About Mark Mulligan

Music Industry analyst and some time music producer. Vice President and Research Director with Forrester Research

MIDIA RESEARCH 2018–2026 STREAMING MUSIC FORECASTS

MIDiA has just published its global music forecasts, with revenue and user numbers projected out to 2026 (a list of data items and countries covered are listed at the bottom of this post).

2017 was the last year of strong streaming revenue growth. From 2018 onwards streaming growth will lessen each year, falling from 29% in 2018 to 7% in 2026. The slowdown in revenue growth reflects maturation of developed streaming markets such as the US, UK, Sweden, Netherlands and Australia. Longer term growth will be driven by emerging markets such as Brazil, Mexico and Indiaas well as later adopting major markets Germany and Japan.

midia music forecasts

Even with this slowing rate there is a lot of growth left in the market – so much so, in fact, that the market will grow from $19.6 billion in 2018 (in retail terms) to $45.3 billion in 2026. This means the market will more than double.We also have all of the market numbers in label trade revenue terms, but we are focusing on retail revenues for a crucial reason: the difference between trade revenues and retail revenues will widen between now and 2026. This reflects a number of factors that will see streaming services improve their margins and thus widen the gap on label revenues:

  1. Rising publishing related rates
  2. Increased share of royalty pot going to non-music content
  3. Increased share of royalty pot going to non-label music
  4. Potential long-term future label rate cuts (e.g. relief for price hikes)

The last item won’t happen in the current round of negotiations with rights holders, but at the next round more power will lie with streaming services. Right now, Spotify poses a lot of threat in rightsholder eyes, but its actual power is more limited. Streaming revenues accounted for 51% of label revenues in 2018 and with Spotify accounting for less than half of that, this means that Spotify accounts for less than a quarter of total label revenues. The labels, publishers and right bodies need to maximize their negotiating power now, while they can.

This will change, and when combined with the other three factors, the conclusion is clear: label trade revenues are becoming a progressively less useful way of measuring the future size of the streaming market.

In subscriber terms, at the end of 2018 there were 278 million paid subscribers. However, due to the impact of family plan accounts, unique subscriptions were only 242 million. The 2 biggest streaming markets in 2018 (US and UK) will remain the largest by 2026 while large markets such as Germany and France will also still be large, leading markets. Brazil, India, MENA and China will all be established as top 15 global markets by 2026, with the first four each more than doubling revenue compared to 2018.

MIDiA clients can access the report and full dataset right now here. Clients can also explore the forecastsin our forecasts viewer on our data portal Fuse here.

The report and data are also available for purchase on the MIDiA report store here.

List of data points and markets:

  • Subscription revenue (retail values)
  • Audio Ad supported revenue (retail values)
  • Video Ad supported revenue (retail values)
  • Total Ad supported streaming revenue (retail values)
  • Total streaming revenue (retail values)
  • Downloads revenue (retail values)
  • Total digital revenue (retail values)
  • Physical revenue (retail values)
  • Other revenue (retail values)
  • Total recorded music revenue (retail values)
  • Subscription revenue (label trade values)
  • Audio ad supported revenue (label trade values)
  • Video ad supported revenue (label trade values)
  • Total ad supported streaming revenue (label trade values)
  • Total streaming revenue (label trade values)
  • Downloads revenue (label trade values)
  • Total digital revenue (label trade values)
  • Physical revenue (label trade values)
  • Other revenue (label trade values)
  • Total recorded music (label trade values)
  • Subscribers
  • Subscriptions (unique accounts)
  • Ad supported audio users
  • Ad supported video users
  • Subscriber ARPU (USD) – Retail values
  • Ad supported audio ARPU (USD) – Gross revenues
  • Ad supported video ARPU (USD) – Gross revenues
  • Subscriber ARPU (USD) – Trade values
  • Ad supported audio ARPU (USD) – Trade revenues
  • Ad supported video ARPU (USD) – Trade revenues
  • Audio streams
  • Spotify subscribers
  • Apple Music subscribers

Countries covered

  • US
  • Canada
  • Austria
  • Belgium
  • Denmark
  • Finland
  • France
  • Germany
  • Italy
  • Netherlands
  • Norway
  • Poland
  • Spain
  • Sweden
  • Switzerland
  • UK
  • Other Europe
  • Australia
  • China
  • India
  • Indonesia
  • Japan
  • Philippines
  • South Korea
  • Taiwan
  • Thailand
  • Other Asia
  • Argentina
  • Brazil
  • Colombia
  • Mexico
  • Other Latin America
  • Russia
  • Middle East and North Africa

 

  • North America
  • Europe
  • Asia Pacific
  • Latin America
  • Rest of World
  • Global Total

Datasheets included in the Excel document:

  • Global Summary – Retail Values
  • Retail Summary
  • Global Summary – Trade Values
  • Trade Summary
  • User Summary
  • Spotify + Apple
  • Country Summaries – Retail
  • Country Summaries – Trade
  • Top Streaming Markets – Retail Values
  • Subscriber Market Shares
  • Methodology

Preparing for the Post-Album Industry

This is a guest post by Keith Jopling, MIDiA’s Consulting Lead. It is a follow-up piece to What’s Next In Playlist Innovation?

Every week I’m still excited to check out the latest album releases. They are the gift that keeps on giving. But that gift feels different these days, more like receiving flowers or chocolate and less like anything you might say is a keepsake.

It’s becoming very rare these days for me to fall in love with an album the way I used to. I miss it, but there it is. Some of this is a conscious trade-off, since I enjoy compiling and curating playlists. But to some extent, it just feelslike I don’t have the time to give (i.e. invest in repeated listens) in the way albums – good ones – truly deserve.

