Spotify Teardown: Inside the Black Box of Streaming Music

by Maria Eriksson, Rasmus Fleischer, Anna Johansson

An innovative investigation of the inner workings of Spotify that traces the transformation of audio files into streamed experience.

Spotify provides a streaming service that has been welcomed as disrupting the world of music. Yet such disruption always comes at a price. Spotify Teardown contests the tired claim that digital culture thrives on disruption. Borrowing the notion of “teardown” from reverse-engineering processes, in this book a team of five researchers have playfully disassembled Spotify's product and the way it is commonly understood.

Spotify has been hailed as the solution to illicit downloading, but it began as a partly illicit enterprise that grew out of the Swedish file-sharing community. Spotify was originally praised as an innovative digital platform but increasingly resembles a media company in need of regulation, raising questions about the ways in which such cultural content as songs, books, and films are now typically made available online.

Spotify Teardown combines interviews, participant observations, and other analyses of Spotify's “front end” with experimental, covert investigations of its “back end.” The authors engaged in a series of interventions, which include establishing a record label for research purposes, intercepting network traffic with packet sniffers, and web-scraping corporate materials. The authors' innovative digital methods earned them a stern letter from Spotify accusing them of violating its terms of use; the company later threatened their research funding. Thus, the book itself became an intervention into the ethics and legal frameworks of corporate behavior.

Genres: Music, Nonfiction, Technology, Business, Science

272 pages, Kindle Edition

First published January 1, 2019

About the author

Maria Eriksson is a social anthropologist and a PhD candidate in the Department of Culture and Media Studies at Umeå University.

Reviews

[David Dinaburg · Rating 4 out of 5]

“Platform scholars need to triangulate by relating user participation, computing technology, and economics in one way or another.” I went for Spotify Teardown: Inside the Black Box of Streaming Music because of positive experiences with I AM ERROR: The Nintendo Family Computer / Entertainment System, Spam: A Shadow History of the Internet, and Racing the Beam: The Atari Video Computer System. This genre—Platform Studies—has become a favorite, though I wouldn’t recognize its roots spanned over a decade; the urtext Twisty Little Passages: An Approach to Interactive Fiction was written by the co-author of Racing the Beam, and that book, which I uncovered because of an adolescence spent with Multi User Dungeons [MUDs], primed me for scholarly breakdowns of how modern commercial platforms work. I knew things in this niche field were interrelated, but imagine my joy when Teardown cited Racing the Beam by name within the first dozen pages!

A deep dive into Spotify’s software—software I have used over the course of my life for more hours than I ever played the NES (heck, just on The Lumineers’ album Cleopatra alone)—is catnip. What makes Teardown unique in the world of platform studies was how it peeled back layers of purposefully misleading market positioning and dug at the nature of branded content and playlists:

Not only are the in-house playlists (Filtr/Sony; Digster/Universal; Topsify; Warner Music) thematically tailored to match advertisers’ potential target groups, they can also be sponsored by advertising clients. Moreover, as musical discovery through playlists is a prominent selling point for Spotify, playlists work as promotional devices for record labels and musicians. Because curation “has become a neutralised marketing term for taste-making and gatekeeping,” the selection and inclusion of specific artists on Spotify-curated playlists—some of them with millions of followers—have enormous effects for building a fan base and for increasing the number of streams and generating more revenue...meanwhile, Spotify keeps asserting the independence of its in-house content curators.

Spotify is not a “platform” for musical access in the way that an Atari or Nintendo were designed to give users access to video games, but a brokerage for advertisers to attach their wares to thematic events that are delimited by musical cues. Make a playlist for “morning commute” and you’ve got a self-selected market ready to go, primed to hear ads about office clothes and weekend getaways.

This non-optional opt-in of user data—I can’t use my iphone if I don’t give Apple access how often I touch it, I can’t fix my John Deere without proprietary diagnostic software—is about a phony as it gets; the illusion grows fainter and fainter when it is access to user data that makes Spotify valuated at billions of dollars. This more recent push for “music for all moments” is a transparent bid to increase the consumption of music and expand the reach of their tracking metrics by positing ubiquitous streaming as essential to productive life. If I go for a run, there is a playlist that matches my targetted run pace; If I’m studying, the Brain Food soundtrack has me covered:

While the idea that music can be used to control one’s body and mind is not new, the mode of “ubiquitous listening” facilitated by streaming services seems to correlate with a broader turn toward a utilitarian approach to music, whereby music consumption is increasingly understood as situational and functional for certain activities (rather than, for instance, a matter of identity work or an aesthetic experience).

