Author's Note: This essay was published almost a year ago. The goal was to predict what the team at project Daisy, now called Beats Music, would need achieve in order to crack the subscription music model and reach the mainstream music listener. Now that Beats Music has launched, it looks like I got a few things right...
I. Invisible Funnel
Industry pundits and executives looked at the file sharing and torrent downloading craze, and saw millions of people who desired unlimited access to music but lacked a legal service to satisfy their needs. Stockpiles of MP3s and bloated iPods evidenced a potential market for an entirely new model where every song on iTunes could be streamed from the cloud for a monthly fee. This fantasy of a “celestial jukebox” has always appealed deeply to fanatic listeners, who dreamed of Santa giving them the entire inventory of Amoeba Music for Christmas. But what about the mainstream market — the millions of casual listeners? Can this jukebox attract their interest?
To date, more than a decade into the digital revolution, no subscription music service has broken through to mass appeal. Spotify, a media darling and royalty scapegoat, boasts five million paid subscribers around the world, one million of which reside in the U.S. Rhapsody, the longest-running service, claims to stand above the one million mark, while newer entry Rdio, which has never shared numbers, is suspected to have far fewer subscribers. Meanwhile, Muve Music, a music plan bundled with Cricket Wireless phones, recently announced that it has surpassed 1.1 million subscribers since its early 2011 launch — more than any rival company in the U.S.
Beats Electronics, the creators of the high-end headphone brand, sees opportunity in the lack of any real subscription-model success to date. Since acquiring the fledging music startup MOG in 2012, Beats has hired Ian Rogers from Topspin Media and positioned him at the helm of a new initiative, code-named “Daisy.” In tandem with this move, the company made a substantial investment in Topspin and named it a strategic partner that will supply photos, videos, and products from artists to the revamped MOG service. Both companies said that this collaboration stems from a “shared belief” that music services should “do a better job” at connecting fans to artists.
According to Bob Moczydlowsky, SVP of product and marketing at Topspin, music listeners should be able to find out when their favorite artist is performing nearby, be able to view an artist’s Twitter feed, and see what products they have for sale, all while playing songs in a music service. Essentially, listeners should be able to dive as deeply as they wish into a newly discovered artist without having to leave to a site or start a Google search. Currently, listeners cannot navigate Taylor Swift’s site or Facebook page from Pandora or Spotify. Such services are, in the words of Moczydlowsky, “a walled garden” where the apps are a mere “utility that delivers the sound.”
For fanatic listeners, discovering a new artist opens a rabbit hole that leads to others. But before fanatics take that journey, they often want to learn everything they can about their initial find. With such listeners willing to engage, the fear is that current services do not give them enough to explore, which can hinder the discovery experience, as well as the potential for them to be converted into consumers of an artist’s products. Subscription services may have gotten better at making the actual music available — as catalogs continue to grow — but spreading the word to listeners about upcoming tours and releases remains an uphill battle.
The industry has long urged Spotify and Pandora to help artists to bridge this gap, and in recent months, the situation has improved. Pandora has revamped its web and mobile apps, making it easy for users to find lyrics, a short biography, and an overview of discography. Spotify, of course, allows users to play all of an artist’s songs without limits, but that is the main benefit to the otherwise limited offering. It has also enabled developers to create apps centered on artists and record labels for the platform, which is a huge step forward.
Rogers and Trent Reznor, the industrial rock icon and newly appointed executive at Daisy, have a much bigger vision. They will attempt to connect the artist layer to the music service. The thinking is that by integrating ticketing and merchandise sales directly into Daisy, listeners will move through the marketing funnel more readily and convert at a higher rate. With credit cards already on file, such sales would be friction-free — like Amazon or iTunes checkout — and lucrative for everyone involved. Such an arrangement would align Daisy’s business model more closely with artists and make it less dependent on advertising and subscriptions for revenue.