Broader listening trends confirm that this applies to people in general. The penetration of adults that claim to listen to whole albums monthly, stands at just 16% (Q4 ‘18 data from MIDiA, a drop from 22% in the previous quarter), compared with say, 35% listening to music on the phone.

This is self-reported of course. Behavioural data on actual album listening is a patchwork of proxy measures, such as ‘listens (to individual album tracks) from the artist’s album page’ on Spotify. To be fair, we never ever really knew how consumers who bought albums actually listened to them. Survey data I saw a long time ago, before the streaming era kicked-in suggested that some purchased albums were played on average, just over once.

Competing in the attention economy

As the management consultants say, what can’t be measured won’t get done, and so in a world of real-time statistical feedback, why make an album if you cannot know who is listening and how? There is already a creative conversation in the industry about “skip-rate reduction” and one way to achieve this is to front load albums with the catchiest tracks, but where does that leave the art of album sequencing and story-telling?

The album’s competitive pressures go wider than music. In the attention economy, albums compete with Netflix, Fortnite, TikTok and Instagam. In music’s own attention economy, albums compete with singles, playlists, games, podcasts and box sets. And those latter two categories are literally stealing the show. This is unsurprising in the streaming world – a better way of coping with the content waterfall is a ‘consume-once-only’ approach. The idea of spending the same 45 minutes over & over to build up familiarity with a record seems taxing.

Albums are no longer water cooler moments

What seems particularly telling for music, is that ‘Netflix and shows’ is the water cooler conversation now. Pop culture talk is all about what you’ve watched, are watching or should watch, and any similar conversation about listening is conspicuous by its absence. The ‘event album’ seems to be over. Is it just me or do artists seem to want to drop albums with less fuss now anyway? Perhaps the element of surprise (Bowie’s legacy yet again) is smarter than facing the “aftermath of promotion”. But it’s also not without risk given the tonnage of new music flowing through. Coupled with this, some artists are either eschewing the format or at the very least questioning it. If we take the world’s biggest artist right now, Ariana Grande – what role did her album play in the scheme of things?

What choices does this leave labels and artists?

When the CD began to give way to streaming, it’s fair to say the labels embraced it. In some ways, the CDs decline was ushered in rather than managed out. Back around 2012 I was in the room when one label boss took hold of a CD and flung it across the room smashing it into shards. I was impressed if surprised. But are labels taking the same level of aggressive-progressive when it comes to succeeding the album?

I’ve already argued previously that the main format to succeed the album is the playlist, and that there is further innovation to go (also acknowledging Apple’s release regarding a more creative approach top playlist cover art). Meanwhile there is no doubt that the EP has made a comeback, and has become a useful vehicle for new artists to drop a collection of songs as a showcase of their repertoire.

But let’s take a look at some other options:

  • (Bring back) Album Exclusives:With some services so favouring the single track, could the labels divide & rule with full album licensing fenced off to other album focused services? Now I know we’ve been there before, and ‘nobody’ liked it much, but things are different now, and platform differentiation is a strategy both labels and platforms need when it comes to content. If some services stepped up in full support of the album, it stands a better chance longer term. A more radical option? Sell full albums exclusively on label-owned streaming services.
  • Physical Exclusives (through vinyl):Favouring both the true fan and the artist, perhaps vinyl should become the exclusive way to hear the whole collected work. With traditional vinyl at capacity one innovation I am keeping an eye on is Virylsteam-based manufacture. For artists looking to do something truly different, this is an option. Pre-order, custom versions, pre-sale and merch during tours, and pop-ups, as well as sell-through Amazon, Bandcamp and brick & mortar, the product sale potential is larger than it looks.
  • More visual outputs:Universal has doubled-down on video. Apple has doubled-down on video. Video remains huge, and continues to grow in short-form, and now long-form too. TV and theatrical productions are hunting very actively on music’s turf. It’s fascinating but by definition, not the same ubiquity potential of audio formats. Music films can do wonders for song catalogues however, Bohemian Rhapsdoy has proven that.
  • Experiment with entirely new formats:Easy to say, harder to do. New platforms such as voice and the car provide options without a doubt. Of course, an option for the producer sector is to do nothing much – simply wait for the technology sector to stumble on the next scaleable format. They are certainly trying, from Spotify’s Canvas moving images, to Pandora’s new Stories format, to Apple’s continued video format innovations such as Up Next. The platforms pitch these formats to artists as much as labels. One problem for labels is keeping up with the tail wagging the dog. Without knowing if the format will last, should they invest and convince artists to make stuff the platforms want? For example, should they make vertical videos just because Spotify wants vertical videos that month, or podcasts because it’s now all about podcasts, until it isn’t?
  • Expand into the live sector: Not easy. Wider representation of artists got a bad rap with the ‘360’ degree deals, but yet again, times have changed and a re-evaluation is due. With money coming in from streaming as well as outside investment, labels could buy smaller live promoters and venues.

What does this mean to the artist proposition and label economics?

Okay, so now we are down to the rub. The album is still the format that drives industry economics as a whole. The conversation around the artist proposition (and therefore the deal) has been changing for some years, but still essentially centres on the album as the economic unit. This must surely see more rapid change, to be replaced by agreed song numbers, or simply a time period covering numerous ‘artist projects’. We’ve already seen the first ‘lifetime’ deal between Elton John and Universal.