Algorithmic recommendations lean hard into the recombination of tracks into playlists, and Teardown answered my single greatest Spotify question—does listening to the personalized “Discover Weekly” playlist solidify the algorithmic trajectory of my profile, self-reinforcing its decisions like a little musical filter bubble?

I can safely say, “Maybe.” I know the format of Spotify before their curatorial turn—type in a song and it plays right away—had a place in the world for people who knew their own music taste; it was perfect minus the lack of the Violent Femmes live album which has been unavailable for streaming since forever and I would really like to hear it again. But if you don’t know what you want, don’t leave it up to the algorithmic crawl of the robots. It’s pretty clear they don’t care what you they feed you, as long as the consumption is measurable.

[Artem Gordin · Rating 2 out of 5]

Don't get me wrong - maybe it's a good academic publication. But it's a bad book: as someone mildly interested in Spotify as a force in music, culture and business, I gained zero new insights from it. The book is academically thorough, but absolutely descriptive, telling any informed reader only what they've already known, but in a way you can quote in further articles.

To my dismay, it made no mention or attempt to look at Spotify internal tools (like PUMA, BaRT etc) that shape the music habits of millions. It doesn't show anything about how Spotify manipulates the vagueness of their popular playlists definitions to insert artists pushed by labels, or how big three gain advantages over indie labels through Spotify systems (or maybe they don't?).
The book mentions artists experience many times, but makes no effort to analyze the actual impact of Spotify model on music - like how 30 second payout rule actually shapes songwriting in 21st century and how some changes in algorithms impacted even established artists careers through weird "butterfly effects".

I guess all of the above is hard to do with proper academic rigor, as the book talks at lengths about how hard it is to apply proper research methods to an opaque and even hostile private organization, but it just makes for a terrible read.

All in all, if you're in academia and want to study music streaming, this book can be a good start and basis. If you're a layperson who wants to know how Spotify actually works and how it impacts culture, you're better off with many other media sources.

[Erkan Saka · Rating 3 out of 5]

This is definitely one of the most detailed accounts of Spotify's progress. Despite the average rating I gave, I suggest anyone who is interested in the relevant issue to have a look at this book.
What I could not ignore is the fact that the authors have such an intense ressentiment against Spotify. One cannot help but feel it throughout the book. I wish the authors could have a more distanced positioning.
Moreover, I admire that the authors engage an algorithmic deciphering of Spotify's working. However, I could not help but feel again that the authors were beginners in quantified/ experimental methods and I was not persuaded with their findings and approaches...

[Arielle · Rating 3 out of 5]

I think I went into this with the wrong expectations. I thought it would be more of a narrative but it was written like an academic paper (which makes sense, considering that it's research). It was insightful but they really took a topic that I think would interest a lot of people and turned it into something that will make people's eyes glaze over as they read.

[hennec · Rating 4 out of 5]

Hab mich für meine Bachelorarbeit durch diese Buch gefressen! An vielen Stellen gibt es einen kritischen Blick auf Spotifys Praktiken, liest sich gut (nur an Stellen, in denen es sehr wirtschaftlich wird, ist es bisschen pain)! Untersuchungen zu Spotify sind schwierig umzusetzen und stehen sowieso noch sehr am Anfang! Dafür bildet dieses Buch einen guten Einstieg!

[Tefo Mohapi · Rating 5 out of 5]

The matter-of-fact style that the book is written in is quite refreshing. It is understandable once you learn that it is written by researchers who used various methods including reverse engineering and bots to learn more about Spotify's inner workings. Also refreshing is how the authors always admit their biases and write about where they were wrong. For me, having tracked music streaming for over half a decade, some of the revelations were already known but nice to have hard evidence to back them up. In short, Spotify is not what it says it is, it is actually more like Facebook. The book details almost exactly how Spotify ACTUALLY works.

[Artem · Rating 2 out of 5]

Combative, dilettantish, barely technical and naive.

[Emily · Rating 3 out of 5]

Lot of interesting ideas and questions in here, but it's way too much of an academic research study to be something you can just...like...read. The language is clear enough but the content is always just a few degrees sideways of what I'd be interested in, the focus too much on the academic literature and research variables to go down cool tangents. I just happened to see the book referenced in this article and thought it sounded fascinating, but the audience is definitely folks well versed in digital humanities scholarship.

