COVID-19’s Effect on the Global Music Business, Part 1: Genre

The partner of DNBB Group: Chartmetric's data-driven analysis of COVID-19’s effects on music-related consumption helps artists, songwriters, labels, agencies, distributors, and other entertainment-related entities sustain and improve their well-being during these unprecedented times.

To help address the global music industry’s concerns about the COVID-19 pandemic, US-based music analytics company Chartmetric is continuously monitoring 2M+ artists across 20+ streaming and social media data sources.

We hope our data-driven analysis of COVID-19’s effects on music-related consumption helps artists, songwriters, labels, agencies, distributors, and other entertainment-related entities sustain and improve their well-being during these unprecedented times.

All data is collected, organized, and analyzed by Chartmetric. Data science by Nuttiiya Seekhao. Analysis by Jason Joven and Rutger Ansley Rosenborg.

Key Takeaways:

  • Spotify listenership appears to be widening for Classical, Ambient, and Children’s due to COVID-19.

  • Spotify listenership is relatively unaffected for Pop, Country and Dance during COVID-19, but Country seems to be demonstrating the greatest resiliency.

  • Spotify listenership appears to be narrowing for Latin, Rap, and Rock during COVID-19, but potentially due to other factors and not necessarily a result of the global pandemic.

The COVID-19 Context:

If you’ve been a part of the global music industry, you have been well aware of the damage COVID-19 has done to music from multiple industry angles. While streaming has been relatively untouched from a revenue standpoint, its most popular artists’ music has tended to receive less play. On the other side of the spectrum, the live music sector has taken the brunt of the industry trauma, with public companies like Live Nation taking huge stock price hits, and other larger-than-life firms like talent agency Paradigm suffering layoffs. New shows and movies can’t shoot and public venues playing music can’t open, so licensing has taken a hit as well, and because music studios don’t likely qualify as essential businesses, their futures are uncertain.

The worst hit are our everyday artists, but workers who make the live scene possible fare no better. Music, in its highest form, is about community, as many of COVID-19’s unexpected moments have shown us: a Spanish fitness instructor leading a class from his rooftop, apartment building sing-alongs, and our beloved medical workers taking a well-earned moment of levity.

***iMarkkeyz' "Cardi B Coronavirus Remix Video" (2.3M YouTube views) first blew up on the DJ'sInstagram(March 13) and then made it to Cardi B's ownprofile(March 16)

Previous industry news has already shown us the broader, market-level strokes: Video-streaming platforms like Netflix and YouTube are getting more play while most of the world shelters in place. The assumption is that streaming music nowadays tends to be an activity that co-exists with others: commuting on the train, driving in the car, working out, socializing with others. With our public transportation abandoned, reasons for going out limited, gyms closed, and citizens in quarantine, it’s no wonder people’s attention gets shifted toward mediums that tend to require multisensory engagement.

COVID-19 and Genre: Considerations:

If we want to develop a nuanced understanding of music consumption in the COVID-19 era, we have to lead with a manageable perspective. Here, we start with streaming insights at the genre-level.

Baked into music genres is an assumption of listener populations that share some sort of geography, culture, and entertainment consumption habits. By examining Spotify Monthly Listeners as measurements for these factors during the pandemic’s spread, we can attempt to put metrics to these dimensions.

First, a timeline: We determined March 3, 2020, to April 9, 2020, as our window for this genre-level look. To extend back to Jan. 1 or earlier would be valid, but we would also risk losing some of the trends that become more visible during a shorter period of time. Too late, and we may cut short trends already in progress.

***A worldwide Google Trends “Coronavirus”searchpeaked on March 15, 2020, suggesting an apex of global curiosity and effects of government-enforced (or fear-induced) lockdown.

Though internet searching is one of many ways to measure public sentiment, a Google Trends search on “Coronavirus” is a simple (if imperfect) way to summarize how much news and Internet-driven concern there is about a topic. You can see that at the global level over the past three months, initial curiosity peaked on March 15 (the graph does not have a “unit,” as it is normalized from 0 to 100). While there were small waves of interest in mid-January and mid-February, the real surge occurred in early to mid-March when governments at international, national, and local levels began enforcing shelter-in-place or self-quarantine measures to help “flatten the curve” of rising virus infections among the public.

Given the fact that we are searching for the effect on genres from a streaming perspective, it makes sense for us to observe any possible activity changes focused on this time period in particular. When lifestyle changes, it’s reasonable to expect an effect in many parts of life.

A New York Timesgraphshowing the different levels of coronavirus deaths at the country-level.

Geographically speaking, with respect to COVID-19’s effect on society, it’s important to note that different regions of the world have had vastly different responses and histories to epidemics threatening their citizens. At the time of writing, South Korea and Taiwan are a couple of the nations who have been lauded for their expedient and successful handling of the crisis, while other countries like Brazil have only now begun to implement public restrictions.

