The Mechanics of Music Streaming: The Five Trends Shaping the Streaming Business
Originally published at soundcharts.com
The adoption of streaming is by far the most significant shift in the industry in the last ten years — and that’s a shift that is still ongoing. Even today, streaming services are looking for ways to expand their user-base, develop their product, grow the revenues, and find a sustainable long-term business model. The music industry is caught in the middle of that process: streaming has already changed the way the recording industry operates and the way we consume, share and experience music — and that might be just the beginning. That’s why we’ve decided to take a close look at the mechanics behind the music streaming and answer the question: how the streaming business works and how will it continue to impact the music industry?
History of the Music Streaming Market
So, let’s rewind the last 20 years or so. The year is 1999, and there’s a new exciting service on the web, getting traction amongst the college students in America — Napster. It allows to exchange data through peer-to-peer networks with ease, and since .mp3 files are so well-compressed, music becomes the primary content transferred through the P2P. Almost immediately, RIAA files a lawsuit, that will drag on for eight years, costing Napster and its investor Bertelsmann over 300 million in settlements. However, the content of P2P music sharing is out in the open — and the attempts of the recording industry to shut down the pirate websites are a bit like the Lernaean Hydra fight. As soon as a pirate service is shut down, two new ones take its place. LimeWire, Morpheus, BearShare, Shareaza — the list of Napster offspring could go on.
Streaming as an Answer to Piracy
The legal approach proved fruitless, and revenues were plummeting across the recording industry. Various legal download-to-own alternatives entered the market in the early 2000s, with the most notable example of iTunes, integrated into Apple’s ecosystem — but even it couldn’t match the appeal of free pirate services. By 2008, by IFPI estimations, 95% of all digital music was downloaded illegally. The market was in desperate need of business models, that could compete on a par with the pirates. Thankfully, it wasn’t long until the answer was found: around 2007 Music as a Service, Open Music Model, or simply music streaming came to the rescue of the recording industry.
It’s hard to nail down what was the first streaming service. YouTube launched in 2005, so did Pandora’s interactive radio service. Soundcloud, the first on-demand service to focus on music content, launched back in 2007. However, if the question is “what is the first DSP to apply the streaming model the way we know it today”, the answer is clear. Spotify was founded in 2006 and publicly launched in Europe by 2008 as the first legal streaming service with an extensive, all-encompassing catalog.
The Complete Classification of the Music Streaming Services
1. Spotify as the Pioneer Brand
As the first product in the category, Spotify earned itself a status of a reference brand: if you live in Europe or the U.S., you probably think Spotify when you hear “music streaming” — just like you think Google when you hear “web search” (although Spotify hasn’t become synonymous with the category, as Google did). This fact alone became a huge pioneering advantage of Spotify from a branding standpoint. None of the competitors were able to catch up with Spotify, at least in terms of paying subscribers — although NetEase Cloud and Tencent Music have long passed it in terms of monthly active users.
The streaming market is very complex: as of July 2018, there were over 200 DSPs that offer music streaming capabilities. All those streaming services are put into the position of late-entrant, having to differentiate their offer from Spotify in one way or another — simply because no one would use a streaming service X that would copy all the features of Spotify. Why would you, if there’s Spotify already? At the same time, when Spotify first entered the market, it has defined what the music streaming service is. So, if the competitor would make its product too different from Spotify, it would “fall out of the category”, meaning that it wouldn’t even be considered by the consumer as an alternative to the Swedish startup. Those two factors are the essence of the Spotify’s advantage: all the other players on the market have to find a balance between following in Spotify’s footsteps and distancing itself from it.
In that sense, all of the brands on the market can be classified based on how they overcome Spotify’s advantage.
Source: MIDiA Research; Music Industry Blog
2. Ecosystem Brands
The first group is the streaming services that are built as a part of a bigger ecosystem. All the major international streaming services fall into that category. Apple Music, Google Play Music, Amazon Music — all those platforms are able to challenge Spotify head-on by offering a streaming service, integrated into the bigger product universe. Of course, there are other reasons: from internal funding and talent capital to the brand image that comes with the name like Apple. However, imagine the promotional effect of having your streaming service pre-installed on every iPhone ever, packaging a limited version of Amazon Music into every Amazon Prime subscription, or leveraging Google Home user-base to sell premium subscriptions. Such synergies allow Apple, Google and Amazon to compete with Spotify directly, without differentiating their product too much. The various telco’s bundled streaming services, from Japanese D-Hits to Canadian Bell, would also fall into that category.
