In the simplest terms, the five models below represent different approaches to customer acquisition and marketing. (The royalties paid, and their role in these models' success, is mostly excluded from this discussion.) In survey after survey, consumers profess far more interest in subscription services, and their willingness to pay for them, than is illustrated by companies' actual subscriber numbers. A recent Nielsen survey found 78 percent of music fans are either somewhat or very likely to pay for a music streaming service in the next 6 months. Numbers like that are encouraging, but the odds are close to zero so many people will actually become subscribers. The challenge is getting consumers to try a service and learn about its features. Just getting to the awareness stage is a difficult and expensive task. Most of the business models outlined below are built to moderate these problems.
Optimists might say all five models can comfortably compete in a growing market. Pessimists could argue that the subscription market requires such massive scale to succeed that only a handful of services will exist in the type of winner-take-most market, such as what we see with Netflix and paid video streaming. The likely outcome is in the middle. The models with the most advantages and industry support will help advance some services and send others to the digital music ash heap.
The Pure Freemium Model
Free access often leads to paid access. It's called a freemium model. Businesses work under the presumption not all users will turn into paying customers. The trick is getting some fraction of free users to upgrade for a paid service that offers more features. This model is often seen in cloud-based services that scale well and have a low cost of adding an incremental user.
Spotify is the standard example of the freemium model. It offers unlimited, ad-supported free streaming that generates about 9 percent of its annual revenue. That free service is an entry point for its subscription service, which generates the other 91 percent of revenue. It's impossible to overlook the controversy surrounding Spotify. Many rights owners and creators oppose freemium business models (this and the one below) for the amount each ad-supported stream generates. According to David Lowery's Trichordist blog, Spotify's average per-stream royalty is 0.521 cents, well below paid-only services like Rhapsody (1.12 cents), Xbox Music 0.32 cents) and Google Play (0.46 cents).
YouTube and SoundCloud are also in this category. YouTube has been a free video streaming service since launching in 2005. Late last year, YouTube added a premium layer, called YouTube Red, that gives subscribers an ad-free experience and exclusive content. SoundCloud is another free service that sells advertisements against some of its streams. Compared to Spotify, its catalog is more expansive in independent and derivative works (remixes) but smaller in record label catalog. It intends to offer some sort of paid service to complement the free service.
The pure freemium model attracts a large enough number of users to have created an opposition which threatens its very existence. Some record labels want to scale back unlimited, on-demand listening in favor of trial periods before requiring listeners to pay for access. The creative community, most notably Taylor Swift and, more quietly and recently, Adele and Rihanna, have railed against the royalties from the freemium model's ad-supported service. Although listeners could easily switch to YouTube if Spotify is forced to place limits on its free access, there is a good chance Spotify will be forced to alter its free offering to appease the parties that supply its music. If this happens, it could limit the amount of free listening or allow artists not to make their music available to free listeners (this was reportedly considered for Coldplay's album Adventure of a Lifetime).
The Limited Freemium Model
The limited freemium model is the less controversial version of the freemium model. The key difference is their approaches to upselling listeners to a paid, on-demand service. Slacker already employs this model. Pandora will do so later this year.
For Pandora, the freemium model is limited by the restrictions placed on the listener, operating a radio service that is non-interactive in nature, not on-demand. Listeners are limited in their selection of songs, ability to skip songs and the frequency an artist can be heard. Section 114 of the Copyright Act establishes the limitations (no on-demand song selection, no pre-published playlist, limitations on the frequency of songs by the same artist or from the same album) placed on non-interactive services that wish to utilize the statutory license. A service that opts for direct licensing over statutory licensing, like Slacker, is required by record labels to abide by the same or similar rules. Outside of the United States, where the statutory license does not apply, direct licenses are required for both the ad-free radio and on-demand services.
Pandora has a unique advantage here: its starting point is its massively popular Internet radio service. It takes a subscription service years to build brand awareness and entice consumers to either a free service or free trial of a paid service. Pandora is starting with an established user base of nearly 80 million monthly listeners and an unknown number of registered users. Brand awareness is not a problem. The task of signing up users for the free service has been completed. In fact, Pandora has a big lead in the United States. According to Triton Digital's December numbers, Pandora had almost 50 percent more session starts than Spotify and nearly seven times the session starts as the next largest non-interactive radio service, iHeartRadio.
Given the composition and size of Pandora's user base -- a large number of low-value monthly listeners that pay little or nothing for music -- a mass migration from the free to paid service is unlikely. Most free radio listeners will be content remaining free radio listeners. (As a point of reference, about 4 million of Pandora's 80 million active users pay the ad-free Pandora One service.) Even so, the paid on-demand service will allow Pandora to retain some users, capture the on-demand spending of existing Pandora listeners and attract new customers.
It's a relatively safe approach, too. Because Pandora already specializes in monetizing free listeners, it won't be dragged down by content costs. And because its free service is based on the statutory license, at least in the United States, record labels can't threaten Pandora's model. (Publishers have more leverage. Some of them negotiated a rate increase late last year.)
