Big data, music streaming platforms and the social dynamics of music taste
By Jack Webster
Big data in the age of music streaming
The rise of music streaming platforms, such as Spotify and Apple Music, has contributed to an explosion of new forms of digital data about music consumption practices. As the digital platforms through which consumers access and engage with recorded music and creators distribute it, they are uniquely positioned to create immense volumes of data about what and how people consume music, individually and at scale. From data about what music people search for and skip, to demographic information about who is consuming what, music streaming platforms generate data about almost every micro-interaction with music, amassing enormous databases ripe for further value-extraction.
The proliferation of big data in the music marketplace has the potential to transform the production and consumption of recorded music in new and interesting ways. From the production perspective, big data is shaping how music is created and composed. It has paved the way for AI-generated music compositions, which have the potential to take over the composition of more incidental forms of music, such as peaceful piano music. For the consumer, big data has enabled music streaming platforms to personalize the experience of consuming music in increasingly intimate ways through the creation of personalized playlists and recommendations. Unsurprisingly, the explosion of big data in the musical field has attracted much scholarly attention. One question social scientists have begun to consider is how music streaming platforms are shaping the social formation and function of music taste (I provide a more detailed examination of this here).
The social dynamics of music taste
Cultural taste – what and how we consume culture – is a product of our social background. The experiences, ideas and cultural forms we are exposed to growing up, such as from the TV programs our parents put on to the schools we attended, shape the formation of our taste. As we take these preferences into our everyday lives, they mark us as belonging to particular social groups or as coming from particular social backgrounds. Stereotypically and in a particularly British context, we might think about ‘posh’ people listening to classical music and eating foie gras, with ‘working class’ people being more interested in ‘popular’ forms of culture, such Top 40 music and pie and mash (my favorite!). Over the years, social differences in cultural taste have become more nuanced (for example, read about the rise of cultural omnivorousness here), but the point remains the same that taste is tied to social difference. When all is said and done, having ‘good’ taste matters because it helps us to access social and economic opportunities, such facilitating entry into the education system or helping people to network by being able to ‘talk the talk’ of those in positions of power (for an excellent study about this, check out this book). In age of music streaming and big data, has this relationship between music taste and social background changed?
Social leveling of music taste
On the one hand, music streaming platforms have democratizing potential. Putting aside debates about how much these services compensate musicians, where some people have challenged the democratizing potential of the format (you can read more about this here), and instead focussing on abundant choice, these platforms achieved something quite radical. Music streaming platforms have made available the history of recorded at little or no cost to anyone with an Internet connection. They have further untethered music access from physical (and social) spaces, making it less risky for us to indulge in our ‘guilty pleasures.’ And they provide forms of curation (such as editorially-curated playlists) that provide us with efficient means to broaden our musical horizons, should we want to do so. In principle, anyone, irrespective of social background, can listen to any kind of music and cultivate broad and diverse taste. This undermines the potential for differences in music taste to reproduce social differences because the notion that any style of music belongs to any one social group is brought into question. This is best captured in one of Spotify’s old tag lines: ‘music for everyone.’
Reproduction of social divisions in music taste
On the other hand, however, when we look closer at how music streaming platforms use big data, their democratizing potential is brought into question. In particular, examining how big data is used to personalize the experience of consuming music highlights how these platforms have the potential to reproduce social divisions in music taste at an unprecedented rate and scale. Take collaborative filtering, for example, which is a widely used technique for generating personalized playlists and recommendations. This technique works by examining the similarities in the listening histories of groups of users and uses the listening histories of those deemed to be similar to identify content to recommend to an individual. These kinds of techniques are problematic because they take music taste, which we know is socially-divided, and uses it as a basis for grouping people as similar or dissimilar. As a consequence, music recommendation technologies may be trapping us in musical ‘filter bubbles,’ exposing us to music that is reflective of our taste (and by extension our social background), rather than introducing us to the unfamiliar.
Looking to the future
These two contradictory forces make it unclear whether we have – or will – see a break between music taste and social background. Sociology has taught us that social divisions tend not to radically disappear; rather, the ways they manifest change in response to broader shifts in society, culture, politics, and the economy. With these words of caution in mind, we should think about how music streaming platforms are becoming a part of, and in turn shaping, how taste is socially formed and performed. What are people doing with the abundant access to music made available to them, or how are they interacting with and responding to the increasingly personalized experience of using music streaming platforms? Focussing on changes in practices may help us to better understand how big data and music streaming platforms are transforming the social dynamics of music taste now and in the years to come.
About
Jack is a social scientist with experience researching in the music and technology industries. His research addresses the impact of platformization on the production and consumption of cultural forms.