“The challenge of the funnel — be it the traditional sales funnel studied in business school, or one specific to digital music — is that moving people from each phase gets harder and harder based on transactional friction,” says Jason Feinberg, VP of Digital Strategy at Epitaph Records, whose roster includes punk legends like Bad Religion, Social Distortion, and NOFX. “Fans have to be reached, made aware, made interested, given relevant options, and be able to easily and conveniently conduct a transaction.” Combining all these elements into a music service, he says, could lead to “a noticeable increase in conversion.”
How much money is on the table? Dave Kusek, a business consultant and former CEO of Berkleemusic.com, says there is “a complete disconnect between where most music is discovered today, and the $2.2 billion in annual merch revenue. The vast majority of merch is sold at the venerable merch table at any given concert.” If merch distribution were aligned with the direction the online music market is heading, Kusek believes that it “would serve artists and merch companies extremely well, and potentially unlock a flood of new revenue.” Added on top of this fair sum is the $2.34 billion in tickets sold to the Top 100 U.S. tours in 2011, according to Pollstar, a trade magazine.
More importantly, Daisy could enable artists to take ownership of the fan relationship through every stage of the funnel, or at the very least, see inside the funnel and reach listeners for a small price. As Rogers has lamented in the past, the fatal mistake many artists make is they send fans to iTunes to buy music or Facebook to friend them, and lose the opportunity to capture their email address for future contact. If artists do not collect and own this valuable dataset, they forfeit the ability to grow and monetize their audiences. Access to contact info and listening statistics in the Topspin dashboard would allow artists to more effectively target and market to Daisy users.
II. Trusted Sources
The opportunity for Daisy is to “create a trusted brand” in music, said Rogers in a conference call shortly after a flurry of news reports that he was taking the Daisy job. He said part of that trust is listeners knowing they are going to receive high-quality, human-curated recommendations and not simply an infinite search box. Meanwhile, Reznor told the New Yorker in an interview that rather than relying solely on algorithms to power the music Daisy suggests to listeners — if you like Taylor Swift, then you might like Sugarland — a network of music experts will point them down less traveled paths. So instead of the obvious fare, a country music aficionado may send listeners to Reba McEntire or Dixie Chicks.
Of course, the notion that music services need programming — as Beats Electronics CEO Jimmy Iovine put it — is not necessarily new. Rhapsody and eMusic employ editorial teams to write blog posts about the trending and popular music, and Slacker and Songza rely on professionals to seed their playlists with the freshest and hottest songs. So too, Rdio has developed social discovery features that enable users to follow tastemakers and receive updates about music they listen to or add to their collections. In the coming months, Spotify, which offers numerous curated playlists through third-party apps, is expected to roll out a similar offering.
Despite the number of programed services already available, the right formula — the “killer app” of discovery curation — remains elusive. The problem is that listeners who like the notion of “millions of songs in your pocket” tend to be music fanatics with insatiable appetites and deep knowledge. They prefer to dig through record bins and program playlists themselves. They already have “trusted sources” like KCRW and Hype Machine who assist in exposing them to new artists. No one in the know about the latest music and apps, to put it bluntly, is missing out or seeking a different solution. Spotify and Rdio work just fine for fanatics.
But Rogers and Reznor want to “take subscription music to the mainstream, and really get it to scale.” To accomplish this feat, Daisy’s staff needs to program the songs that casual listeners care most about hearing: the freshest, hottest songs by top-ranked artists. Such advice may sound rudimentary, but Daisy must develop a Now That’s What I Call Music sensibility. The great irony of Spotify is that the songs that get played on Top 40 radio are exactly what casual listeners play there. Yet curation and discovery apps like Shuffler.fm and We Are Hunted cater to fanatic listeners and mainly focus on surfacing songs that are trending in the music-geek blogosphere.
Why is this? Startup founders and app developers often succumb to a “fanatic fallacy” that leads them to believe casual listeners can be re-educated to like “good” music through the labor of sifting and searching for songs. But they rarely convince anyone put in such effort. So they resort to taking the work out of music discovery, often by hand-curating the most beloved songs on blogs or by aggregating them through various algorithms. The hope is that if make music discovery easier, casual listeners will teach themselves to like “good” music. What founders and developers come to find, though, is that casual listenersjust don’t care enough to make that investment.