For the vast majority of artists, revenues are now following a pareto curve, with their ‘top songs’ (between five and ten say) making up a fraction of their catalogue but the large majority of their streaming income. When an established active artist releases a new album, the impact is often on those jewels in the crown more than the new collection, and if one or two songs make it into the crown, then bingo! The project pays off. The goal of any artist project is to get another jewel in the crown, but is an album the critical vessel to achieve that goal?

Perhaps the golden rule here is that the one size fits all model is becoming unfit for purpose. We’re already seeing some innovation, especially from Hip Hop artists like Migos, Drake and Kanye, but this trickle needs to turn into a flood. We’re getting to the time for ‘throw everything at the wall and see what sticks’.

The last days of All Killer No Filler?

I still use my primary source of the past 20 years to find new records, The Guardian G2 and Pitchfork. I can’t find any new albums via Spotify’s personalised feeds, because that’s mostly now just ‘for me’, or singles. Apple Music does a better job of presenting new release albums, and I guess both the streaming rivals are serving their own listener base appropriately.

Despite everything you read above, I feel the album will endure. The digital streaming age is a challenge to artists to make better albums, the ‘all killer no filler’ approach. The album as a canvas still feels relevant for some artists, but only some.

The album is already a niche in consumption terms, unmeasurable in streaming terms, but still the essential deliverable in the deal. That’s out of step, and those labels and artists with one eye on the mid-term future will already be planning for what the game looks like going forward. It’s what Reed Hastings calls “constantly worrying about what’s next” and it’s worked for him so far.

We don’t just write it. For a ‘post album world’ conversation with Keith & Mark contact us at MIDiA. We’ve already been helping labels, artists and managers rethink the album, drop us a line at info@midiaresearch.comto see how we could work with you.

 

Calling all Artists!

In partnership with independent distribution company Amuse, MIDiA Research is undertaking a detailed study of the music artist landscape. We are fielding a survey to the artist community, exploring issues such as:

  • What success looks like to you
  • Career aspirations
  • The importance of signing to a record label
  • Financial wellbeing
  • Maintaining creative control

If you are a singer, DJ, producer, performer, or in a band, then we’d love to hear your views. Just click the link to take the survey.

All of your responses will be treated as strictly confidential and will only ever be presented in aggregate as part of results for the entire survey – so never attributable to any individual. We will not use any of your responses to contact you again for any purpose, unless you specifically provide your email address to us in order to be interviewed in more detail for the research project.

We will also send you a summary of the findings so that you can see how you fit into the picture amongst your fellow performers, and benchmark yourself against their aggregate responses.

If you have any questions concerning the survey the please email us at info@midiaresearch.com

2018 Global Label Market Share: Stream Engine

Recorded music revenues grew in 2018 for the fourth consecutive year, reaching $18.8 billion, up $2.2 billion from 2017. Streaming was the engine room of growth, up 30% year on year to reach $9.6 billion. For the first time streaming became the majority of label revenue (51%), and its growth continues to outpace the decline of legacy formats. Major label rankings remained unchanged in 2018, but the majors enjoyed varying fortunes and the continued meteoric rise of Artists Direct points to market transforming changes that are reshaping the entire business of record labels.

2018 was shaped by three key factors:

  • Continued growth: Global recorded music revenues grew 7.9%. Though 2017 revenues grew by a higher 9.0%, the market grew the same in absolute terms in 2018, adding $1.4 billion of net new revenues as in 2017. Since 2015 the total market has increased by 26%, adding $3.9 billion of net new revenue.
  • Stream powered: Though relative growth is slowing, streaming added the same amount of net new revenue – $2.2 billion – in 2018 as it did in 2017. Though 2019 will see mature streaming markets such as the US and UK slow, mid-tier markets such as Mexico and Brazil, coupled with Japan and Germany, will ensure that streaming revenues grow by another $2 billion in 2019.
  • Artists Direct:The major record labels retained the lion’s share of revenues in 2018, accounting for 69.2% of the total. Changes in global market shares typically move at a relatively slow pace, particularly at a major vs independent level. However, Artists Direct – i.e. artists without record labels – are changing the shape of the market, growing nearly four times as fast as the total market to end 2018 with $0.6 billion of revenue.

midia music market shares 2018

There were mixed fortunes in terms of market shares. Universal Music and Warner Music both gained 0.6 points of market share in 2018, up to 30.3% and 18.3% respectively, with Sony Music losing 1.5 points of share in 2018. Though Sony’s 2018 revenues were constrained in part by the company implementing new revenue recognition practices in 2018, Universal’s market share lead over the second placed label is now an impressive 9.7 points.Artists Direct and Independents together accounted for 30.8%, though this figure is measured on a distribution basis (i.e. Major revenues include independent labels distributed by majors and major owned companies). The independent share based on an ownership share will therefore be higher.

More of the same, but change too

In many respects 2018 was a re-run of 2017: total revenues grew in high single digit percentage terms; streaming was the engine room of growth and added more revenue than the prior year; Warner Music gained most major market share; Universal Music added more revenue than any other label; Artists Direct gained most market share.  But it is this latter point that may say most about where the overall market is heading. The range of tools now available to an artist are more comprehensive than ever before, while deal types that labels are offering (e.g. label services, joint ventures) are changing too. Artists are effectively able to custom-build the right model for them. The market will always need labels, but what constitutes a label is becoming a fluid concept. And in so becoming, it may put us on the verge of the biggest shift in record label business models since, well, ever.

These findings are highlights of the MIDiA Research report: Recorded Music Market 2018: Stream Engine. If you are a MIDiA client you can access the full report, slides and datasets here. You can also purchase the report and all its assets here.