There are signs the limited freemium model, in various shapes, will win out over the unlimited freemium model. The major labels want to scale back Spotify's unlimited free listening. The creative community, mostly notably Taylor Swift, also opposes the freemium model and its offering of free, unlimited music. The Pandora approach is like a compromise between limited and unlimited models. The limited model provides unlimited listening but without the features of paid models. As a result, it may be the more palatable approach to acquiring listeners.
The Vertical Paid Model
A number of large technology companies have standalone subscription services that complement their products. These offerings are integrated vertically, one stacked upon another, as to own multiple steps in the value chain that connects products with consumers. In one example, the value chain starts with the manufacturing of computers and portable devices, often purchased directly through the company, continues to an app downloaded at that company's app store, then to its music service. The steps are connected with branding and marketing efforts.
The above example describes Apple's approach to its music subscription service, Apple Music. Apple has a unique set of strengths. Its footprint grew to 1 billion active devices in the quarter ended Dec. 31. It has distinct advantages in marketing the service to consumers worldwide. It is the world's largest technology company (and most valuable company in the world) with vast resources and the ability to generate widespread media attention for even the smallest development.
Apple Music and Google Play Music are small tools in the very large toolkits of the world's first- and fourth-largest technology companies, according to Forbes. Some messaging apps are also using music to complement their core businesses. LINE offers a music subscription service in Thailand and Japan. And earlier this month, Kakao acquired South Korea’s largest music streaming service, MelOn.
The Standalone Paid Model
Some subscription services are islands unto themselves. They do not complement the products or services of a large corporate parent, nor do they use unlimited access to attract listeners. They ask only that people pay a subscription fee. Netflix has been able to do it in video streaming, but music services have had less success.
The term "standalone" refers only to ownership. Companies that employ the standalone paid model tend to partner with telco companies to reach consumers. These partnerships mimic the relationships in the vertical model. While Apple, for example, can leverage its products and marketing for the benefit of Apple Music, a standalone service like Deezer reaches consumers through its relationships with Orange and other telcos.
Until Apple Music, consumers have been slow to pay for subscription services. But relatively few of them are paying for the subscription services of standalone companies that aren't owned by a large technology or e-commerce company. Deezer has nearly 4 million revenue-generating subscribers. Rhapsody announced in December it was nearing 3.5 million and Tidal has surpassed 1 million. Based on its bankruptcy court documents, Rdio probably had fewer than 200,000 before it closed down. These numbers are insufficient to run profitably, and lag far behind Spotify and Apple.
The trial period is a key tool of the paid model. It creates familiarity and, if all goes well, a desire in the consumer to pay for the service. Duration of trial periods vary. Apple Music comes with a three-month trial period. One week was common up to a few years ago. Now, one month is a common length for a free trial. Longer trials can be found with telecommunication companies and other partnerships.
A subscription service that truly stands alone is destined to fail. As was the case with Rdio, one can also fail even with partnerships.
The Soft Paid Model
Music need not be the focal point in the soft paid model. It can be a reward, a feature or an enticement to enroll in a larger program. Even though it feels free, the price is hidden from the customer, the rights have been licensed and rights owners are paid. Rather than requiring hard cash, the soft paid model is available whether or not the consumer wants it.
The best example is Amazon Prime Music. In ten years, Amazon Prime has evolved from a free shipping program to a feature-filled membership plan that also includes video streaming -- some of it exclusive content -- e-books, unlimited photo storage and music. Amazon uses music and other Prime features to encourage Amazon customers to stay longer and spend more. Estimates for the number of Amazon Prime members in the United States range from 44 million to 54 million and 60 million to 80 million worldwide. Amazon has been rumored to be developing a full-feature subscription service, a la Spotify.
Another example of the soft paid model was Muve Music. It started as an on-demand service created by prepaid mobile carrier Cricket Wireless. Muve Music was originally a soft-bundle that was sold as a $5 add-on to monthly fees. It later transitioned into a hard bundle that was added free of charge to a number of mobile plans. Deezer acquired Muve in 2014. Now the Deezer service is soft-bundled into Cricket Wireless plans for $6 per month.
A single company can dominate its segment of digital entertainment. iTunes captures roughly 70 percent of the world's music download revenue. Netflix's share of North American download traffic (from fixed access sources) of 36 percent is more than double the next source, YouTube, and 18.5 times greater than Amazon Video. Pandora has 58 percent of Internet radio listening and 9.5 percent of all radio listening -- terrestrial and Internet -- in the United States. The music subscription market doesn't yet have a runaway leader but is still top-heavy. Billboard estimates Spotify has roughly 50 percent of all subscriptions globally, even after Apple's strong entrance.
The business model matters here. The difference between the first and, say, third places will be hundreds of millions of dollars in the next few years alone. Third place should afford a comfortable business, but the tenth-ranked service won't be as fortunate. The larger industry may not care who survives as long as there are survivors. But for any single service, especially a standalone, the model that drives customer acquisition could be the difference between life and death. For subscribers, which services survive will determine how they will experience music.