My recent hack at the MIDEM music hack day, DJ Spotify, seemed to peak the interest of quite a few corners of the music and technology industry. In the week leading up to the hack day I put some thought into the business implications of a fully integrated DJ-friendly streaming service, thinking through some of the potential benefits, pitfalls and possible financial models.

Throughout this post I will use Spotify to refer to all streaming services of a similar nature, and Traktor to refer to all DJing software packages of a similar nature. There are of course other services and pieces of software available (e.g. Deezer and Omnifone for streaming, Serato and VirtualDJ for DJing etc.) however Spotify and Traktor represent their respective fields sufficiently well to paint the picture.

The industry is already moving into cloud-based, streaming solutions. Pulselocker, for example, is already offering a streaming solution tailored for DJs, with an offline mode, integration with Traktor and Serato and a community of users on its site. In terms of licensing, however, Pulselocker is approaching the problem from the other end of the spectrum, effectively hashing out new agreements with labels to form an ever growing collection of dance music in a range of genres, rather than partnering with any of the existing streaming services. Full credit to Pulselocker for charging into the space, and some of the features listed below are fully present in their software, but I list them here anyway so as to give a full view of the benefits of a fully integrated system (i.e. one where the streaming catalogue and DJing software are in the same software package).

Services such as Spotify already have APIs, and programs like Traktor already have great, industry standard software DJing solutions, so, technically, a fully integrated system is already possible. The last remaining barrier that the streaming service providers have to overcome would be the licenses that they have in place with the rights holders. These currently prohibit any type of audio manipulation, so there would have to be some license renegotiations.

Pacemaker iOS App is the first to offer DJing capabilities using Spotify

Interestingly, Pacemaker, an iOS app that allows you to DJ using Spotify content, came out yesterday, so those license renegotiations must have happened somewhere, and hopefully Spotify will apply them to their public API soon. Also, Pacemaker is decidedly a casual/consumer product, so there is still much to play for when it comes to the pro market.

Bearing all of this in mind, it’s interesting to consider what new opportunities streaming DJ programs will offer. A bit of a thought experiment, if you will…

A cautionary tale

The suggestions below are predominantly relevant to the preparation process and workflow of the modern DJ. I do not foresee this new system making significant changes to how DJs actually perform live, and the implication is certainly not that having instant access to millions of tracks will mean that DJs should pick out and play new tracks that they’ve never heard before while standing in front of a dancing audience. At the end of the day, DJs will still need to trawl through hundreds of tracks to find the right ones for their sound, they will still need to sort them into playlists, tag and rate them, and practice mixing in and out of them, just as they do now, but combining streamed content into the workflow will make all of those tasks easier and hopefully more enjoyable. Having said that, internet connected streaming services will allow DJs to better handle requests, but I wonder if that’s a good thing…

Features

In short, the catalogue of Spotify needs to combine with the power and feature set of Traktor, along with a sprinkle of Beatport, a dollop of SoundCloud, and a dash of Google Analytics.

Adaptations of existing pieces of software

Instead of dreaming up an entirely new feature set to build on top of Spotify, it’s more helpful to consider how software like Traktor could be adapted to suit this new method of DJing, seeing as Traktor already has all of the features that professional DJs want and expect to see. However, adding the ability to stream content necessitates some departures from the current streaming service feature set. Some of these include:

Higher-quality format options
The debate of audio quality is incredibly hot among DJs, and it divides the community quite fervently. Having the option for WAVs would be a necessity in order for this to be taken at all serious by a good proportion of DJs, while other compressed formats such as FLAC are also widely discussed and should also be offered. This new wave of software options could potentially make it as easy as flicking a switch in a preferences panel to choose what format of audio you want to access, allowing the debate to continue, for better or for worse…

Offline mode
Although it’s not unreasonable to expect clubs to have solid internet connections in the future, a robust offline mode will be essential as a means of back-up, especially if higher quality formats are being used. An initial thought would be to automatically ‘offline-sync’ each and every playlist that a user of the software has saved, with an option to opt-out of the offline sync on a per-playlist basis to save on hard drive space if desired.

Better browse and search functionality
Traktor will need to build more advanced browse and search functionality into it’s software so that DJs can easily traverse a catalogue of 22m+ tracks. Being able to sort by label, genre, release date etc. would be essential, and being able to browse trending tracks and DJ charts would be extra ‘nice-to-have’s. Basically, recreating all of Beatport’s industry standard functionality.

New opportunities – the fun stuff!

All of these elements coming together will offer some really exciting, revolutionary opportunities:
User accounts
By it’s very nature, streamed content is centrally stored and served, meaning that the software accessing the content needs an idea of who is using the service i.e. a user account. With Spotify, the data associated with an account includes playlists, subscription level etc., so the next wave of DJ software would need to adopt such an element, too. The added benefit of having user accounts is that you can log-in on a different computer and have instant access to all of your playlists and settings.