Here’s How Spotify Can Fix Its Songwriter Woes (Hint: It’s All About Pricing)

Songwriter royalties have always been a pain point for streaming, especially in the US where statutory rates determine much of how songwriters get paid. The current debate over Spotify, Amazon, Pandora and Google challenging the Copyright Royalty Board’s proposed 44% increase illustrates just how deeply feelings run. The fact that the challenge is being portrayed as ‘Spotify suing songwriters’ epitomises the clash of worldviews. The issue is so complex because both sides are right: songwriters need to be paid more, and streaming services need to increase margin. Spotify has only ever once turned a profit, while virtually all other streaming services are loss making. The debate will certainly continue long after this latest ruling, but there is a way to mollify both sides: price increases.

spotify netflix pricing inflation

When Spotify launched in 2008, the industry music standard for subscription pricing was $9.99. So, when its premium tier was launched in May 2009, it was priced at $9.99. Incidentally, Spotify racked up an initial 30,000 subscribers that month – it has come a long way since. But now, nearly exactly ten years on, Spotify’s standard price is still $9.99. Its effective price is even lower due to family plans, trials, telco bundles etc., but we’ll leave the lid on that can of worms for now. Over the same period, global inflation has averaged 2.95% a year. Applying annual inflation to Spotify’s 2009 price point, we end up at $13.36 for 2019. Or to look at it a different way, Spotify’s $9.99 price point is actually the equivalent of $7.40 in today’s prices when inflation is considered. This means an effective real-term price reduction of 26%.

Compare this to Netflix. Since its launch, Netflix has made four major increases to its main tier product, lifting it from $7.99 in 2010 to $12.99 in 2019. Crucially, this 63% price increase is above and beyond inflation. An inflation-adjusted $7.99 would be just $10.34. Throughout that period, Netflix continued to grow subscribers and retain its global market leadership, proving that there is pricing elasticity for its product.

Spotify and other streaming services are locked in a prisoner’s dilemma

So why can’t Spotify do the same as Netflix? In short, it is because it has no meaningful content differentiation from its competitors, whereas Netflix has exclusive content and so has more flexibility to hike prices without fearing users will flock to Amazon. If they did, they’d have to give up their favourite Netflix shows. Moreover, Netflix has to increase prices to help fund its ever-growing roster of original content, creating somewhat circular logic, but that is another can of worms on which I will leave the lid firmly screwed.

If Spotify increases its prices, it fears its competitors will not. Likewise, they fear Spotify will hold its pricing firm if any of them were to increase. It is a classic prisoner’s dilemma.  Neither side dare act, even though they would both benefit. Who can break the impasse? Labels, publishers and the streaming services. If they could have enough collective confidence in the capability of subscriptions over free alternatives, then a market-level price increase could be introduced. Rightsholders are already eager to see pricing go up, while streaming services fear it would slow growth. Between them, there are enough carrots and sticks in the various components of their collective relationships to make this happen.

However – and here’s the crucial part – rightsholders would have to construct a framework where streaming services would get a slightly higher margin rate in the additional subscriber fee. Otherwise, we will find ourselves in exactly the same position we are now, with creators, rightsholders, and streaming services all needing more. When Netflix raises its prices it gets margin benefit, but under current terms, if Spotify raises prices it does not.

The arithmetic of today’s situation is clear: both sides cannot get more out of the same pot of cash. So, the pot has to become bigger, and distribution allocated in a way that not only gives both sides more income, but also allows more margin for streaming services.

Streaming music in 2019 is under-priced compared to 2009. Netflix shows us that it need not be this way. A price increase would benefit all parties but has to be a collective effort. Where there is a will, there is a way.

Kobalt is a Major Label Waiting to Happen

Disclaimer: Kobalt is a label, a publisher as well as a Performing Rights Organisation (PRO). This post focuses on its label business, but does not presume to overlook its other aspects.

Lauv Kobalt

News has emerged of Kobalt potentially looking to raise an additional $100 million of investment, following a 2017 round of $89 million and a 2015 $60-million round led by Google Ventures. Kobalt has been the poster child for the changing of the guard in the music business, helping set the industry agenda by pursuing a creators-first strategy while

building an impressive roster of songwriters and artists at a scale that would have most indies salivating. But it does not have its sights set on being the leading player of the indie sector, instead playing for the big game: Kobalt is the next major label waiting to happen.

So, what makes Kobalt so different? In some respects, nothing. Most of what Kobalt is doing has been done before, and there are others plotting a similar path right now (e.g. BMG, United Masters, Hitco). What matters is how it is executing, how well backed it is and the scale of its ambitions:

