• Follow Curiously Persistent on WordPress.com
  • About the blog

    This is the personal blog of Simon Kendrick and covers my interests in media, technology and popular culture. All opinions expressed are my own and may not be representative of past or present employers
  • Subscribe

  • Meta

Will Spotify revolutionise the music industry?

spotify-logo1This may sound like a sponsored post, but I can reassure that it isn’t. I unapologetically love Spotify – it is the best new service I’ve used for quite some time. It has completely changed my listening habits. And while I accept I am not the average consumer, I do wonder whether it can represent a major shift in the composition of the music industry.

What it is

For the uninitiated, Spotify is an incredibly simple to use online music service. Based around an interface similar to that of iTunes, it allows you to either search through its enormous back catalogue (all the major labels have signed up) to pick out a specific artist or album, or use the “radio” feature where you can specify a genre or time period (though this doesn’t work so well since Spotify takes the date of a track as the most recent reissue). Arguably the greatest feature is the playlist function – these can not only be created and shared, but open playlists allow anyone to contribute.

spotify-screengrab

With fairly low barriers to entry and high levels of satisfaction among users, it has recently passed 1m users. Still fairly small, but not bad considering it is only available in a few territories (due to licencing issues, which could ultimately puncture plans for world domination).

The API has also recently been opened, so we will be soon be seeing new applications and functions based around the software in a similar way that Twitter has. This also opens up the possibility of Spotify being made available on any web-enabled device from games consoles to car radios to refrigerators.

The business model is a variation on freemium. The majority (including me) use it for free in exchange for around 3 ads an hour. But there is also the option to upgrade to a day or monthly pass, which is ad-free and offers some additional functionality. Furthermore, they recently announced a partnership with 7digital to offer downloads.

The competition

Can, and will, Spotify grow further? I think so, though as a disruptive service it faces some major competition.

Music consumption is largely a mature market. Even with game-changing services it is unlikely that the overall size of the market will increase by any great extent. Listening on Spotify will largely be replacement activity. There are several things it could replace:

Can it replace ownership? CD sales are falling, and being replaced by digital versions. Is there a difference between owning a digital copy and being able to access it through the “cloud”? I would say yes. Spotify giveth and Spotify can taketh away. Despite issues over DRM or hard drive corruptions, there is still something to be said for “owning” something. However, using Spotify as a “try before you buy” will arguably reduce album sales to only the essentials. A streamlined, but high quality, collection is good for the consumer, but not for the retailers.

In the “try before you buy” category, Spotify competes with other online services ranging from Youtube to Hype Machine to Myspace to Last.fm. At the moment, Spotify wins on depth of content, but as yet doesn’t offer the breadth of functionality of these sites. Youtube et al offer a means of listening to music, but this is then wrapped inside the social elements. Spotify can’t yet compete on this front.

Ultimately, I see the biggest loser being radio. I haven’t been an avid radio listener for several years but do appreciate the strengths of the platform – locality, portability, editorial, serendipitous discovery, and original content production among others. These are big strengths but they aren’t insurmountable – all can be replicated. Radio is inherently a push medium – this appeals to some but equally others prefer a pull. If Spotify can evolve, it can offer a very real alternative.

The business model

Unfortunately, the biggest barrier is monetisation. The music industry is accustomed to high revenues through monopolised distribution. This is being eroded but the ability to make money through music isn’t changing commensurately, with the PRS running both Pandora and Youtube (for the time being) out of town. (for the record, I completely agree that artists should be compensated but something, somewhere needs to give).

This coupled with the bizarre truism that growing too fast actually hurts the business makes it difficult, but not impossible, to achieve profitability. This is because more users listening to more music means more bandwidth charges. Credit Suisse estimate Youtube’s losses are at $470m on costs of $711m. Spotify isn’t on a scale anywhere near this, but it is a concern.

However, there are many potential sources of revenue. Many people smarter than me will be working on realising them, but off the top of my head:

  • The three ads per hour frequency is evidently an artificial limit and this could change, though at the moment there seems to be more of an issue filling their inventory rather than expanding it
  • People need to sign in to use Spotify. Sign-ins mean targeted advertising, which will fetch a premium
  • Similarly, adverts can be targeted by particular artists, genres or even songs
  • Display advertising’s effectiveness will be limited as it is less interruptive, but specific call to actions (e.g. accessing an artist microsite, or competition) could prove effective
  • I’ve already seen a sponsored playlist for a film, and this is evidently an area that will be capitalised on
  • The download element will have a better chance of success if it can offer things iTunes et al are unable to
  • Could there ever be paid placements of songs within the stream? Well, payola is a word that exists…
  • Additional premium services could persuade people to upgrade, though I’m sceptical of this.  Firstly, what would be the “killer app” that would persuade people to subscribe to a streaming music service in the way that the Premier League and Hollywood films lay the seeds for people to accept paying for TV in the UK? For me, it would either be unique archive content – not easy – or portability. But, secondly, it is inevitable that they would eventually filter down to the free version lest competitors jump in.

I see Spotify’s success being reliant on a sustainable ad funded model rather than subscriptions. As a service, it is still new and there are several directions it could go in.