3. Local Brands
The second group is the local brands that have developed as an alternative to Spotify within the confinements of the local market. Chinese TME streaming services, with their 600 million users are perhaps the most prominent example, but others can be found all over the world. Russian Yandex Music and Zvuk, Indian Gaana and JioSaavn Music, Korean Melon, African BoomPlay Music, Anghami in the Middle East and North Africa — the list could go on. While Spotify and the ecosystem brands are ruling over Europe and the Americas, a lot of the markets outside of the western world have their own streaming landscapes, developed in the absence of global players.
As of late, international streaming brands turned their attention towards those developing countries. The industrial media are bombarded with news of Spotify, Amazon, YouTube Music and Deezer entering new markets — with a strong focus on India and MENA countries. Most of those regions already have established streaming brands of their own, which means that the global services (including Spotify), are put into the catching up position — and the question remains if they would be able to beat the local competition. One one hand, global players can come in strong, wielding their subscriber bases and unmatched marketing budget. On the other hand, they might still struggle with the local catalog and culture-specific curation — which is a huge success factor when it comes to domestic markets. Developing markets are likely to become the main battleground within the streaming industry in the coming years — we’ll get to it a bit further down the road.
4. Niche Brands
The third approach to overcoming the Spotify’s advantage is in differentiating the product itself and appealing to a niche audience. The core offer of the most streaming services is unlimited access to the complete music library of the world. Even if it isn’t complete per se — the point is you can find any song you can think of across Spotify’s 40+ million tracks. However, that all-fits-all approach of global streaming services leaves space for improvement, and here’s where the niche streaming services come in. The idea of niche streaming is that if you choose a specific customer segment — usually fans of a particular genre — and build a dedicated service for them, you would be able to create a better listening experience (including discovery, recommendation and so on) within that niche.
Classical music is the lowest hanging fruit, but several apps are applying the same niche approach across different genres. We now have BeatportLink for electronic music, GimmeRadio for Metal and even Tidal can be put into this category through the strong connection of its brand with the hip-hop world — although its catalog isn’t limited to the genre. The various Hi-Fi streaming services like QoBuz are also virtually niche brands, although focusing on audiophiles instead of the genre fans. Let’s face it, charging users $24,99/month for 24-bit FLAC streaming is hardly a mass-market positioning.
5. Semi-music Streaming Services
Last, but not least, we have the digital service providers that, while used for on-demand music streaming, doesn’t exactly fit the definition of a streaming service. The most prominent examples of such brands are the historical players on the market, that were here even before Spotify came around: YouTube, Bandcamp, SoundCloud, Pandora, and alike. Take an example of YouTube: even though it is primarily a video-sharing website, 47% of all music consumption in the world happens through the platform. Semi-streaming services have the most precious resource of all — active users, who employ the platform as the source of music — which means that they can leverage that audience and try to convert it into paying subscribers. YouTube Music, SoundCloud GO+, Pandora Premium — most of the historical DSPs on the borders of the streaming market have now crossed over, entering the fight for the consumer’s $9,99/month.
The 5 Trends Shaping the Music Streaming Industry
According to the analysis of international streaming market by Music Business Research, powered by IFPI data, as of 2011, over 80% of all markets in the world were dominated by the physical sales — that is if we don’t take into the consumption via pirate services, which were still in their prime (especially outside of Europe and North America). Fast forward to 2017, 85% of the markets were digital-first, and for all of those countries (except for Indonesia and Malaysia) streaming is the primary source of digital revenues. Streaming is now the primary source of recording revenue around the globe. The recording industry revenues, raising another 9,7% in 2018, seems to be on its way to match or even top the 2000–2001 pre-piracy peak — and the power behind that growth is the rapid development of streaming. However, what does the future hold for the likes of Spotify?