  • Moving beyond masters: In the old model, artists signed away their rights in perpetuity to record labels, with nine out of ten of them permanently in debt to the label not yet having paid off their advances. The new model (i.e. label services) pursued by the likes of Kobalt, reframes the artist-label relationship, turning it one more akin to that of agency-client. In this rebalanced model artists retain long-term ownership of their copyrights and in return share responsibility of costs with their label. This approach, coupled with transparent royalty reporting, lower admin costs and continual tech innovation has enabled Kobalt to build a next-generation label business.
  • Laser focus on frontline: In a label services business the entire focus is on frontline, as there isn’t any catalogue. An artist signed to such a label therefore knows that they have undivided attention. That’s the upside; the downside is that the label does not have the benefit of a highly-profitable bank of catalogue to act as the investment fund for frontline. This means that a label like Kobalt often cannot afford the same scale of marketing as a major one, which helps explain why Kobalt is looking for another $100 million. However, there is a crucial benefit of being compelled to spend carefully.
  • Superstar niches: In the old model, labels would (and often still do) carpet-bomb TV, radio, print and digital with massive campaigns designed to create global, superstar brands. Now, labels can target more precisely and be selective about what channels they use. Kobalt’s business is based around making its roster superstars within their respective niches, finding a tightly-defined audience and the artists they engage with. The traditional superstar model sees an artist like a Beyoncé, Ed Sheeran or a Taylor Swift being a mass media brand with recognition across geographies and demographics. The new superstar can fly under the radar while simultaneously being hugely successful. Take the example of Kobalt’s Lauv, an artist tailor-made for the ‘Spotify-core’ generation that hardly registers as a global brand, yet has two billion audio streams, half a billion YouTube views and 26 million monthly listeners on Spotify. By contrast, heavily-backed Stormzy has just three million monthly Spotify listeners.
  • Deep tech connections: The recent WMG / Spotify spat illustrates the tensions that can exist between labels and tech companies. Kobalt has long focused on building close relationships with tech companies, including but not limited to streaming services. This positioning comes easier to a company that arguably owes more to its technology roots than it does its music roots. The early backing of Google Ventures plays a role too, though with some negative connotations; some rights holders fear that this in fact reflects Google using Kobalt as a proxy for a broader ambition of disrupting the traditional copyright regime.
  • A highly structured organisation: One of the key differences between many independent labels and the majors is that the latter have a much more structured organizational set up, with large teams of deep specialisation. This is the benefit of having large-scale revenues, but it is also a manifestation of ideology. Most independents focus their teams around the creative end of the equation, putting the music first and business second. Major labels, while still having music at their core, are publicly-traded companies first, with corporate structures and a legal obligation on management to maximise shareholder value. Kobalt has undoubtedly created an organisational structure to rival that of the majors.

Earned fandom

Kobalt is a next-generation label and it is plotting a course to becoming a next generation-major. That success will not be reflected in having the rosters of household names that characterise the traditional major model, but instead an ever-changing portfolio of niche superstars. The question is whether the current majors can respond effectively; they have already made big changes, including label services, JV deals, higher royalty rates, etc.

Perhaps the most fundamental move they need to make, however, is to understand what a superstar artist looks like in the era of fragmented fandom. The way in which streaming services deliver music based on use behaviours and preferences inherently means that artists have narrower reach because they are not being pushed to audiences that are relevant. This shifts us from the era of macro hits to micro hits ie songs that feel like number one hits to the individual listener because they so closely match their tastes. This is what hits mean when delivered on an engagement basis rather than a reach basis. Quality over quantity.

Majors can still make their artists look huge on traditional platforms, which still command large, if rapidly aging audiences. But what matters most is engagement, not reach. It is a choice between bought fandom and earned fandom. In the old model you could build a career on bought fandom. Now if you do not earn your fandom, your career will burn bright but fast, and then be gone.

What’s Next For Playlist Innovation?

This is a guest post from MIDiA Research’s Keith Jopling.

In this era of access to all music and everything about it, I do enjoy reading artist interviews, and pay attention to artists’ views on the modern music industry. What caught me recently were Mark Ronson’s remarks on songwriting in the age of the playlist in The Guardian:

“Everything has to be produced so it sounds competitively as loud as possible coming out of an iPhone or as loud as possible when it comes out of a Spotify hits playlist; you have to make sure the kick drum and the guitar have the same loudness and presence all the way through the whole fucking song or you don’t stand a chance. It’s kind of crazy how you have to think about music now.”

I have heard similar views from artists in recent years and they do have grounds, though you’ll be hard pressed to prove that streaming and playlists have truly altered music per se. After all how would that equate with the huge success of ‘slow music’ playlists, from Peaceful Piano to Esquenta Sertanejo to Your Favourite Coffeehouse?

There has certainly been an industry skew towards ‘streaming hits’ though. More artists are making “Spotify-core” (i.e. music that will do anything to avoid being skipped) and that’s a genre perhaps lacking in subtlety, But can it be accused of crowding out other, subtler types of music?

I would argue that playlists have changed the way we listen more than they have changed music itself. If the first port-of-call for music consumers is to hear songs on playlists, a lot of context around the music is lost. Already, research has shown that listeners are hard pushed to recall song titles, let alone the albums those songs are taken from. Speaking of albums, it strikes me as high time the industry now evaluate just what a ‘post-album’ world will look and sound like. So perhaps somewhere to start is to think about more innovation with the format that in essence, has replaced the album – playlists.  

Are playlists now set in stone as 30-50 tracks that are forever subject to optimisation tools and analytics? I very much doubt it. Despite playlists becoming valuable commercial music real estate in their current form (at least on Spotify) it seems to me that there is plenty of scope for further creative development of the format. We are far from ‘peak playlist’. A glance at the latest MIDiA global consumer data for Q4 2018 would back this up, with just 17% of music streamers claiming to listen to playlists regularly. though that number nearly doubles for Spotify subscribers to 31%. Critical mass for Spotify but hardly mass behaviour for the market overall.

That said, playlists do greatly reflect the way consumers interact with music: compiling or selecting, listening and sharing, sometimes adding context along the way. The opportunities these behaviours offer as a vehicle for promoting and monetizing artists are immense and perhaps still largely untapped, especially with regard to contextualization of new talent, as well as the creative resurfacing of songs from within music catalogues too.

A slowdown in playlist innovation

Spotify came to be the platform that really owns the playlist format, and it achieved this through a breakneck period of innovation between July 2015 and September 2016, with curated playlists brands like Rap Caviar and Today’s Top Hits augmented with the personalised “made for you” power triangle of Discovery Weekly, Release Radar and Daily Mix.

