Wednesday, 17 November 2010

Music Futures #2

Last night was the second in a series of 4 seminars hosted by Generator exploring the future of the music industry.  The session featured a number of established and relative new (but influential) players in the digital music market place.

Handmade Music 8/23/07 with Etsy Labs, CDM, and Make

The guest panelists include Dave Haynes (Soundcloud), Atan Burrows (mFlow), Ian Greaves (Napster) and Colin Rice (We7).  The general theme of the session was similar to the previous night: digital downloads versus streaming services, looking to address the opportunities presented by services driven by retail and those that work on an access-basis (and whether or not the two models are mutually exclusive).

In a slight departure from my last post in this series, I've attempted to precis the discussion that took place last night in a rather linear structure, adding a few observations and comments where appropriate.

Killer/filler tracks?

The first issue to be raised revolved around the decline of the album sale and the emergence of a la carte download services which permit music consumers to cherry pick individual tracks (eg iTunes).  The panel was asked to consider whether or not this has led to a position in which people now have more control over the quality of content they access rather than paying for an artist's entire catalogue (underpinning this assumption is the notion that albums contain 'filler' tracks that casual listeners may not want or need).

Atan (mFlow) responded by suggesting that shopping for music online today has become boring and sterile. The shopping experience lacks innovation - it's like shopping for groceries. Indeed, many supermarkets (like Tesco and Asda) have taken this logic and applied it to CD sales.  Shopping for music needs to be more integrated into the social life world of music fans - something mFlow aims to capitalise on.  Users can 'flow' (share) tracks they like on the service with their followers (it's also integrated into social networks like Twitter and Facebook) who can then listen to each track in full.  If they go on to buy the track the person who 'flowed' the track gets 20% of the cost of the tune - quite a unique proposition.

Colin (We7) suggested that many companies are looking at 360 degree deals as a way of taking a slice out of merchandise and live performance opportunities in the face of declining physical sales. There are now a variety of different sources from which artists can generate revenue - record labels still have a role to play (eg skilled at marketing large acts like Madonna etc) but that role has changed in lieu of the DIY ethic. They aren't dead just yet despite it being frequently predicted. Going down the major label route is useful for certain kinds of artists but there are more tools available to help empower musicians today, enabling them to have greater control over licensing their content.

Ian (Napster) also added that major record labels are good for developing artists through their career but the labels do need to see some return on their investment early on (usually by the second album).

Free or pay?

The question of whether or not people were still willing to pay for music reared it's head.

Dave (Soundcloud) made the point that the growth in digital is not replacing or making up for the decline in physical sales directly. The reason for this is largely due to the industry itself as having been the barrier to the growth of digital. Digital Rights Management (DRM) software in the case of Napster and iTunes are cases in point. Fearful of piracy the major labels insisted that digital files be encumbered with software which monitored users and limited how they used the music they had legitimately purchased often in multiple competing file formats that were not always cross-platform compatibile.  This was a headache for the typical casual music consumer.  The industry effectivly applied the brakes that prevented mass migration to digital.  In recent years a relaxation of the rules governing DRM and the growth in streaming services has acted to redress this - convenience of access should bring more people in and therefore bring in more money from smaller payments in greater volume (eg Spotify's ad-supported streams). We are seeing a shift from  a period of music ownership towards music access.

Atan emphasised that ownership is still important to many music fans, if not all of them, and catering to a diverse set of interests in a flexible manner is what is needed. People will buy content but they are becoming more discerning about what they are willing to pay for.


The chair raised the point that most music listening (80%) is done via the radio and most of that is done in cars. How do companies monetize this?

None of the panelists were able to tackle this mobile issue directly.  Colin was keen to highlight that We7 is adding more social functionality to their newly upgraded radio service which allows people to share their musical taste with friends via Last.FM recommendations, etc.  It also allows people the chance to discover and hear music related to their intersts.  One of the problems many music fans face when confronted with huge catalogues of millions of tunes is the scope of such access results in them freezing up due to the limitless possibilities - this is where profiling and recommendation radio services come in.

Ian claimed that when Napster started (as a legal service!) people didn't really 'get' what a monthly music subscription service meant. Clearly we've moved on since then. Historically people used to trust formats (radio, CD, etc) as you knew where you were - a CD played in a CD player - the proposition was simple. A subscription service didn't always make sense to people and took a while to become simple enough for casual users.  Formats have largely been static. Going forward you'll want to access your music that you've selected wherever you want, whenever you want - that will be where digital will innovate.

Dave returned to the point that early services crippled by DRM that involved platform specific third-party clients installed to Windows partition drives on Apple Macs, where users regularly had to make the effort to reset all the DRM licences on one machine in order to take the music on the go. As an early adopter, he was willing to make the effort for the benefits but he was not typical of the casual market. The dream is a cloud-based solution which includes a mixture of tracks owned alongside those leased via a service in much the same way that Spotify works now; combining a users personal iTunes library with that of the cloud-based service.

Recommendation and discovery

The chair posed the question asking how important recommendation engines will be in the future for music discovery?

Atan talked about some of the research mFlow have put into various algorithms.  This showed that people like to talk to each other about music and share their thoughts. This (sharing) aspect is one of the most important issues that couldn't be talked about a few years ago, when the industry associated sharing with giving away content for free, piracy and the fear of declining sales.