With respect to Spotify’s user base, consider whose app usage we are measuring. Note below Spotify’s Q4 2019 Monthly Active User (MAU) base, or the number of users who have used Spotify in the past month, broken down by international region.

***Spotify’s Q4 2019 Monthly Active User (MAU) base favoring Europe and North America at a 62 percent combined total.

So, the varying behavior measured in this time period will contain more than 60 percent of a North American/European bias. Latin America is at 22 percent (though it’s unclear if Mexico, containing one of Spotify’s biggest markets, is included in North America or here), while huge swaths of populations in Africa, Asia, the Middle East, and Oceania all get binned into the 16 percent “Rest of World” category. We should note that any Spotify analysis is going to strongly minimize most of these markets’ tastes and preferences.

While Spotify is global, it’s worth mentioning that the platform is still limited to its market reach. Knowing these limitations, we can resume our interpretation of the genre-based insights we uncovered.

Our ‘Trendy’ Approach:

  • Spotify Monthly Listeners (MLs), as used here, is a Spotify-generated statistic showing the cumulative number of unique listeners for an artist in the past 28 days. It is about how “wide” a listenership is (while stream count can be considered how “deep” an audience listens). Its “reach” nature is largely fueled by a combination of new releases and editorial playlisting. For this analysis, we collected MLs data for the Top 100 artists of each genre, ranked by Spotify Follower count.

  • Genres are social constructs and inherently subjective, so we’ve attempted to optimize our categorizations to traditional industry expectations. Here are some sample sub-genres from each of the nine genres we looked at:

1) Classical, Ensemble & Opera includes romantic era, early modern classical, a cappella

2) Ambient, Relaxation & Experimental includes compositional ambient, lo-fi beats, new age

3) Hip-Hop & Rap includes trap music, gangster rap, underground rap

4) Pop includes dance pop, post-teen pop, pop rap

5) Country includes contemporary country, country road, nashville sound

6) Dance & Electronic includes edm, tropical house, funk carioca, drum and bass

7) Latin & Caribbean includes reggaeton, latin trap, latin pop

8) Rock, Punk & Metal includes modern rock, protopunk, alternative metal

9) Children’s includes children’s music, musica para niños, kindermusik

  • Normalized” measurements mean that we mathematically translated those absolute numbers to a uniform scale in order to zoom in on subtle dynamics and trends. For example, the chart on the bottom left is absolute, and the chart on the bottom right is normalized, but it’s the same data!

***Spotify Monthly Listeners for Classical music (March 3-April 9, 2020). The left graph is raw data, and the right graph is normalized to show the subtle trends. Don't get intimidated: it's the same data!

The Top 100 Classical artists according to Spotify Followers that make up the above two graphs. This is how we'll examine each of the nine genres below.

Here is a link to all the charts with the Top 100 artists in each genre. We will truncate the artist name charts below to focus on the genre trends.

Classical and Ambient: The Big Quarantine Winners

According to a number of different sources, Classical, Ensemble & Opera ( “Classical”) and Ambient, Relaxation & Experimental (“Ambient”) music seems to have really won out in these dark times. Working from home has become the new norm for many, and apparently, lyrics just get in the way — either that, or listeners are seeking a sense of calm and order when work, school, social, and family routines have been turned upside down.

Both genres have helped determine an important marker for when Spotify users started to exhibit unique behavior, relative to other genres. This inflection point occurs roughly around March 18th, and near where the “coronavirus” Google Trend peaks, as noted above.

***A look at the Spotify Monthly Listeners trends for the Top 100 Ambient artists suggests a genre-wide lift during COVID-19 lockdown for many European and North American countries.‌‌

In Ambient’s Top 100 artists, you can see a grouping of high-flying artists around the 0.9-1.0 range, and then a band of artists between the 0.6-0.8 marks. There is even a loose category of artists who trend down strongly from the beginning of March, losing MLs quickly.

But the interesting inflection point is between March 16 and March 23, where all three groupings trend high and to the right, peaking at the April 9 end of the timeline. Approximately 80 percent of the artists here do so. So, no matter what their MLs momentum was, they seemed to march to the beat of the same drum at the same time.

***A look at the Spotify Monthly Listeners for the Top 100 Classical artists suggests a genre-wide lift during COVID-19 lockdown for many European and North American countries.

With Classical, an even stronger lift trend reveals itself around March 18. Here, one grouping of artists have decent MLs, while a heavy clustering of artists have minimal MLs starting from the beginning of March. Yet again, we see another coinciding trend up in mid-March in both bands, also terminating on April 9 at their highest level within the time period. Even some of the “noise” that does not fall into either of the groupings trends up slightly in mid-March, with only 10-15 percent of the artists showing no coinciding behavior.

We’ll say it a million times: correlation is not causation. It’s a basic tenet in statistics and data science. But it’s certainly a notable coincidence. And in case you’re wondering, “Well, maybe this is what always happens this time of year”.... We wondered the same.