01-27-2018 11:48 AM - last edited 10-22-2018 02:19 AM
I wanted to use Power BI's Bookmarks to show something fun and interesting. In the case of music data, it's apparent when famous musicians die--but are they famous everywhere? In this report, viewers can see how an event that might be newsworthy in a country like the United States may not even register in other countries.
Spotify publishes data for the top 200 tracks in each region with an archive going back to 1/1/2017 at https://spotifycharts.com/regional. Various sources have made the entire archived dataset available in one place. In this case, I used a dataset from Kaggle that compiles the Spotify data: https://www.kaggle.com/edumucelli/spotifys-worldwide-daily-song-ranking/data The "story" source file that I wrote is available here: https://1drv.ms/x/s!AkhaW9EUmKGTgvl6p6A47HAn1VuHgA
The largest consideration about this dataset is that it only includes Top 200 data for each region by day. The Streams measure does not truly represent the total number of streams for a track each day--only when it cracked a region's Top 200. As a result, the report does not focus on actual totals but instead tries to show a visual comparison of peaks and trends for various artists across the year. I broke some rules and did not even label axes. This is not a low-level, drill-down to detail type of report. There are lots of things that I could have done to dig deeper into this dataset, but I left it as more of a visualization than an analysis. You can expand from Date on the axis into Track though. I ultimately did not want to lose focus on the original intent--visualize the peaks of popularity surrounding a musician's death.
One decision that I made when creating this report was to use small multiples to show the relative influence of a musician in various countries. Not having each country in the same chart helps minimize the effects of population on the visual, which allows the viewer to focus on the high-level picture for each artist in each country.
I was interested in telling this story after seeing this dataset and discovering that Tom Petty had a Christmas song that was wildly popular internationally but which does not receive much (if any) airplay on the radio in the United States.