The big question is who do they want to directly compete with. Is it traditional radio stations, iTunes, Last.fm or a combination thereof?

Future extensions

In pure fantasy land, ways that Spotify could be made even better include:

  • Portability – most likely through a smartphone. Of course Apple are unlikely to let an iTunes killer onto their phones, but then Google, Palm et al finally have something to offer that the iPhone doesn’t
  • Original content – not necessarily DJs or Radio 4 style comedies, but being able to opt in to news, weather or even traffic updates if there is an in-car version (which could be both sponsored and localised through an IP address) would take away one of radio’s major USPs
  • Recommendations – there is already a basic recommendations page, but this could be hugely expanded either through an algorithm or some form of UGC tagging
  • Statistics – already available in terms of top artists/albums but there are plenty of data geeks out there who appreciate the option to catalogue their habits (though the last.fm plug-in does do this already)
  • Sharing – the playlists are a great idea but being able to “friend” those that have similar music tastes takes it to the next level. Going down the social route, with UGC comments, reviews etc, would create “stickiness”
  • A UGC section allowing unsigned bands to upload material and try to make a name for themselves
  • Integration with user libraries – particularly if the download store takes off. Combining the streaming service with a straightforward playback service allowing you to switch between your music library and the Spotify catalogue would give a one-stop shop for listening
  • Integration with other services – with an open API, Spotify could theoretically be available on games consoles. The next step is integrating it into games e.g. as an alternative to the radio stations within Grand Theft Auto

Summary

This post is probably better suited on a fansite than this blog. But I really, really like Spotify. At the moment I am listening to Edvard Grieg. Previously, I was listening to Scout Niblett, Less than Jake, Nina Simone and the soundtrack to Dexter. Aside from Scout Niblett, I wouldn’t normally buy any of these, yet I am able to broaden my music horizons and investigate new artists and even new genres.

My media consumption habits aren’t anywhere near an approximation of a mainstream average, but I still think radio stations, other websites and even retailers should be concerned. The labels should embrace it – they have already given consent but it represents a tremendous opportunity to both push new content and remind listeners of the extensive back catalogue available in their publishing archives.

Spotify is still a new service, and for all I know the site may remain true to its roots. But it has the potential to be so much more.

For those yet to try it, I wholeheartedly recommend it.

sk

Reblog this post [with Zemanta]

Measuring the success of free music

As an addendum to my prior post on free music, Kevin Kelly has posited a “true fan” model. He argues that having a core following of around 1,000 people per artist would bring in enough revenues to sustain a career. Nine Inch Nails appear to have managed this – their run of 2,500 deluxe ($300!) editions of their latest release sold out straight away, and John Otway has sustained a career this way. Of course, this requires either your fans to grow up with you, or a constant stream of new fans to replace disillusioned ones. NIN’s early music was quite angsty – will their fanbase remain loyal as they enter maturity? The evidence so far suggests they will.

I’m interested in how one can make this a robust business model. In TV land, there is increasing talk about moving from eyeballs to engagement. It is no longer enough for someone to have a TV programme switched on – they need to be attentive, to be interested, to interact. But measuring this is tricky. With music, as the methods of distribution increase, it becomes more difficult to know exactly who owns, or who listens to your music. What proportion of tracks owned by music lovers came into possession through legal (and measurable) methods? And how does ownership intersect with passive listening – through TV, radio, Myspace and so on.

Could musical success be measured by engagement? If the Seinfeld Curve is borne out, the record labels (or their successors) will need greater certainty in predicting ticket sales, not to mention advertising on artist websites and other revenue streams. A minimum number of units sold will be meaningless. Some artists (I can’t remember their names) are now getting people to pay up-front for tours or CDs. If enough people sign up, it happens. If not, the transaction won’t take place. This may work in the odd case but is too rigid to be sustainable for the majority of the market. A model of engagement therefore needs to be found.

A lot of data is already available on music consumption and engagement, but it is proprietary. Could iTunes or Last.fm become an industry currency? If they asked all of their users whether they would be willing to share their data with the music industry, a lot of people would refuse. Understandably so, given the amount of pirated and leaked music available. But if incentives were offered, I believe a a reasonable number would. Interested users could fill in an additional form for demographic information, and this can then be calibrated to the wider audience for greater accuracy. If this became an industry standard, there would be no competitive advantage. The more parties that sign up, the cheaper subscriptions could be and the lower the barrier to entry

Some of the possibilities of data analysis this could offer include:

  • Knowing number of plays, not just number of units shifted
  • Finding out the top rated tracks for each artist
  • Seeing how quickly new tracks are forgotten about and never played again
  • Profiling the most avid fans
  • Segmenting these fans by geography to plan tours
  • Getting similar artists to join the tour
  • Measuring the speed at which new music travels globally
  • Tracking and predicting popularity of different genres
  • Last.fm event attendance can be correlated to music libraries
  • Using the messageboards to gather information on buzz
  • Surveys can be used to gather opinions to supplement the raw data

I realise this is just a pie-in-the-sky idea and that there are many barriers to this actually being implemented. But if music does become the loss leader to the experience’s premium product, then there is no point in restricting the distribution to official methods. If the music can be democratized, can the information?

sk