“I believe that as an industry we put too much faith in music discovery,” says Feinberg. “I have no doubt that there are many fans that do want to discover new music, but I find far more ‘average users’ — i.e. people not in our industry — that simply want to pick something they love and hit play.” Listeners can tune into genre radio stations on Spotify and Pandora, but these are often pigeonholes that range from vast tunnels to tiny cages, either too generalized or too specific. Likewise, there is no way to fine-tune a station beyond rating songs with a “thumbs up” or “thumbs down,” which can lead to listeners protecting the songs they already love (thumbs up), while dismissing the unfamiliar before really giving it a chance (thumbs down).
Many broadcast radio station lessons have not been applied to subscription services. 1) People want to listen to popular music and hear a few old favorites and new hits along the way; 2) The songs they discover should be from the best album releases by their favorite artists; and 3) The songs that listeners give “thumbs up” to on a station are likely the same songs that they want to hear in similar experiences. The best experience for casual listeners who love Nicki Minaj may be to recommend that they listen to Melanie Fiona and to seed her single “4AM” into a station, and if listeners favorite that song, alert them the next time a new single debuts.
Raw discovery — say, for example, listening to Hype Machine or We Are Hunted — is a bad experience for casual listeners. The songs are unfamiliar and eccentric. To make music discovery more accessible, it should be directly integrated into Daisy. A listener’s library should be the “musical brain” that powers the experience, shuffling in old favorites and next hits. It should be the central location where the songs a listener has collected from the physical and digital worlds are, and where disparate online identities are seamlessly unified. It should be a map that guides them — enhancing activities and brightening moods — without needing to be “seeded” with artists or songs to grow.
“In order to succeed, [Daisy] would need to be designed largely for casual listeners, because fanatics are by their very nature the rarer creature,” says Aaron Tap, a musician who is best known for playing guitar with Matt Nathanson and Paula Kelley. “The big question hanging over every music service is what, really, does the average listener want?” The answer, according to Rogers, is “curation by trusted sources,” which he calls “the next phase of internet distribution.” On this point, Tap agrees, adding that there is “not much personality” in current services, and a well-curated one, along the lines of Jack FM or KCRW, “could be all things to all people.”
This is why Rogers and Reznor cannot just strive to make Daisy a “better” Spotify, with a simpler design and stronger curation. The truth is that Rdio has already created a better Spotify, and Slacker Radio has always been a better Pandora. iHeartRadio sits somewhere in the middle, with a combination of broadcast radio and custom stations, and Songza and 8tracks probably tie for having the best curated playlists. YouTube and Vevo, the largest services, display music videos, which continue to be the most popular media among younger listeners. Not one, though, does a great job at curating and surfacing old favorites and new hits, i.e the songs that casual listeners want to hear.
III. Pieces Together
Billboard writer Glenn Peoples says that subscription services have not reached a wider market because they are not “built, packaged, and priced” for mainstream listeners. The leading services — Spotify, Rhapsody, and Rdio — are for fanatic listeners who will “invest a great deal of time and money in the product.” Peoples argues further that such services “need to become easier to use and a better value for the price.” Aping iTunes may have been the best way for rival companies to translate the idea of unlimited music for their listeners, the Iovine pitch goes, but Daisy will take a different approach, possibly bundling the service with mobile devices and pricing the plan very low.
During a Twitter exchange, professor and researcher David Touve suggested that the price of music services should “match the nature of demand” — $5 a month rather than the industry standard of $10 — because casual listeners only spend less than $60 a year on music. Rhapsody recently introduced a new plan with MetroPCS that dropped rates to $5 a month, proving that such a price is possible. But whether it can help Rhapsody convert and retain subscribers is the big question. The greater problem is that making subscription services cheaper to buy and easier to use will not magically enthuse apathetic casual listeners.
