Indie Artists, Labels thrive by focusing on off-center niches, new genres, and local music [Chartmetric]

The team at Chartmetirc analyzed 6,000 playlists to understand the relative power of major vs. indie catalogs in Spotify Recommendations across music genres.

by Dmitry Pastukhov from the Chartmetric Blog

In the post-streaming era, streaming recommendation algorithms have become key mediators in music consumption and discovery. According to Spotify’s “Made to be Found” report released in January 2022, more than a third of all new artist discoveries on the platform happen through personalized features like “Made for You” playlists. Today, recommenders are more than just tools, enabling users to navigate and filter the ever-growing body of music. Listeners built trust in these systems: a song that made it to the top of your Discover Weekly can be akin to a recommendation from a trusted friend—the one who already introduced you to some of your favorite artists.

Yet, while the algorithmic landscape might seem infinite compared to the crowded editorial spaces, at the end of the day, it really isn’t. One song getting recommended always means that another won’t make the cut–and the favor of the algorithm can have an immense impact on the release trajectory. As a result, DSPs face growing pressure to ensure these systems are fair—but given the lack of transparency around streaming recommenders, the industry is left with little insight into these systems.

In the few years Music Tomorrow spent deciphering music recommenders, this is one of the most frequent questions we get: how fair these systems really are? Do recommenders level the playing field, carving out a space where independent and unsigned artists can compete on par with the major-backed projects? Or, is it the same “the rich get richer” narrative, with algorithms driving listeners to the most popular tracks, making them even more popular? 

On the surface, it seems like there’s no way to answer such questions. Recommenders are often perceived like black boxes: how can you assess the landscape of algorithmic recommendations when it’s split across millions of personalized playlists, instant radio queues, and autoplay sessions? Well, the personalized recommendations are, indeed, beyond our reach. However, the recommenders generate millions of data points, and if you know where to look, you can assess recommendations as a whole.

To address that fairness question and assess the relative power of independent catalogs in Spotify recommendations, Music Tomorrow partnered with Chartmetric and retrieved a dataset of over 6K Every Noise at Once playlists. Each of these playlists is an algorithmic representation of a particular Spotify genre showcasing the most listened tracks in that niche: from the broadest tags of popand rap down to the nitty gritty of neue neue deutsche welle and egg punk. Here’s what we’ve learned about the relative power of major and indie catalogs in Spotify recommendations across the most prominent music niches and communities.

About the Dataset

Every Noise at Once is an independent project developed by Glenn McDonald, a veteran of the Spotify personalization team. Initially conceived to represent the genre convention developed by the Echo Nest, Every Noise at Once remained active as Spotify acquired the platform back in 2014, to lay the foundation of its nascent recommender system. Today, it exposes a regularly updated ecosystem of more than 6K unique genres, used by Spotify to identify distinct musical niches and enrich asset and user data powering the recommender. For every genre tag on Spotify, Every Noise at Once generates a set of at least four playlists:

• The Sound of XXX, showcasing the most popular tracks that define the sound of that genre
• The Pulse of XXX, showcasing the tracks currently listened to by the fans of that genre
• The Edge of XXX, showcasing some of the less-known tracks currently listened to by the fans of that genre
• The Intro to XXX, showcasing tracks that are considered the best introduction to a genre (e.g. those likely to be recommended to listeners engaging with the niche for the first time)

To limit the scope of the analysis, we’ve decided to focus on the “Sound of” playlists, that showcased the key tracks, most popular with the fans of a particular genre, regardless of their release date (while “Pulse of” and “Edge of” playlists focused on recently released tracks, currently getting traction with the fans of the genre). For each of the 6K+ “Sound of” playlists, we’ve extracted up to 24 monthly snapshots, stretching back two years, from January 2020 to December 2022. Each snapshot contained all tracks featured on the playlist, alongside their position and label affiliation—allowing us to differentiate between major and independent catalogs. 