I hope I can read a more conversational / journalistic book soon that builds off of what these authors are doing, because it's all information worth unlocking from the "black box." (For example, yes, the algorithmic recommendations are exactly as repetitive as you think they are.) Or questions worth asking about a service I use daily (hmm, what exactly is Spotify's product and who are they selling it to? Access to music? Music curation? User data?).

[Daniel Cukier · Rating 5 out of 5]

I liked this book, not only because of the deeply description of how Spotify works from inside, but also because of the scientific rigor on which it is written. It is a must read for all music intusiasts, as it marks one of the most important happenings of our time related to music industry and consumption.

We must put a special highlight on the debate the book brings about research ethics boundaries. The authors tried to contact Spotify to get support for their research. Spotify ignored the call and researchers had to push the limits of the Spotify Terms of Service in order to deliver a high quality research. Maybe, if Spotify decided to collaborate, the research wouldn't go so far... Paradoxes of science...

[K · Rating 4 out of 5]

3.5 stars. Groundbreaking observations in a study that's shockingly narrow in scope. The quality of writing made me sad and will probably severely limit who ends up reading this. That's really too bad because some of the group's conclusions should be shared far and wide.

[Joey Lee · Rating 3 out of 5]

A bit disappointing in the terms of the quality & results of their studies, but good in terms of explaining the history of spotify, how they have changed and why.

[Maks Kolodziej · Rating 4 out of 5]

The book "Spotify Teardown…" by Eriksson et al. is an overtly thorough research book about the music streaming service Spotify, exposing Spotify's shady beginnings amidst the reign of terror during the Pirate Bay era, delving deep into the technological, financial, cultural, and political aspects of and motivations of the company while providing an explanation of Spotify's impact on the music industry.
Throughout these heavily researched chapters, the author's approach to creating an engaging narrative of the book is to create an intriguing structural pattern of its chapters. The book's main chapters lay a foundation for a specific idea or theory, which later is explained in the chapters and subchapters of the book, known as 'interventions,' after which an extensive analysis of their experiments in the 'intervention' subchapters is made from their hypothesis results. A conclusion of their findings later follows the hypothesis results. I found that, while the lengthy chapters were manageable to read (such as 33 pages in Chapter 2: When Do Files Become Music?), for an average reader, many of these chapters and 'interventions' could be regarded as 'too technical,' where their complexity grows exponentially with the introduction of new terminologies and concepts, like in the case of chapter' Interventions: Too Much Data.'. I found myself needing to re-read certain parts, especially this chapter, which could have been avoided if the authors had provided more concise resolutions to their chapters.
Overall, "Spotify Teardown" is an insightful read that thoroughly analyzes one of the most influential companies in the music industry in the 21st century. I appreciated this book's analytical moments as they came from a genuine interest and investigative nature. However, the overtly scrutinized tone intertwined with humorous moments could be regarded as discrediting the book's seriousness. The book's target audience is mainly aimed at students and professionals in the field. However, casual readers who are interested in the inner workings of tech companies and the music industry might also enjoy it.

[Dan Watts · Rating 2 out of 5]

I ended up browsing the book more than actually reading it. It has 2 problems. First and foremost, the writing is overly academic, focusing more on the social implications of streaming music than a description of how Spotify actually runs the business. Secondly, it is heavily biased against Spotify, without convincingly explaining what the problem is.

[Antoine Balaine · Rating 2 out of 5]

Darn this read was wordy. The dry info could have fit in 50 pages.

Though interesting, the playful tone that the book announced had a hard time staying on track and buried its findings under lengthy, conceptually vague term definitions. Discussing why your methodology is «gonzo» for pages on doesn't help your research.

I suspect a lot of this wordiness was due to extra caution out of fear for a spotify lawsuit.

