If you listen to music right now, there’s a good chance you haven’t chosen what to put on – you’ve outsourced it to an algorithm. The popularity of recommendation systems is such that we’ve come to rely on them to serve us what we want without even having to ask, with music streaming services like Spotify, Pandora and Deezer all use custom systems to suggest suitable playlists or tracks for the user.
Usually these systems are very good. The problem, for some, is that they might be just too good. They understand your taste, know exactly what you’re listening to, and recommend more of the same until you’re stuck in an endless pit of ABBA recordings (just me?). But what if you want to break out of your usual routine and try something new? Can you train or trick the algorithm to suggest a more diverse range?
“It’s tricky,” says Peter Knees, assistant professor at TU Wien. “You probably need to point it very directly in the direction you are already interested in.”
The problem only gets worse the more you rely on automated recommendations. “When you keep listening to the recommendations that are being made, you end up in that feedback loop, because you are providing additional evidence that this is the music you want to listen to, because you are listening to it,” Knees says. This positively strengthens the system, prompting it to keep making similar suggestions. To get out of this bubble, you’re going to have to listen to something different quite explicitly.
Companies like Spotify are secretive about how their recommendation systems work (and Spotify declined to comment on its algorithm specifics for this article), but Knees says we can assume most are heavily based on collaborative filtering, which makes predictions of what you might like based on the tastes of other people who have similar listening habits to you. You might think that your musical taste is something very personal, but it’s probably not unique. A collaborative filtering system can create an image of taste groups – artists or songs that appeal to the same group of people. Really, says Knees, it’s not that different from what we did before streaming services, when you could ask someone who liked some of the same bands as you for more recommendations. “This is just a continuation of that idea backed by an algorithm,” he says.
The problem arises when you want to step away from your usual genre, era, or general taste and find something new. The system is not designed for this, so you will have to put in some effort. “Frankly, the best solution would be to create a new account and really train it on something very different,” says Markus Schedl, professor at Johannes Kepler University in Linz.
Otherwise, you should actively seek out something new. You can search for a new genre or use a tool outside of your main streaming service to find artist or track suggestions and then search for them. Schedl suggests finding something you don’t listen to as much and starting a “radio” playlist – a feature of Spotify that creates a playlist based on a selected song. (However, these can also be influenced by your broader listening habits.)
Knees suggests waiting for new releases or regularly listening to the most popular songs. “There’s a chance the next thing that comes up is your thing,” he said. But moving away from the mainstream is more difficult. You will find that even if you are actively looking for a new genre, you will likely be pushed towards more popular artists and tracks. It makes sense – if a lot of people like something, you’re more likely to, too – but it can make it difficult to find hidden gems.