In the modern music industry, the lines between major and indie catalog can get extremely blurry. For the purpose of this project, we’ve defined the major catalog as tracks released by major labels & major-affiliated imprints, or tracks partially owned by major players. For instance, Joji’s BALLADS 1 has been originally released through 88rising & 12Tone Music (which has been acquired by Warnerback in 2021). In such a case, the label affiliation of “88rising Music/Warner Records” would be mapped to Warner Music Group. 

With the total dataset adding up to over 60 million lines, we further enriched the data by tapping into Spotify API to get each track’s popularity index, measuring the song’s current popularity on a scale from 1 to 100. Finally, we ranked all the genres according to the Every Noise at Once internal popularity rank.

The relative power of major and independent catalogs across Spotify genres

First things first, we’ve studied the connection between Every Noise at Once internal popularity rank and the average popularity index for the tracks on the “Sound of” playlist for that genre, finding the two closely correlated:

With the Every Noise popularity rank established as a good proxy for the genre listenership, we’ve looked into the top 500 most popular genres and calculated the major label share, expressed as a percentage of tracks on a “Sound of” playlist owned by either Sony, Warner, or Universal. Then, we plotted the major label share as a function of the genre popularity.

Perhaps unsurprisingly, the share of the major catalog (abbreviated as MS along this article) declined as we moved from broad tags of pop and hip hop into the niche spaces. For the top 10 most popular genres on Spotify, featuring pop (64% MS), dance pop (68% MS), hip hop (43% MS), rock (70% MS), and EDM (27,5% MS), the average major label share added up to about 48%. Then, as we moved further down the popularity rankings, the average major share gradually decreased, reaching an average of about 25% for genres close to the 100th popularity rank (such as europop, indie rock, and neo soul).

Even with those preliminary insights, we could tell that the catalog distribution across algorithmic niches tends to skew indie. For example, if we look at the recording market as a whole, most estimates put the major market share at around 70–75% for 2021—and Spotify recently reported similar figures as a share of revenue distributed to majors or Merlin affiliates. Yet, across our dataset, only the most major-heavy genres—such as classic rock and mellow gold—have crossed the 70% major share threshold. 

Sure, the revenue estimates quoted above include formats beyond streaming, and “catalog distributed by majors” is not necessarily the same as “major catalog.” Many independent labels distribute their catalogs through major-affiliated partners—meaning that majors will often include the market share of the independent-owned content they distribute in their own investor reports. However, if we aggregate the major share across the top 100 Spotify genres, only 43% of the tracks could be attributed to the big three, with the major share further dropping to 27% if we expand the sample to the top 500 genres. Such discrepancy is hard to write off on external factors—leading us to believe that, on average, the genre niches tend to be driven by independent catalog. 

To gain a more detailed view of the power dynamics and major share distribution across the top 100 genres on Spotify, we also examined the outliers in terms of the major share. More specifically, we asked ourselves: which genres deviated the most from the observed linear trend?

The most (and least) major-heavy Spotify genres

Our analysis of the outliers revealed that while majors thrived in UK and US, catalog-heavy spaces, the indies took over when it came to local musical contexts, off-center niches, and emerging genres. Looking into the high outliers (highlighted in green on the graph below), we could immediately detect a distinct group of genres. Classic rock (77% MS), mellow gold (79% MS), permanent wave (73% MS), new wave (60% MS), hip pop (68% MS)—all of these sonic spaces were rooted in a pre-streaming era. The “sounds” of these genres have solidified in the days of the “old industry”: back in the 60s through to the early 2000s. These were the days when the evergreen catalogs, often quoted as one of the key advantages of the major labels, were forged—and the era when the music business looked much different. 

On the other end of the spectrum, the first group we found related to non-US/UK music contexts: filmi, desi pop, and modern Bollywood (all connected to the Indian music market); narrative driven Mexican folk pop of corrido, and the global phenomenon of K-pop. These niches were rooted in local music contexts and stemmed from the markets with mature local recording chains.  