“The idea of being able to hear anything they want, whenever they want, appeals to the mass market — but it’s not actually what most want in practice,” says Jason Herskowitz, who contributes to Tomahawk. “I think what they want is a few giant play buttons that just play them stuff they love . . . with the majority of it being stuff they already know they love with a few prospects of new love sprinkled in between.” Entrepreneur Bruce Warila summed up this concept perfectly, saying casual listeners want a “fader” that calibrates “more to less of the music I already love and less to more of new tracks you think I love.”
The music industry has bought into this idea that casual listeners are fanatic listeners with less time. If we just make it easier for them to harvest the fruits of fanatic labor they will love the same music. In effect, the “programmer” needed for music services is a fanatic listener that takes them on a guided tour of their rabbit holes. Casual listeners, in this context, are viewed as lazier or older listeners (often both). With more ambition or time, they too would research and discover new music, but the effort is too great. Rather than accept that casual listeners are different listeners, we have wrongly labeled them as “lesser” listeners.
But casual listeners never shopped at record stores and sought advice from the clerks. They walked into big box retailers like Wal-Mart or Best Buy — where the staff is mostly clueless — and grabbed the album with the single they liked off an end-cap. They do not spend their free time researching on music blogs and programming playlists. They turn on KIIS FM in their car and Ryan Seacrest introduces the next Justin Bieber or Rihanna single. They have seen Nickelback perform live — twice — and eagerly captured Instagram photos of singer Chad Kroeger dousing the entire front section with Bud Light. The “soundtrack” to their life is “Bartender Song” by Rehab.
To be clear, even casual listeners are “fanatic” about some artist or band. They might discover One Direction and grow so passionate that they read the latest gossip, print out and hang up pictures, and share lyrics on Facebook and Twitter. They might even become so committed that they lash out against “enemies” in online feuds. But it is their orientation toward culture more broadly that is different. They will not also become passionate about activities like discovering songs on blogs or listening to records. A fanatic of this sort is something that you are, not someone you become. Fanaticism toward music has never been the destiny that pundits have made it out to be.
In a presentation in 2006, Rogers accused iTunes of being “a spreadsheet that plays music” — adding in a recent blog post that subscription services have yet to find “a way to surround music with context.” The crux of his claim is that listening to music used to consist of physical media — the object that produced brilliant sound, and the ever-important artwork and album booklet that, taken together, formed a cohesive experience. When fanatic listeners bought an album and played the songs, they pored over the photos and lyrics inside. To see the faces of their heroes and read their words (maybe for the first time) gave the music additional meaning.
Rogers (of course) is not the only voice in this choir. Speaking of a recent re-design, Rdio VP of product Malthe Sigurdsson said the company’s new web and desktop apps transform “the boring, spreadsheet-like way of consuming digital music into something visual and dynamic.” The word choice here is very specific andrevealing. It is tempting to argue that listening to music should be emotional and engaging, but that is also a fanatic fallacy. Casual listeners often want music to be anything but those things; they are way too busy working and studying. The problem with Turntable.fm, after all, is that fanatic listeners love the experience, but stop getting anything done and leave.
This is not to say that improvement to existing music services is not needed, or that it should not be attempted. Rogers and Renzor have good reason to deem the state of online music unacceptable and to challenge the status quo (they both share a well-documented history of doing the latter — sometimes even together). The future they hoped for and the present they occupy are vastly different; a music utopia that is neither lost or found. The web did not create a middle-class of musicians and “1,000 True Fans” to support each of them; it gave us a relatively successful handful of artists who did all right, but many more who did not, and an upper class that largely continues to control tours and charts.
After years of fighting with record labels and downloading millions of songs, music listeners have been given the “celestial jukebox” at least some of them always fantasized about. They can pull out their computer and have unlimited access to music on Spotify or Rdio, for free. Hell, Amoeba Music has even made much of its brick-and-mortar store available for sale online, too. Everything is amazing and most listeners are either blissfully ignorant or very happy. Mainstream or not, the “promised land” already exists, and it would seem worth asking, as actor Jack Nicholson once did, “What if this is as good as it gets?”
Perhaps we’ll find out later this year, after Daisy sprouts.