It’s been nearly three weeks since I’ve joined OmniSci’s Community team as a Developer Advocate Intern and I’ve had a delightful time using OmniSci Immerse- a web-based data visualization interface- to analyze Spotify’s global music trends and explore how artist/song popularity varies over time. In this post, I’ll demonstrate how I was able to import and visualize my data on Immerse and how you can too!
Importing Data into OmniSci
The goal for this project is to visualize Spotify’s data with OmniSci Immerse to analyze the trend of artist/track popularity from 2017–2018. I downloaded this Spotify dataset as a CSV from Kaggle, which contains the daily ranking of the 200 most streamed songs in 53 countries from 2017 and 2018 by Spotify users. It contains over 3 million rows, comprising 6629 artists, and 18598 songs for a total count of one hundred five billion streams.
Now that the data is on my local machine, it’s ready to be imported into OmniSci Immerse. Inside the OmniSci Cloud interface, select the ‘Data Manager’ tab from the navigation bar, then click on the blue ‘Import Data’ button on the upper-right corner:
We’re given two options for importing data: from a local file on your machine or from Amazon S3. Since my data is on my local machine, I’ve selected the ‘Import data from a local file’ option, and then selected the Spotify CSV file I downloaded from Kaggle.
After OmniSci has finished uploading the file, you’ll see a preview of the data, have the option to rename the table, and finally we can select the ‘Import Data’ button.
After the data has successfully imported, you can return to the Data Manager page and see that your new table is included. Now you’re ready to create your very first dashboard.
Creating a Dashboard
Dashboards are used to create and organize the charts that represent your data. The coolest thing about Immerse is the wide range of chart types you can choose from to visualize your data. Immerse provides standard visualizations, such as line, bar, and pie charts, but also complex data visualizations, such as geo point maps, geo heat maps, choropleths, and scatter plots.
To create a dashboard, you’ll need to click the blue button that reads ‘New Dashboard’ in the upper right of the Dashboards page, give your new dashboard a name, then click the Save button.
To begin analyzing the Spotify dataset, we’ll need to add a chart by clicking the button that says ‘Add Chart’ in our dashboard, selecting a chart type (e.g. bar), and specifying that we’d like our newly imported Spotify table as our data source.
After we’ve specified our chart and data sources, we can set the dimensions and measures. The goal for this chart is to display the highest streaming artists by setting our measure to the sum of all streams for each artist.
Now that the dimensions and measures have been specified, we can rename our chart, click the Apply button, and save all changes to our dashboard.
This single chart has only scratched the surface of what’s possible with Immerse. Within minutes, I added four additional charts to create the dashboard below:
Dashboards aren’t automatically saved in Immerse, so if you leave your dashboard page before you save, all changes will be lost. If you make any changes to your dashboard, be sure to Save before you leave.
Ed Sheeran on the A Team
With the visual representation of our data, we can determine which artists are the most streamed around the globe.
The Chainsmokers, Drake, and Ed Sheeran were the most streamed artists of 2017–2018, but the clear frontrunner is Ed Sheeran who was streamed nearly 9 billion times (more streams than Drake and The Chainsmokers combined!).
The primary factor of Ed Sheeran’s success was due to the release of his third album Divide in March, which won four awards including the Grammy Award for Best Pop Vocal Album at the 60th Annual Grammy Awards. We can confirm this by looking at the dates lining up to the release of his album with a Line chart in Immerse:
As we can see in the line chart above, Ed Sheeran previously garnered 25+ million streams daily until his album release date, where he gained over 91 million streams in a single day.
Filtering Our Data
With the visual representation of our data, we’ve been able to determine that the three most streamed global artists are Ed Sheeran, The Chainsmokers, and Drake, but what if I only wanted to see the three most streamed artists in Mexico or Hong Kong?
Immerse lets users add filters to their dashboard in a few easy steps, so let’s filter our data to see what tracks and artists are most streamed in Mexico. We’ll first need to click on the ‘Filters’ icon, then click on the ‘Add Filter’ button. From here, we’ll select our data source (e.g. spotify_data), select what column we’d like to filter, and finally what value we’d like to filter. Since we only want to view the trends in Mexico, we’ll type “mx” as the value.
Now that we’ve filtered our data, we can see that the three most streamed artists in Mexico are J Balvin, Maluma, and Ozuna- a complete contrast from our global charts.
Another great thing about Immerse is cross-filtering. Dashboards automatically cross-filter when interacting with data, so when you add more than one chart to a dashboard, filters on one chart affect all other charts on the dashboard.
As I select different tracks on the “Highest Streaming Tracks by Region” chart, the other charts are filtered, as well.
This introductory analysis of the data is only a fraction of what’s possible with OmniSci Immerse. If you have a dataset you’re interested in, I encourage you to take advantage of OmniSci Cloud’s free 14-day trial to find quick insights in a matter of a few steps. If you create a great dashboard that finds interesting insights, feel free to stop by our Community Forum and share your results or feel free to send us an email directly for any questions you have about the OmniSci platform.