DashboardWeek Day 3. I arrived in the office and my laptop got taken away. But no worries! Our coach had an idea. After 4 months of training and every day looking into a computer screen for several hours, our first hour today was offline! We worked old-school with pen and paper, which was a really refreshing task that even reminded me that it can be really creative and sometimes is even a quicker way to sketch something or get ideas together without a screen.
And what was our task today – a Spotify analysis. So I started with up to 20 ideas of what I would like to analyse (we didn't see the data yet) and sketched out different visualisations, from bar charts to advanced types like a waffle or Sankey chart – everything was included. I guess I overestimated what was possible. What is likely for me: there are so many options, and I love implementing a lot of features and interesting visuals – that is probably the reason why I like working with data, the possibilities are endless. But it always depends on the data available. And most of the time, not everything I would like fits in the time available – probably one of the top things I have to get happy with. And as well the reason why we are expected to deliver MVPs. It is better to have a fully working MVP, you can use and see the further steps or ideas popping up in your head for improvement , than to have only 25% of your dashboard looking and working fine.
But before we start, we need data! We could use our own Spotify data or take the data from our coach. I took the one from our coach and connected the data with Tableau. First, I checked some insights to see what would be possible to visualise of all my earlier ideas. Obviously not that much 😄. Unfortunately, the API we thought of using to get more data about one individual track didn't work anymore, so some of my ideas were not possible anymore. After discovering the data, I went online to get some inspiration from Tableau Public. Some dashboards looked really nice and had some good visualisations. I realised that not all my plans were going to work due to data and time available.
I started with some quick visualisations of the Top 10 Artists and which platform was used to stream. Then I made a heatmap to show the time listened over the year and at what time of the day. I found online a little implementation of an embedded Spotify player that I included as well. After the presentation, I got the feedback to connect the Top 10 Songs to the embedded Spotify player so that it changes the songs in the embedded Spotify player – which was a great idea and made the dashboard even more interactive. If you would like to see how to implement it, check my profil for the blogpost to this topic.
If you wanna see the dashboard or get some nice music to listen to:
https://public.tableau.com/app/profile/stephan.christner./viz/PetersSpotifyInsights/PetersSpotify
Have fun! 🎶