1. The Maturation of the Market
The lion’s share of the streaming growth we see today is due to the new consumers, making a switch from the previous generations of music distribution — whether its physical sales or digital downloads. Once that shift is completed, the growth is bound to dial down — in fact, we are already at the point where some of the more developed markets are reaching the max streaming adoption rate. Take an example of Scandinavian markets: in Finland, Sweden, Norway, Denmark and Iceland, according to IFPI data, streaming accounts for more than 90% of digital music sales, or about 80–85% of all recording revenues. The Scandinavian markets are the first mature streaming economies that completed the user-transition mentioned above — and in that sense, they can provide a sneak peek into the future of the global recording market.
Source: MIDiA Research
Take a look at the home of Spotify, Sweden. In 2017, streaming sales grew just 7%, down from 23% in 2015. By global standards, a 7% annual growth is almost a stagnation, given the fact that the worldwide revenues were up 41,1% over the same period. However, that’s an inevitable future of the streaming industry — the question is just how and when the various local markets will reach that mature state.
The streaming industry is fragmented: while some markets, like Scandinavian countries and, to a lesser extent, the U.S. are reaching adulthood, many regions are just entering their adolescence. The maturation trend, however, seems to take the upper hand. In 2016 and 2017, streaming revenues grew 60,4% and 41,1%, accordingly. In 2018, it was down to 34%. MIDiA Research forecasts that this trend will persist in the foreseeable future, with the global market reaching 7% growth rate by 2026.
2. Streaming and the Cost of Music
At the current state of the business, most streaming services struggle to find a sustainable financial model. Spotify reported a profit only once in over ten years of its history, in Q4 of 2018 — and the company projects to get back in the red in 2019. There’s every reason to believe that not just Spotify, but all major western services are yet to reach profitability — it’s just that Spotify doesn’t have a massive company behind it to absorb the losses like Apple Music or Amazon Music do. However, profitability is not imperative for streaming services — much more important is the valuation of the business. Which means that the question isn’t how will streaming services turn a profit — it’s how will they keep the momentum, given the slowing down market.
Source: Spotify
So first, to understand how the streaming services make money, we need to take a look at the cost structure across the segment. The first line of cost for any streaming service is the payouts to the right holders. In general, the payout is divided into:
- Mechanical royalties, paid to the songwriters for the right to reproduce of the recording every time the user chooses to play a song. That means that non-interactive plays on DSP, like ad-supported radio-stream of Pandora, doesn’t generate that type of royalties.
- Performance royalties paid to the songwriters for the right to publicly perform their music. Since the streaming distribution model implies that the consumer doesn’t own any of the songs, every stream is rendered as streaming service publicly performing a song to the user — even if it is enjoyed in the privacy of one’s headphones.
- Payout to sound recording owners that makes up the primary share of the overall per-stream payout. It is paid to the owners of the recording copyright — whether it’s the label, distributor aggregator or the artist itself ( if they went with a direct Spotify deal, for example). The recording payout is usually divvied up between the different parts of the recording chain, and the splits will hugely depend on the contract in place — which means that the recording artist ends up making anywhere from 20% to 100% of that money. For more detail on how the recording business works, check out our article on the Mechanics of the Recording Industry.
On the artist side, what the streaming service will pay you per stream will depend on the vast number of factors, from the stream type (ad-supported vs. premium) as well as the country where your song was streamed. While Spotify charges $9,99 for its premium subscription in the U.S., in India, the price goes down to just $1.67 — and the payout will differ accordingly. Not all streams are born equal.
On the streaming side, however, calculating the overall artist payout is much simpler — the DSPs would usually pay a set percentage of its revenue to the right holders. This percentage is determined through an ongoing stream of negotiations between the streaming services and the right holders (or their representatives): most notably the big three (which owns ⅔ of the streaming catalog) and global digital rights agency for the independent labels Merlin.
The exact figures that are a subject of those negotiations are rarely publicly disclosed. In 2018, according to the company’s annual report, Spotify had made €5,259 million in revenues, and €3,906 million, or about 74,2% of that pie was immediately deducted as the cost of revenue — a total cost of manufacturing and delivering the product. Apart from the payouts themselves, that figure would also include some expenses linked to maintaining the networks and delivering the content, which are hard to estimate. The total 74,2% figure can give us a top threshold for the total costs of payouts.