[Ike Velez · Rating 3 out of 5]

I approached this book with heightened interest as a artist with a curiosity for media studies and digital culture. I wasn’t exactly a layman, so the academic approach (and extensive wordiness) didn’t immediately put me off at the start. There were plenty intriguing aspects to Spotify and their inner workings that were hinted at being explored, enough to keep me going through the prolonged tangents into their own research methodology. About midway through the book, I was beginning to think that the researchers were more interested in explaining how they did the experiments than their actual conclusions and findings, which even by their own admission left more questions than answers. The conclusion section itself seemed more concerned about explaining away the legality/ethics of their project than the actual results of the project, probably in the hopes of avoiding a lawsuit or absolve themselves of any guilt for taking a research grant for a study that they admit is incomplete. That last point may be too harsh, because I do believe the research was purposely obfuscated by their study subject Spotify by their design and through their legal actions. But throughout this read with its seemingly never-ending tangents, and overtly technical ‘interventions’, I really felt the writers were compensating for the underwhelming lack of true insight and revelations into the ‘black box of streaming’, the teardown of which was the main selling point for book.

[Jamie · Rating 2 out of 5]

While I am not an academic, and therefore may not be the authors' intended audience, I am a technologist at a leading music streaming service, and I was unimpressed by both their technical capabilities (they describe difficulties running simple scripted bots) and business familiarity. This was an illuminating survey of Spotify's history, but for anyone with any experience in any of the business or technical domains they're actually trying to understand (personalization, digital supply chains, etc.), I don't expect you'll learn much. I think their work would have been dramatically improved had they begun with background expert interviews -- even if Spotify didn't officially collaborate, jump on LinkedIn and find some people with industry background.

[Seth Holland · Rating 3 out of 5]

This book was alright. A little outdated with spotify being post-IPO and very profitable now. Some interesting thoughts from an STS perspective, what it means to be a media/tech company profiting off of music that is not your own, data and privacy concerns, the general juicy big tech dilemmas. Occasionally I thought it was too academic for it's own good, exhausting definitions simple words like "market" and "free" in a manner that may have sounded fine in an academic journal but seem long winded in a book. Glad I finally knocked this one off my list after having it on my shelf for about five years.

[James Mishra · Rating 1 out of 5]

This book operates at the wrong level of abstraction

The authors of this book are highly-credentialed but don't seem to think very clearly when it comes to matters about business. They tend to treat marketplace businesses as the same as businesses where the supply is under monolithic control.

They also waste a lot of time and space on highly-technical matters like running Wireshark on the Spotify Desktop client. I'm a software engineer and I think this is cool, but there isn't a whole lot in the pcaps when it comes to corporate strategy.

[Ogi Ogas · Rating 4 out of 5]

My ratings of books on Goodreads are solely a crude ranking of their utility to me, and not an evaluation of literary merit, entertainment value, social importance, humor, insightfulness, scientific accuracy, creative vigor, suspensefulness of plot, depth of characters, vitality of theme, excitement of climax, satisfaction of ending, or any other combination of dimensions of value which we are expected to boil down through some fabulous alchemy into a single digit.

[Mugren Ohaly · Rating 3 out of 5]

We take things for granted. Companies just want to make money and have access to people’s information. If we all stopped to consider how unhelpful Spotify is at introducing us to new music that we’d like and how poorly they pay musicians, we would surely stop using the service and go back to traditional methods of buying and listening to music.

[Kevin Hodgson · Rating 4 out of 5]

Fascinating look at the streaming music world, and a single company constantly repositioning itself, through an active research lens ... complete with complaints from Spotify itself about the ways the research was conducted with bots and fake record labels as a means to track the flow of data ...

[Matthieu Picard · Rating 2 out of 5]

While there were some neat ideas and findings scattered throughout, the tone/pace would be more fitting for an academic journal than a paperback. The book also felt significantly biased against Spotify.

[Andrey Goncharov · Rating 4 out of 5]

Minus one star for the way it’s written: after all, it’s a research documentation, not a detective story.

[Aj Sharma · Rating 1 out of 5]

Dry, and uninteresting.

[Eli Pales · Rating 4 out of 5]

Good research on Spotify as a social phenomenon and its history. Unfortunately, very little on the relationship between artists/labels and Spotify itself.

[Jenny · Rating 3 out of 5]

Interesting, but not very easy to read and somewhat inconclusive.

[Mark Southby · Rating 2 out of 5]

Somewhat interesting but seemed to switch between entertaining the reader and almost academic text.

[Nikoo · Rating 3 out of 5]

Skipped around a bit, but some interesting research in here for sure

[Nicole · Rating 3 out of 5]

What a weird book. Part academic exploration, part performance art. I’ll have to come back and write a more thorough review later.

[Maya Ordonez · Rating 5 out of 5]

senior thesis inspo