryx to do some quick prep, before spending most of my time thinking up loads of different dashboard ideas before actually settling on one! The following documents this process throughout the day.
Step 1 - Data prep
The first step was to clean the data in Alteryx: first joining the boy band members with the data on the bands and their top hits, before doing some basic preparation, such as converting a field into a proper date field, and converting the band member heights into centimetres. My completed workflow, which I outputted as a .hyper file is below.
Step 2 - my thought process
There were many iterations of dashboard design before I sort of stumbled on to my final dashboard. Unfortunately I didn't save any of the previous versions of any plans, but some of the ideas I had were:
● Creating a dashboard where you could choose different physical metrics (such as hair type, hair colour, eye colour etc) to filter a dashboard to show how those characteristics would fair in a boy band.
● Showing the hairiest boy band members since 1980
● Creating a dashboard which just showed data for the 1990s as this was the most popular decade for boy bands
● My final idea which I finally settled on: Highlighting the year 2000 because it was the most successful year for boy bands.
Step 3 - Dashboard creation
After going through many sheets on Tableau, and thinking up different questions I could answer with the data, I decided to focus on the year 2000 and create a dashboard which offered specific information on the year 2000, before investigating a few traits about the different boy bands over the years such as hair types and successful clothing combinations.
At the top of my dashboard, there are a few BANs which highlight the year of 2000 and some key statistics for it. Below that I created a marginal heat map that shows the popularity of different hair types, as selected by the user.
Finally, I wanted to look at which clothing combinations (top and bottom styles) which were seen the most in Hot 100 hits.
The main challenge I came across today was just coming up with an idea which wasn't awful.