However, there are reasons to believe that the actual payout is below 70%. Around 4–5 years ago, Spotify communicated the payout rate of 70% — but since then the streaming service has negotiated the decrease of the payout rate across all of the major’s catalog on a condition of meeting specific revenue and subscriber growth targets. The negotiations have affected the cost of revenue: the current cost of revenue is, in fact, significantly down from 88,4% back in 2015. So, for Spotify, the reasonable guess would be to put the payouts at somewhere between 60% to 70% of the total revenue — and there’s every reason to believe the payout rate would fall in that range for all primary streaming services out there. Apple Music, for example, used to pay out 71,5–74% of its revenue (58% and 13,5–15% to the recording owners and songwriters, accordingly), but, presumably, it has negotiated the rates down on a par with Spotify. Given the power tensions between the recording industry and streaming services, major labels are interested in keeping the playing field leveled and competitive — so it’s reasonable to believe that approximately equal rates are applied across the board.
Source: Spotify
The rest of the streaming costs are hardly worth mentioning in that context. The market is still in its early stage, so the key to a company’s success is not in optimizing the internal costs structure. To come back to the concept of the value of the business as the primary financial goal for streaming services: in its core, the market valuation is a communal estimate of all combined future profits whether they are 10, 15 or 20 years from now given the fact that the streaming market is a growing industry with huge potential. So, the streaming services can play a very long game, securing the share of the market that they will turn into profits 10 or even 20 years from now. Profitability and customer acquisition costs are hardly relevant at this point — what matters is the share of the global market (with the reasonable assumption that streaming will become the global standard).
If we would simplify things a bit, every streaming service has two main ways to grow the market share. First is the horizontal expansion of the market — entering new territories to unlock new growth opportunities and broaden the addressable market. Second, is strengthening the positions on developed streaming markets: either by winning over the users of competing services or expanding the product to attract the non-streaming audiences.
3. The Global Expansion of Streaming
As we’ve mentioned before, more developed streaming markets are reaching the max adoption rate. That is to say that, even though the penetration of streaming in the U.S. was at 47,5% in 2018, the rest of the population is not interested in the current value of music streaming services. The maximum adoption rate of streaming is not 100% of the population — it’s the point when the new user flow will dry out (at around 50% penetration rate, based on the mature streaming markets stats).
There’s not much to be gained in terms of “new blood” in the U.S. and Europe, but at the same time, the streaming penetration in India is just 6,1%, which means that the streaming offer is yet to be introduced to the mainstream audience. MIDiA Research predicts that the long term growth of the global streaming market will be driven by emerging markets: Brazil, Mexico, China, and India — as well as major markets that are a bit late to the streaming party, namely Germany and Japan. Each local music market is a universe of its own and a topic for a separate discussion, but quickly skim through the more interesting ones:
1. China:
China is the fastest growing music industry in the world right now. It went from 12th to 7th place in the IFPI global recording market rankings, and the vast rise of streaming powers a big chunk of that growth. There are now over 600 million people logging into the streaming services every month in China. However, due to the local legislation and the country’s closed off digital space, those audiences are not accessible to the global streaming players. Instead, they are users of the local streaming services Q.Q. Music, KuWo, KuGou and NetEase Cloud. Despite the wide adoption of streaming, there is still a huge room for growth on the Chinese market, as the average premium conversion rate is just 3% (compared to 46% for Spotify). However, the premium subscriptions are not even the focus of the local players, as TME, the most significant player on the market, takes an alternative path to the valleys of its customers. We did a great piece on the Chinese streaming services — so check out our latest Market Focus: China if you want to know more about the country’s unique digital music market.
2. Japan:
The second biggest music market in the world, Japan is still a predominantly physical market, with the entire industry built around the CD-format. As of 2017, more than 80% of all recording revenues were generated by CD sales. However, the physical market is slowly deteriorating — and streaming remains the only real answer to falling CD sales. All major players are already on the market, from Spotify to Apple Music to Google Play and YouTube Music — as well as the local solutions, trying to tap into specifics of the market, like Sony’s Hi-Res streaming service Mora Qualitas. However, the existing services are yet to find the key to the Japanese market, as streaming a minor power with less than 10% revenue share. As Goshi Manabe, the international rep and adviser of the country’s leading digital music provider RecoChoku, has pointed out in our interview:
[Transition to the streaming model] has to happen, it’s inevitable — it’s just about how that’s going to happen. Even in Japan, nobody knows. Everybody’s just trying to position themselves.