As for international genres, if the higher end of major share distribution revolved around rock-adjacent music niches, the lower end featured the styles connected to all things electronic. EDM, for instance, was by far the most independent out of the top 10, with only a 27,5% major share. To explain such an extreme outlier, we’ve dug deeper into the label affiliation data for EDM. What we found was a picture of a “centralized independent market”: a significant portion of non-major tracks on the Sound of EDM gravitated towards a few key independents in the niche, such as Monstercat, Armada Music, and Revealed Recordings.

A similar structure was true for most electronic genres in the top 100 (e.g., house, tropical house & electro house) as well as art pop, with a wide range of powerful independent players taking over the niche:

However, if we were to name a single most independent popular genre in our dataset, that would have to be underground hip hop (and closely adjacent tag of viral rap). These genres feature only 10% and 19% of major-affiliated tracks, respectively and the independent share is also evenly distributed, with no particular players centralizing the power. 

In the case of underground hip hop, for instance, the latest snapshot featured 713 independent tracks split between 440 different imprints—with a healthy portion representing self-releasing artists. The viral nature of these genres could perhaps explain such catalog structure: these genre spaces were not only defined by their sound but also by the approach to marketing and promotion favored by these artists. Relying on digital word-of-mouth and emerging digital platforms— remember Soundcloud rap?—the artists in that niche were able to carve out a uniquely independent, decentralized space for themselves.

How did the relative power of major catalog shift in the past two years?

Lastly, we wondered if there was a clear trend we could detect regarding the major label share across the Spotify genre niches. With our dataset containing 24 monthly snapshots for each Sound of playlist, stretching from January 2021 to December 2022, we were able to calculate the monthly average major label share and track its development over time.

Plotting the trendline for the top 100 Spotify genres, we could see a clear pattern emerge. Across the two years, the major label share steadily decreased, dropping 3,2 points from 46% to 42,8%. This trend was further confirmed as we expanded the analysis to the top 500, with the average major share dropping from 30,6% to 27,2%. Moreover, the observed trend has aligned with some industry figures reported earlier this year: according to Spotify’s annual fiscal report, for instance, the revenue share of the Major/Merlin-backed catalog on the platform has dropped 3% from 2020 to 2022. A similar pattern could be found if we focus solely on the Sound of Pop.

The trend we found closely mirrored the general trend observed across the top 100/500 Spotify genres: over the past two years, the major label share for the Sound of Pop gradually decreased, falling a total of 3,8%. Looking further into the data, we found that most prominent global genres—such as dance pop, hip hop, rap, or r&b—also fit this pattern.

To sum up, our analysis of the algorithmic representation of genre niches on Spotify has revealed that, on average, the Spotify genre niches tend to be driven by the independent catalog. The share of major-affiliated tracks has sharply decreased as we moved away from the most popular genres, such as pop, and into more granular niches. Further, we found that most major-heavy niches on Spotify were connected to the “old” music industry: genres like mellow gold and classic rock, that cemented their sound back in the 20th century when the recording market was much more centralized. However, the days when radio and TV ruled the game are largely behind us—the modern music market is orchestrated by digital platforms, hosting colossal volumes of content, and their personalization systems, driving consumers into the niches rather than towards the top-40 hits. 

The reason majors still hold such a vast share of the music market is not only their marketing budgets and leverage. As our analysis showcased, big independents can compete on par—or even outweigh—the majors across many genre niches. A critical component of the major’s competitive advantage is the evergreen catalogs held by the big three—and the “passive” market share that comes with it. However, if we move away from these “golden” genres, independent players will often carry more power than the majors. And when it comes to emerging genre spaces, such as underground hip hop and viral rap, things can get uniquely independent—one might even say self-released.

What is a niche today, can quickly turn into a global, mainstream phenomenon—we’ve seen this happen with genres like K-pop or reggaeton. With time, emerging genres like alt z & underground hip-hop might be absorbed by the major-affiliated structures, or centralize around a few independent players that will master the niche. Today, through these spaces, we see a new generation of artists-entrepreneurs releasing self-funded projects relying on organic and viral reach, and utilizing emerging platforms and alternative record label models to build their careers.