But since then, in over two years there has been little of that same calibre, other than somewhat lighter initiatives such as “Your Time Capsule”. More recently, Spotify has begun to fuse some curated playlists together with algorithmic suggestions (‘algotorial’). The strategy to personalise more playlists not only makes it harder for its competitors to copy, but introduces many more song slots for direct or independent artists to fill. The strategy is smart but perhaps lacks imagination.

Playlist Innovation Timeline: There has been a slowdown following 18 months of blistering innovation led by Spotify between July ‘15 and September ‘16.

Playlist-Timeline

Despite this, remarkably none of the major streaming competitors have made moves to innovate in the playlist space either, which seems a shame when streaming services are desperate for differentiation strategies. Instead, Amazon, Apple and Deezer seem to have largely played a game of follow the leader, with similar playlists themes, visual presentation and even ever blander names. In recent days Pandora has launched something called Stories, a mesh of podcast + playlist, which seems not only to lack imagination but to be muddled. Isn’t that just a tweak on radio?

Many artists and labels are concerned about the power that services have with regard to playlisting. While getting tracks on to the current raft of major playlist properties represents a challenge for A&R and marketers the music industry, the more accessible platform-hosted independent curated playlist have been around since Napster.

There are few, if any, limits to the potential reach of these playlists, and their value is determined in part by how creative they are, and in part how shareable they are. The recording industry’s earlier efforts to build playlists or influence playlists have fallen short though lack of imagination, global thinking, context and shareability, but that just means that the industry is free to work more creatively with the wider playlist community. It surprises me that Spotify, which plays host to independent curators still (and hopefully will continue to) doesn’t do more to tend to the community. But this perhaps creates room for others to do so.

Personally, I can see new exciting playlist platform tools like Soundsgood helping global playlist curators reach audiences more easily, providing many more opportunities for artists, labels and creators in the process. Other platforms like Stationhead and Lost are doing interesting things too with curators and tracks, but perhaps could get more traction if they switched focus to playlists?

Bringing back context

I like playlists as much as the next music fan, but like Mark Ronson and many other artists and music fans, I also feel they have commodified music somewhat. One thing I would like to see much more of for playlists is context. Why can’t we have sleeve notes for playlists about how they were put together (let alone by whom and why?). It looks to me like there is an opportunity to bring some context and personality back into playlisting. Mixtapes from the cassette age were lovingly compiled, often with handwritten notes and homemade cover art.

I’ve played with some of these ideas with my own playlist site the Song Sommelier. The Song Sommelier was created as a passion project to bring mixtape and vinyl values back to playlisting. Each playlist is accompanied by bespoke artwork by ‘resident’ artist Mick Clarke, and a reader or ‘digital sleeve notes’ by the playlist curator. For this project we chose to use the Soundsgood player to allow users of any streaming service access the playlists, doing away with those widgets that direct the music fan to either Spotify or Apple. It’s still more than a two horse race after all.

My hope is that with the artwork, sleeve notes and more original themes for the playlists, will make fans and listeners listen to playlists more like they used to with albums – paying more attention and hopefully connecting with the song collection on a deeper level. At least that might make Mark Ronson cheer up a bit (though his new “Club Heartbreak” playlist on Apple Music suggests it might be a while).

Perhaps a key question is how important is context to the music itself? It depends on your point of view, but context may be the thing that divides the passionate music fan from the casual one. Really, who wants to listen to a collection called “Get Home Happy” or “Prime Chill”, whether it’s curated or personalised? Except I guess some people do. Mostly, I would welcome some more creative thinking and innovation that can take playlists to another level.

Why India Matters to Spotify, and Why it May Not Deliver

Warner Music and Spotify have been involved in a rather unseemly and very public spat this week over Spotify’s India launch. I’ll leave for someone else, the discussions of the potential implications of a blanket license for songwriter rights in India for an on-demand streaming service. Suffice to say, the words ‘can of worms’ come to mind. Instead, I am going to focus on why India matters so much to Spotify.

The next one billion, perhaps…

Spotify’s Daniel Ek has made much of addressing the next one billion internet users as part of Spotify’s long-term opportunity. Given the fact that China is effectively off the table for now and that sub-Saharan Africa is probably a generation away from being a major streaming market, India is the key component of that next one billion.

Europe and North America accounted for 69% of Spotify’s subscriber growth in 2018. While this was hugely positive for those regions and delivered high-value subscribers – declining ARPU notwithstanding, growth in those regions will slow down towards the end of this year. Next tier markets – Brazil, Mexico, Germany and ideally, though probably not, Japan – will pick up much of the slack. But to sustain the growth rates its shareholders require, Spotify needs other large markets to start building real momentum by 2020/2021. India and the Middle East represent the best options. However, the Middle East already has a strong incumbent – Anghami – and a potentially resurgent Deezer, newly empowered by its exclusive deal with leading local label Rotana. So, India is effectively the last bet on the table.

India is a very competitive but problematic market

India, however, is a problematic market. It has a host of well-backed incumbents – Jio Music, Saavn, and Tencent-backed Gaana – as well as solid performances from Apple and Google. Yet despite all this robust supply, the market heavily underperforms, registering only 1.7 million subscribers in 2018 with a monthly label ARPU of just $0.74. 1.7 million may sound like a solid enough base, but it represents just 0.1% of the total Indian population. There are two key reasons for such weak uptake to date:

  1. Music plays a different role in India:Bollywood and devotional are two of the most widely listened to music genres, neither of which are mainstays of subscription services, nor streaming music consumption in general.
  2. Income levels are low:the average per capita income is $553 a month, with the luxury of a music subscription far out of reach for most Indians, other than urban elites. Spotify’s $1.80 price point in India may sound cheap, but relative to average income, it is 9.3 times more expensive than $9.99 is in the US. So, Spotify would need to be priced at $0.19 to be the same relative affordability as in the US, which coincidentally is the price for its day pass.