For more information on the Japanese music market, home of the idol-system, that has gone global in the last few years with the likes of BTS and BLACKPINK, check out our Market Focus: Japan article.
3. India:
India represents a massive opportunity for the global streaming market in terms of new blood. According to local industry body IMI, only 150 million — or about 10% of it’s 1,3 billion population — are using music streaming services. On top of that, less than 1% of those 150 are paying for standalone streaming subscriptions — which means that the market has a vast untapped potential, both in terms of streaming adoption and conversion to the premium subscriptions. With such promise, no wonder that both Spotify and YouTube Music have launched their services in India in 2019 less than a month apart.
Here’s where the YouTube’s ecosystem service model shines: while Spotify made the headlines by getting over a million users in India in its first week, YouTube Music actually tripled that number, leveraging the 245 million local YouTube audience. However, global players will face a massive competition in the local brands such as Gaana and JioSaavn Music — which means that India is likely to become the main battleground of the global streaming market for the next few years, alongside the MENA markets, where competition is structured similarly. The local market is calling for an extensive standalone analysis article, which might follow shortly — so stay tuned.
Aside from entering new markets, any streaming service can expand and strengthen its positions in developed streaming countries. That, of course, has to do with attracting switching users of other streaming services. However, let’s leave the internal competition aside for now — we will explore that topic in the future — and focus on the market opportunities across the board. After all, the market is just 12 years old. The product potential has not been fully realized — and so there is still room for the development of the streaming offer, which would allow to attract new, non-streaming audiences into the system.
4. Smart-Speakers and Voice Controlled Devices
The latest forecasts project that by 2022, smart-speakers will reach 66,3 million households or 167,7 million people in the U.S. alone — matching projected streaming user base for the same year. Smart-speakers are one of the hottest topics in the music industry right now, promising a shift in music consumption (here’s our take on it) which will surely make an impact on the streaming market.
The studies have already shown that smart-speakers tend to drive music consumption, as the users notice that they’ve listened to more music, with extended listening sessions and a broader selection of artists — and where there’s music abundance, there’s more demand for streaming. According to NPR and Edison survey, 28% of smart-speakers owners saying that getting a device caused them to subscribe to a streaming service. Smart-speakers bring digital music to homes — and to a new type of lean-back listener. In that sense, the growing smart-speaker penetration holds considerable potential for the streaming market — especially for the ecosystem brands. Three of the Spotify’s closest competitors, Amazon, Google and Apple are also the top three players on the smart-speaker market — while all of the speakers have Spotify integration, there’s a massive synergy for the ecosystem brands. For example, Amazon Echo, the leading brand on the smart-speaker market, comes with 90 days of free access to Amazon Music Unlimited, and that leverage is predicted to put it ahead of Spotify in terms of the growth rate on the U.S. market.
However, alongside new opportunities, there’s are new challenges that come with the voice-controlled consumption, and metadata is by far the biggest of them all. Inconsistent or incomplete metadata is a complex issue, plaguing the entire industry — and a topic for a separate article. However, when it comes to smart-speakers the main issue is not in the inconstancy in the song's credits and the subsequent challenges of revenue attribution. The voice-controlled environment calls for a whole new type of metadata when it comes to music discovery and recommendation. Voice assistant turn people to informal, less structured quarries, which in the music world would mean a key transition: from putting the name of the artist or a playlist in the search bar to asking Alexa to play a type of song. As MD of Techstars Music Bob Moczydlowsky puts it:
“How we prepare for voice? What’s the metadata looks like? What’s the tagging look like? How do we read emotions there? […] In my house there’s my voice, my wife’s voice, my son’s voice, my daughter’s voice. We have one Alexa account, but we all have different preferences. So what I’m excited about [is] when I say “Alexa play a sad song”, I get something out like “Blood and Chocolate”. When my son says Alexa play a sad song, he gets “When will I see you again”. Then my daughter says “Play me a sad song” — and she gets the Abba song, because she is in love with that at the moment.”
To achieve that effect of personalized search and recommendation, the great deal of internal metadata has to be attached to every song in the streaming catalog, which is truly a Sisyphean task. Furthermore, that data should be personalized: essentially, the voice-mediated environment asks for a transition from mood playlists — which are served the same to the entire user-base — to personalized mood tags. There are several approaches to metadata and recommendation out there: from Pandora’s Genome project, based on human classification, to Spotify’s algorithmic recommendation, based on the user-generated playlist data and user taste profiles. However, while the current methods might work well on the smartphone screen, the voice-controlled environment is bound to make recommendation much more complicated. So, the question of how will the streaming market prepare for voice, and how will this new recommendation affect the music industry, remains wide open.
