“Searching for Sugar Man” the long tail of music streaming
In my previous post I spoke about a time when we were exposed to less music, owned a lot less of it but enjoyed it a lot more.
Today we constantly want more and so the insurmountable choice of music available to us should be a good thing, but as we have all discovered, the choice is too overwhelming.
This is where streaming services offer several solutions and do a lot of the heavy lifting by curating their content and serving it up to you on a silver platter. One of the side effects is that there are millions of songs that never get played — and a lot of amazing artists that never get discovered (Read more about the long tail here).
Why is this? Well for the same reason as Radio, streaming services play the most popular music to try and appeal to a wider audience. Most playlists are hand curated so songs that don’t have heavy rotation by users or are not easily visible to the curation teams, get lost in the noise. Spotify has over 4 million songs that have never been streamed. Some could argue that is because they are simply not good enough or it is filled with karaoke versions and covers, which is partly true, but what if 1% of that is simply amazing. Just think about Soundcloud for a second - their main content strength comes from these independent artists which are part of the long tail on other streaming services.
Being a resident DJ in Ibiza, I was lucky enough to be invited to Soundcloud at a very early stage. With a healthy dose of scepticism, I proceeded to upload a mix I had recorded in a club the night before. Within a matter of minutes, I had people asking me for song names and download links. For me, it was a much simpler and cheaper solution than burning my mixes on CD, but more importantly I could talk with my followers and be part of an ongoing conversation. Previously I would hand out a cd with my details on it and never hear of that person again.
After the DJs had joined, new and unsigned artists started uploading their content and promote it through social media. These artists now had their own voice and were in control of promoting their own music. Some of them would also have their music on streaming services, but it was a lot easier to be “discovered” on Soundcloud.
Prior to 2012, you would have likely never heard of an artist named Rodriguez.
Rodriguez is an American Folk Rock artist from the U.S. who at the time of his first album release in the early 70’s was likened to Bob Dylan. His future looked promising but his album completely flopped. He remained an unknown artist in the U.S. and couldn’t make ends meet so stopped performing and worked in construction to provide for his family.
It wasn’t till a certain “Searching for Sugarman” was released in 2012, that the world found out who Rodriguez was…including the cult following he had in South Africa. What is so important about this story, is that the long tail is not necessarily a qualitative issue.
In the last few years, streaming services have been buying up data and analytics companies that can help them spot these trends and look at the social chatter to predict which artist will blow up. Spotify’s “Fresh Finds” pushes unknown artists that they have noticed through their data. The benefit of todays technology is that in the case of Rodriguez, these streaming platforms could have picked up on the sheer popularity of his music amongst young South Africans through their data, which in turn would have pushed him into other territories to be “discovered”.
This doesn’t necessarily solve the whole long tail problem though, as we still need this element of “fresh finds” and random discovery incorporated into our everyday music journeys and playlists. I believe the key is in how we tweak our recommendation engines.
As a DJ I had a formula for the balance of popular songs vs. brand new unknown songs vs. vocal songs vs. instrumental songs that I would include in a set and in which order I would play them (for example: 2 popular, 1 unknown, 2 1 popular, 1 vocal, 2 popular, 2 unknown, 1 instrumental etc.). This was an ever evolving formula and changed with every single gig, but it was so crucial in defining the vibe for that night and actually capturing the crowd and taking them on a journey.
What do I mean by that? Well if you are listening to a playlist that just absolutely nails it song after song and vibe, but then throws a curve ball at you with something you have never heard of, this can be potentially very off putting. But, if this is a song that is selected, because it fits really well with the previous songs and very nicely with the next one, chances are, you will be very accepting of that anomaly track. Two songs fitting together, as if a DJ had picked them, allows for greater “random” discovery.
Pandora opted for the human curated music catalogue which means they have a lot less long tail and a better fit between songs than most of their competitors — the downside is that their catalogue is predictable and does not allow for random discovery. The problem for streaming behemoths like Spotify and Apple Music, is that there is a lot of junk in their 30+ million catalogues, which makes it a lot harder to filter out the gems.
Spotify’s own curation team, use a tool called “Truffle Pig” to find music to add into their curated playlists. Obviously that is part of their IP, but can you imagine the playlists Spotify users could come up with if they had access to the same tools for better search? I think this would be simply amazing.
At Muru, we made a very conscious choice to focus a lot more on the long tail, niche artists and underground tracks — Not because we don’t like the hits, but because we believe this is where a lot of the magic is happening. We want to have our ears to the ground to ensure we can provide recommendations that combine your favourites tracks with absolute gems that you would otherwise never find.
It would be a shame if our generations’ Rodriguez would get lost in the noise and have to give up making beautiful music.