The ARPU challenge

The realistic ambition for Spotify should be to drive five to 10 million subscribers over the next five years or so, primarily pulling from urban elites (essentially a re-run of what has been happening in Latin America). While more credible, this falls way short of denting the ‘next one billion’ opportunity. To unlock the scale opportunity, streaming has to look beyond subscriptions, and also beyond ad supported (India’s 270 million free streamers only generated a monthly ARPU of $0.006 in 2018). The scale opportunity is telco bundles. Reliance Communications’s prioritisation of Jio Music makes it the most likely player to capitalise on this in the mid-term.

Spotify needs to find a similar scale partner and somehow convince label partners to accept an ARPU of say $0.08, which would be roughly in line with US telco bundle ARPU on a Purchasing Power Parity (PPP) basis. This would unlock scale without having to tolerate the catastrophically lower ad-supported ARPU rates. But the odds of that happening anytime soon are miniscule. When, and it is a case of when, not if, that time does come, the scale of adoption could be transformative for the global market. In fact, this is exactly where MIDiA thinks the market is heading. In our just published music forecasts, we predict that by 2026 India will have the fourth largest installed base of music subscribers, anywhere in the world.

What matters most, revenue or scale?

The question is whether Spotify can be a major part of the ‘Indian adoption’. Even if it can, the ARPU will be so small that all the current concerns about Spotify’s falling ARPU will look like a storm in a teacup when compared.

Have no doubt, India can, and perhaps will, become a major player in the global streaming market, even helping reshape global music culture. It can also play a major role in Spotify’s future, but the rules of engagement will need to change to unlock that growth, and in doing so Spotify will be sacrificing ARPU. All of this means, Spotify’s investors and partners need to ask themselves, what do they want Spotify to deliver most: strong revenue growth or strong subscriber growth, because India cannot deliver both.

What the 2018 Success of the Beatles for UMG Tells Us About Where Streaming is Heading

Universal Music Group recorded an impressive €6 billion in revenue in 2018, bolstering a JP Morgan valuation of $50 billion. No doubt, UMG is enjoying a purple patch, riding and driving the wave of recorded music industry growth. But as with an any industry transition, progress is not linear and the past can have a lingering embrace. In UMG’s earnings report lies a small but crucial detail that point to the fact that the music industry’s path ahead may not be quite as straight as it first appears: the continued success of the Beatles.

The Beatles were UMG’s fourth best seller in 2018 

On page 13 of Vivendi’s year-end financial report, the Beatles’ ‘White Album’ is listed as UMG’s fourth best seller in 2018. It finished ahead of frontline artists including XXXTentacion, Migos and Ariane Grande. Above it were Drake’s ubiquitous ‘Scorpion’, Post Malone’s ‘Beerbongs & Bentleys’ and the soundtrack to ‘A Star is Born’. On the one hand this reflects the continued importance of the Beatles as a revenue driver for UMG. The Beatles, along with Abbey Road, were among the ‘crown jewels’ that UMG gained when it acquired EMI in 2012, so it is encouraging for UMG that the Beatles continue to deliver top tier revenue. However, Beatles revenue is not only a very different thing from Drake revenue, it also highlights the earnings divide between physical sales and streaming.

Streaming’s twin promise

The long-term promise of streaming is the combination of:

  1. Delivering larger audiences
  2. Replacing near-term, large volume revenue for a longer-term, annuity-like income model

Item number two happened very quickly; item number one is still in progress, but moving sufficiently enough to ensure many artists are now able to earn meaningful streaming income. However, we are not yet at our streaming destination, which is illustrated by the prominence of the Beatles in UMG’s 2018 sales. A ranking that owes little to streaming.

The Beatles are not a streaming powerhouse

According to the BPI, music from the 1960s accounted for just 3.6% of catalogue streams in the UK, which represents about 2% of all streams. Let’s assume the Beatles account for 40% of those streams – which is probably a generous assumption, this would mean the Beatles represented 0.8% of the $9.6 billion of streaming revenue in 2018, which translates as $79 million, which in turn equates to 2.7% of UMG’s 2018 streaming revenue. A meaningful amount for sure for a single artist, but not that significant in the greater streaming scheme of things. Therefore, the Beatles did not get to be UMG’s fourth biggest seller through streams. Instead, it did so through physical sales.

midia beatles umg

The main release in 2018 was the 50th Anniversary edition of the White Album. This premium physical release includes a $25 edition, right through to a $145 deluxe box set. With such high-unit prices, only small numbers need be sold to generate meaningful revenue.

To illustrate the point, let’s assume UMG collects around $15 of the $25 retail price and $100 of the $145 edition. To generate $7.5 million of label revenue, UMG would need to sell just half a million copies of the $25 edition and only 75,000 of the Boxsets. To generate the same $7.5 million from streaming UMG would need to have 62.5 million people each streaming 15 tracks from the album. $7.5 million is incidentally also roughly the amount a label would earn from selling a million copies of a standard priced album.

Streaming cannot yet match CD-era album revenue metrics

This gets to the heart of the matter of why streaming is creating, alongside a welcome growing body of middle-tier artists, a small handful of megastars. To replicate physical sales success, an artist must have exceptional streaming success. To replicate standout physical success requires as yet ungraspable streaming success.