5. From Music to Audio: Streaming Services to Take on Radio
We live in the attention economy, which means that the streaming services compete with other mediums for the users’ attention — and if we’re talking audio content in general, the biggest competitor of streaming is radio. Often thought to be an outdated medium, radio still holds substantial power on some of the markets. Take the U.S. for example, radio is both the first channel for music consumption and the most powerful medium in the U.S., reaching 92% of Americans every week. In that sense, the radio-audiences are a massive opportunity for the streaming market. However, winning over the radio listener is not an easy task.
Source: Nielsen
First, the streaming will have to take over the car listening. Radio lives on the road: The U.S. 247 million vehicle population is perhaps the main reason behind the medium’s strength in the country. The car is the primary battleground of the streaming vs. radio standoff, as all of the major streaming services now offer solutions for in-vehicle listening. Apple has CarPlay, Google has Android Auto, and even Spotify has made its first steps in that space, confirming tests of its long-rumored voice-controlled smart assistant for cars.
Second, streaming has to diversify the content, moving from music-exclusive streaming to all-around audio streaming. Earlier this year, Spotify has publicly announced its ambition to become “the world’s leading audio platform”. Since then, we’ve been hit with the news of Spotify’s investments in the podcast space, curated podcasts playlists and new app design emphasizing non-music content. Spotify is the first streaming service to make bold moves in the non-music content space, and its ambition is more than winning over the Apple Music users by offering talk-show content.
Image: Spotify
Just a couple of days ago the news broke of the new “My Daily Drive” personalized playlist, that would merge music recommendation with news talk shows. My Daily Drive is the essence of Spotify’s strategy to win over the hearts of radio-audience. It is both designed for in-car consumption (the word drive should give you a hint) and mirrors the format of the music radio, mixing music and talk show content. Non-music content would bring new users into the market — Spotify already stated that focus on podcasts allowed it to “convince people who previously thought Spotify was not right for them to give it a try”. However, there might be more behind the Spotify’s all-around audio content strategy.
Streaming Services on a Lookout for Long-term Profitability
According to the company’s CEO, Spotify expects that in the future more than 20% of all listening on the platform will be attributed to the non-music content — and that is, in fact, a massive opportunity for building a sustainable, profitable financial model. As we’ve mentioned, around 60–70% of the revenues of streaming services immediately goes out of the door in the form of payouts to the music industry. This payout rate is not a problem in the short term since the streaming services don’t need to be profitable right now. In the long-term, however, they will have to find a way to monetize their user-base. Diluting the content is one of the oaths to profitability: non-music content is much, much cheaper for streaming services since it doesn’t have a 70% payout rate attached to it. So, if the company brings down the share of music content, it would be able to negotiate down the payout rates, in the way of “hey, Universal, music is just 80% of the consumption on our platform — it’s only reasonable if we cut the rates by 20%”.
In a sense, the streaming market is just testing out things right now — things that could potentially drive the profits in 20 years from now. There are several paths to profitability: crossing into neighboring sectors of the music industry and selling conjoint goods like concert tickets and merch; simply raising the subscription prices — and Spotify is already experimenting with all of them. In 2018, the company tested 10% price increases in Norway, and the “concerts” tab on the artist page is one payment system integration away from turning into an in-app ticket marketplace. In a sense, the streaming industry is at the crossroads, and it doesn’t rush to leaving it — but be sure that once it does, the shift in the music industry will follow.
From the music industry standpoint, any major decision of streaming services will have a huge effect. Let’s say that Spotify goes down the path of content dilution and cutting the payout rates — totally a double-edged sword. On the one hand, the content dilution would mean that the music business will have a claim on a lesser part of the streaming revenue pie. On the other hand, the pie itself will grow as new users join the system — after all, American radio stations don’t pay out performance royalties at all, stating that they provide “free publicity and promotion to the artist”. In sum, the streaming market will inevitably find profitability in one way or another — and the music professionals will have to stay agile and adapt their strategies accordingly.
