Journal: Music Curation in 2020

Algorithms play a huge role in music discovery (and these algorithms are akin to have a thousand friends working together to select your music), but the personal touch is still king in music discovery.
Share this:

Since I began writing about music twenty-three years ago, I’ve seen the music landscape change dramatically. The MP3 was not yet an industry standard, and YouTube didn’t exist (neither did Google). Sharing music online meant posting a massive WAV file, the raw audio format, which people could download over their dialup modems. This was so impractical that I remember seeing it happen once in 1997 (it was a Jeremy Enigk song from Return of the Frog Queen).

Music discovery relied heavily on word-of-mouth, effective record label marketing, live shows, television, magazines, or perusing through music at a record store. The record store is where I found most of my music at the time. I would walk to the store twice a week and flip through CDs. The friend who knew a lot about music had a limited voice, and could only be heard through one-on-one conversations (unless they had a `zine that they handcrafted, copied on a copy machine, and distributed). Curating music in 1997 largely centered around the person, being together physically.

Over the next five years, this would all change as the MP3 became the standard. It was easier to download music (still over dialup) because the files were small. MP3.com was the first site that successfully made listening to music accessible online, a site that launched in 1998. Artists could upload their MP3s, offer them for download, profit from their music on a “Pay for Play” model, and climb up charts. By 1999, MP3.com went public and raised over $370 million. Artists had the option to sell merch, such as shirts, coffee mugs, etc. with their art, and CDs were duplicated per order by MP3.com. The site peaked at 4 million MP3 downloads per day, an astonishing number even today. In 2000, MP3.com launched a locker service where you could upload the music from your CDs and stream them through your account. This made the music industry nervous, and a lawsuit eventually led to the site shutting down. Before it was shut down in 2003, a site called Napster was launched, and the industry couldn’t put the genie back in the bottle.

I share all of this because it’s important to know how we got to where we are today, if we are to understand music curation in 2020. Between 1997 and 2003, the shift from relying on friends turned to searches on these sites, queries on search engines, discovery on early music sites, and MySpace profiles (MP3 blogs wouldn’t be too far behind). The iPod, released during this time, would further turn music listening into a private and isolated experience. People needed a place to store their growing collection of music files. This was the beginning of the shift toward accessing music that led to music streaming services.

Moving forward twenty years, where are we with music curation?

To understand this, let’s look at where people are finding their music:

1 billion people listen to music on YouTube every month. That’s a staggering statistic.

271 million people listen to music on Spotify every month (124 million are subscribers).

175 million people listen to SoundCloud each month.

63.5 million people stream music on Pandora everything month

60 million listen to music on Apple Music every month.

55 million turn to Amazon Music every month.

14 million use Deezer to listen to music each month.

Bandcamp receives 31 million visits per month, though they don’t release their streaming statistics (which would include streams via their app).

661 million people in China stream from Tencent Music every month (35.4 million subscribe).

152 million people in India turn to Gaana for their streaming every month (1 million subscribe).

(I pulled these statistics from Musically.)

Needless to say, music curation takes place largely in these services. World of mouth is still very important (still the most influential form of finding music). But people will most likely to listen to something when they’re in the music streaming environment.

Spotify claims 36% of the global music streaming share. On average, people listen to 25 hours of music on Spotify each month. One-third of this time is spent listening to a Spotify-generated playlist, another third is spent in user-generated playlists.

Considering these stats, algorithms play an equally important role in music curation as Spotify playlist curators. Our listening habits (both personal and collective) influence what you’ll likely hear in these generated playlists. Once Spotify is able to identify what you enjoy listening to, the listening habits of other people come into play and Spotify will toss you song after song.

Playlists curated by people are equally important. When you consider that these playlists are likely more emotionally driven, the right curator can really connect with listeners. These playlists become a massively important player in today’s music.

Stepping outside of Spotify, let’s look at YouTube. While YouTube curators are able to create playlists, the influencers tend to work outside of the playlist. They post music on a regular basis that they select. When I think of dreampop, shoegaze, and bedroom pop, I think of people like David Dean Burkhart who has over 620,000 followers. He shares an album cover video (or a music video) every three to six hours consistently. You can go to his channel at 4am and find a newly posted song. His selections, and other curators like him, tend to influence the playlist curators on Spotify.

So how do human gatekeepers still influence music discovery in 2020?

Suppose David Dean Burkhart posts an album cover video of a newly released song on his channel. Within three days, the song is streamed 8,000 times. Dozens of curators hear the song and add it to their playlists on YouTube, Spotify, SoundCloud, etc. The song begins to get plays from these playlists. Burkhart also adds the song to his Compact Cassette playlist on Spotify, which has over 40,000 followers. Since the song is playlisted with similar songs, Spotify makes a connection between the new song and the songs in the playlists. In a week or two of gathered listener data, the song begins to appear in artist radio streams, Discover Weekly, and other Spotify generated playlists. YouTube connects the song with songs from David Dean Burkhart’s channel and YouTube playlists, and the song begins to gain plays from YouTube’s streaming algorithm. So while the song receives plays from algorithm dependent listening, it broke into these streams after first being selected by a human gatekeeper.

Other forms of human curation that we hardly think about, but are very influential:
Songs shared in Instagram stories are important. While they might not generate a lot of clickthrus, it is essentially a friend saying, “Check this song out!” If enough friends suggest the same artist or song, you’re more likely to look it up. Facebook tends to have much higher rates of clickthrus, so sharing music on that platform is also critical. Twitter, Reddit, SoundCloud reposts, and Shazam (or whichever service you use) all come into play. Never discount the importance of live shows, and effective marketing from record labels. And don’t discount college radio, whose DJs provide bursts of exposure to thousands of listeners (who Shazam a song to find it on their favorite streaming service).

While I believe that algorithms play a huge role in music discovery (and these algorithms are akin to have a thousand friends working together to select your music), the personal touch is still king in music discovery. When one of these gatekeepers approve of a song, the song will appear on user-generated playlists, which will eventually trickle down into the world of algorithm-based music discovery. Human gatekeepers and curators are still critically important in finding music in 2020.

Share this:
Kevin Flick
Kevin Flick

Kevin Flick founded Puddlegum in 1997 and has been engaging in music journalism off and on since. He's also a recording/mixing engineer and loves to help bands work through the creative process. He has coproduced and mixed for artists such as Saeyers, Midi Memory, and Cathedral Bells.

Rumors are that he's a Brighton Hove & Albion FC fan. He's also obsessed with coffee. Kevin resides in the college community of Bloomington, Indiana, where he studied at Indiana University.

Articles: 310

Leave a Reply!