For example, the number one album in the US in 2000 – NYSNC’s ‘No Strings Attached’ – sold over nine million copies, which would require 600 million people each streaming it all once — roughly 8.5 billion streams — to generate the same income. That is more streams than the entirety of Drake’s 2018 Spotify streams across the entire planet – Drake was Spotify’s most streamed artist. In short, streaming is currently large enough to make record labels grow, but not yet vast enough to create artist-level revenue on the same scale that that the CD peak once did.

Longer-term revenue may, or may not, add up.

The counter argument is that over a number of years the revenue will add up to the equal. But even with that assumption, an album would need to generate around a billion streams a year over eight years to replicate the success of NSYNC’s ‘No Strings Attached’. No easy task when you factor in the dynamics of streaming consumption i.e. playlists replacing albums, new music being pushed over catalogue etc.

None of this is to suggest that streaming is failing, nor that UMG’s revenues are in question. Both are doing well. Instead, it is evidence that we still have much distance to go with streaming before we can start seeing artist-level successes on a par with the peak of the industry. Though of course streaming-level success needs measuring differently than CD-era success, so these comparisons provide context rather than performance targets.

Will there ever be another Beatles’ greatest hits?

One intriguing post-script to all of this is that with download revenue falling by 15% in 2018, and physical by 7%, the days of large-scale album sales are long gone. When this is considered alongside the Beatles’ under-representation on streaming, the elephant in the room is whether UMG would ever risk releasing a Beatles greatest hits album for fear of underwhelming sales numbers damaging the Fab Four’s legacy. The last greatest hits was ‘1’ back in 2000 during the EMI years. Might it just be that UMG bought the Beatles too late ever to release their last ever greatest hits?

Spotify Q4 2018: Solid Growth With a Hint of Profitability But Longer Term Questions

Spotify finished 2018 strongly, overperforming in both subscriber and ad supported MAU additions. This was accompanied by Spotify’s first ever profitable quarter and two major podcast acquisitions early in 2019 hinting at a positive year ahead. However, at the same time premium ARPU continues a long term decline – the price Spotify is paying for maintaining global subscriber market share.

spotify 2018 earnings midia research

Spotify hit just over 96 million subscribers which was an increase of 36% from 71 million in Q4 17. The addition of nine million net new subscribers in Q4 18 was the same amount of subscribers added one year previously. However, while the Q4 17 increase represented 15% growth in Q4 18 the rate was 10%. Relative growth is slowing as the market matures.

Spotify is growing its subscriber base markedly more quickly than it is growing its premium revenue, resulting in declining ARPU. Although subscribers hit 96 million at the end of 2018, premium ARPU declined from €6.20 in 2016 to €4.81 in 2018, a fall of 22%. Over the same period ad supported ARPU followed a mirror opposite trend, growing +22% from €0.96 to €1.17. Spotify routinely explains in its earnings that trials and family plan adoption are driving down ARPU. However, this is not a secular trend but instead a Spotify trend. In retail terms, global music subscriber ARPU actually grew 3.5%. Spotify slightly increased its global subscriber market share in 2018, up to 36.2% from 35.8% in 2017, but it is clearly having to aggressively discount pricing to do so.

Subscriber ARPU continues a downward trend

While ad supported ARPU was up, ad supported revenue grew more slowly in 2018 than 2017 so the increased ARPU is in part a result of users growing more slowly than monetisation. While this is the right balance commercially, Spotify also needs to grow ad revenue more strongly. 

Takeaway: Spotify is maintaining subscriber market share through price discounts while ad ARPU growth owes more to slower ad supported user growth than it does monetisation.

Churn up on an annual basis

Following a peak of 5.8% in Q2 18, Spotify brought quarterly churn rates down, first to 5.6% in Q3 18 and then 5.3% in Q4 18. However, the cumulative impact of churn throughout the year was an annual churn rate of 19.8%, up from 18.1% in 2017. This in part reflects the effectiveness of promotional trials. These trials open the funnel to new subscribers and have strong conversion rates, but because paid trialists are counted in Spotify’s subscriber numbers, any that do not convert become churned subscribers.

Takeaway: Spotify is having to spread its net wider to maintain subscriber growth. 

Profitability has arrived but investment is needed for long term growth

Spotify closed off 2018 in style, adding higher than expected numbers of both subscribers and ad supported users. Also, profitability is on the horizon – Spotify generated a quarterly net operating profit of €94 million in Q4 18 compared to a quarterly loss of €87 one year previously. Spotify is demonstrating that its business can operate profitably even without flicking the switch on new revenue streams, albeit at a modest level. 

Longer term revenue growth will be dependent on a two pronged approach of accelerating subscriber growth in big music markets that are later entrants to streaming – Germany and Japan – while continuing growth in large mid-tier markets like Brazil and Mexico. It also needs to continue its investment in ad infrastructure. Ad revenue is not growing fast enough, nor is Average Advertising Revenue Per User (AARPU), up just $0.15 in Q4 18 compared to Q4 17. This is an increase of just 3% compared to the 26% growth in ad supported MAUs. Spotify understands the importance of building its ad supported business and is investing heavily in ad technology and sales infrastructure. This needs to continue. But it will look to big radio markets  (e.g. the US, Australia and the UK) to drive mid-term growth, not emerging markets as those territories do not have strong enough digital ad markets. So expect AARPU to be hit as free user bases grow in emerging markets.

Takeaway: All in all, a solid quarter for Spotify but with enough softening metrics to suggest that 2019 growth will require more effort than in 2018.

NOTE: these findings form a small portion of MIDiA Spotify Q4 Earnings Report which will be available to MIDiA subscribers next week