Colin claimed that personalisation is important part of the user experience. If a service feels personal and  less mechanistic or pushed at you by the industry, then music fans relationship worth music changes. We7 is increasingly looking at social media integration.

YouTube Fear?

The chair recounted an anecdote about how his 13 year old son consumes music, typically via YouTube, and questioned whether or not the audio quality of the service of the content itself was worthwhile. For him YouTube is primarily a video platform.  As one of the key spaces for new music discovery amongst the youngest demographics is YouTube a threat to digital music distributers?

Dave took umbrage at the chair's intonation that YouTube may be full of 'crap' or questionable content, or even that it is just a video platform rather than music. However Dave sees the platform as an essential creative outlet irrespective of how professional the content is. In this instances YouTube is a democratising tool that allows people outside of the mainstream system to gain some traction. Some smaller acts are seeing 4 figure sum monthly returns from their YouTube content alone. Companies are increasingly hiring youth teams for advice on how to generate more subscribers and connect with music fans. YouTube gives good control to labels or acts in that there are various controls offered to users (eg around video embedding, uploading and playlist controls).  Companies or acts can generate money around their video content by taking advantage of Google Adwords and links to buy.

Atan suggested the reason why YouTube is successful with kids is that it just works.

Ian highlighted a problem within the industry which seems to have a bizarre distinction in how they view video.  They view video as a marketing tool, partly due to legacy reasons associated with promotional campaigns, and often are not overly concerned if music videos leak to YouTube before they've secured an audio distribution deal or licensing with digital retailers.  But the digital start-ups trying to make headway in the audio distribution space -  companies like Napster -  they aren't allowed to carry these promotional releases in audio form until an official release date has been agreed. Even users uploading tracks and videos to YouTube are viewed by many in the industry as an adjunct to marketing.

Dave suggested that it would be in the best intersets of digital retailers to close the release window gap. The industry is configured to work over a staggered release window with their eye on promoing tracks in print or waiting for confirmation they've made it to Radio 1's playlist. A release date may be way off but the viral nature of the Internet means tracks can leak quickly, circumventing the often slow processes of the industry. Some music journalists (the NME was named) still insist on review copies on CD so that journalists can make some money from selling them! This has led to a recalcitrance regarding digital as they'd lose this perk.


The question turned to whether or not the freemium business model was a feasible one?  Typically this is often described as giving away content for free (often ad-supported) to the vast majority of users whilst offering a paid-for premium service taken up by around 10% of users.  The paying customers typically offset the costs of the free users.  I've discussed this model a few times on this site so regular readers will be familiar with the idea.

Ian started by describing Napster's early attempts at dabbling with it a few years ago in the US but they struggled to make enough money required to pay the labels for the licenses from the ad-funded service.

Colin's company, We7, currently works with both freemium and subscription models. They have around 10000 UK paying subscribers with millions more accessing content for free. Back in April We7 announced that they had managed to make the ad-funded model work for them.  This is important as much of their funding is still reliant on venture capital.  Their recent switch to the  radio stream makes a lot of sense as it typically cost 30% less than the licences for subscription services.

Dave pointed out that being free allows you to get heard when radio playlisting isn't working for artists. Soundcloud is less of a consumer destination, rather its a space where musicians (both amateur and professional) can upload and share their content with certain controls, circumventing traditional radio.  Even if you give content away for free you have to be savvy about the social contract involved and get something back - eg data on users, geography etc. This can help musicians targe certain geographies for  live tours or related campaigns. Soundcloud is built of the premise of the social share: ensure you get something in exchange for the download.  You can give users of the service the ability to download tracks providing they tweeting or embed the content in Facebook for increased presence. Location-based data is increasingly important and there is promotional potential in this area.  Recently James Blunt gave away music to people that checked in to a certain place at a certain time using the GPS in their phones and a location-based application like FourSquare, Gowalla or Facebook Places.

From this point on the panel took questions from the audience.  It became clear that one of the biggest barriers preventing start-up companies from joining to what seems like an already overcrowded digital retailing space is the cost of licences from the major labels for delivering digital content to music fans.  The majors seem to think that cannibalisation of the existing business models will occur if the costs they charge come down. Digital Music News recently carried a feature listing 100 companies that have tried to make it and have failed.

Dave noted that some labels' business models revolves around ensuring they extract as much venture capital money as possible from new startups.  This focus on short term economic gains undermines the ability to develop, nurture and sustain new platforms and opportunities for growth. Initially the industry didn't think it needed these new digital startups but the decline in paying customers has forced a redress of the balance somewhat.

The panelists where in agreement that there will still be innovation in the technology going forwards despite the licensing barriers, notably around the ability to build API's around limited types of access to free services.  Features like this helped to sustain Twitter's growth and it isn't inconceivable that something similar could happen with music catalogues, creating interesting data mash-up services.

One current problem that many digital services face is a lack of standardisation around metadata, eg things like playlists, top-rated tracks, etc - a playlist created on We7 cannot currently be taken to Spotify if a user decides to switch music services.  This will become increasingly important as more services will be built upon the same 11 million or so tracks currently licensed.  The prediction here is that metadata will be more important that tracks owned.

You can catch up with the event via the video on the Generator site here.

No comments:

Post a Comment