Essentially, this will mean that, when a critical mass of DJs start using streaming services, clubs and bars will start providing a house computer or laptop for the DJ to use, just like they currently do with house CD decks. A DJ will simply turn up to the venue and log-in to the software on the house machine and start playing.

User accounts will also be great for the distribution of ‘promo’ records. Dance labels regularly distribute unreleased tracks to tastemaker DJs. Instead of manually sending DJs this promo content via e-mail or separate digital services such as Label Worx or zipDJ, labels could simply upload the unreleased tracks to the streaming service and grant access to these tracks only to specific user accounts.

Community
User accounts add a new dimension to what you can do with DJing software, allowing DJs to communicate with each other. Many DJs have a close circle of fellow DJs that they share their current favourite tracks with via group e-mail or other means. This kind of communication could be incorporated into a streaming DJ system, again, centralising everything.

Better pre-production workflows
Diligent DJs will spend hours researching, buying and editing records. A common workflow for a DJ would be to trawl Beatport for hours (even days sometimes) looking for that perfect extra weapon to have in their armoury, but the low-quality two-minute nature of the digital store model means that once they’ve paid for and downloaded a track, it could turn out to be completely wrong, or the two-minute preview was the only good bit of the track. Pulselocker do have a nice alternative on the ‘two-minute preview problem’ by allowing users to skip around the entire track, but only listen to 30 seconds at a time, but this still doesn’t solve the bigger problem.

Effectively combining Beatport’s store into Traktor’s performance tool could allow DJs to do ‘live auditioning’ whereby they play a record or loop that they already know and like, and try out lots of different unknown tracks (the whole track – not just a two-minute preview) at the click of a mouse, complete with automatic beatmatching, effects and looping, making the pre-production process much more efficient.

Unreleased or edited content
Many DJs make personal re-edits of tracks to elongate or shorten certain sections, to make them more useable in their sets, or to simply personalise them a bit. Furthermore, DJs will sometimes use a track that they themselves are currently working on to test out in a club, or share demos with their immediate DJ community. If this solution is to be fully cloud-based, there would need to be a way to accommodate these more ad-hoc uses of music i.e. tracks that are not provided by labels/rights-holders, and this is where the SoundCloud element of the recipe could come in handy. DJs should be able to upload their own content to store in their own ‘locker’ and choose to share it with certain other DJs, or to simply have accessible when they log-in on another machine. Interestingly, the need to allow DJs to make their own edited versions of tracks means that there will still need to be an option to download and ‘own’ the track via the traditional means.

Analytics
For labels, the most exciting aspect of a streaming DJ service is the analytics and reporting that they will be able to get for their content. For the first time, they will know data points such as exactly who is playing their records, how often, where in the world, what time of night, what they played before and after their records, and much more. Labels will need a dashboard to observe this data, however this would make more sense as a separate, web-based service, rather than being baked into the software itself.

Again, this could be very powerful when it comes to promo content. Labels will instantly know which DJs on their lists listened to or played the promo, and this information could potentially replace the annoying questionnaires that labels ask DJs to fill out. There are prototypes for this kind of reporting present in online promo distribution platforms, but a fully integrated software package will be able to go far deeper.

Analytics leads us onto another interesting and new topic – how these services make money for artists, labels and platforms…

Financial model

The potential financial models of a DJ-friendly streaming service are ripe for innovation, and it will be very interesting to see what pricing options the first companies to jump on this go for. Streaming services have a great opportunity to open up substantial revenue streams by capitalising on the professional nature of their target demographic.

While the ‘premium’ £9.99 models happily accommodate the various types of music listeners that currently use streaming services, there is a great opportunity to monetise the DJing ‘vertical’ for significantly higher monthly subscriptions. Furthermore, having higher fees for such a service could open up the possibility of higher royalty rates and therefore more revenue that the streaming platforms can offer labels, which will surely help with license renegotiations. Higher royalty rates seem like a necessity in this future vision, as the labels will be granting far wider reaching licenses than the ones in place in current streaming services, what with the ability to commercially exploit and manipulate the audio.

Broadly, I foresee two potential financial models coming through – a premium monthly subscription for unlimited access, and a pay-per-play model, where users pay a monthly fee based on how many tracks they play in that month.

Fixed-fee subscription

This is effectively an unlimited monthly subscription model which doesn’t need much explanation, but suffice it to say that it would need to be a significantly higher price than the £9.99 norm we see with ‘consumer’ streaming services – this is a service for professionals who can afford it (and it’s not uncommon for DJs to spend £40+ buying tracks from Beatport prior to a single gig), and building all of the above features will justify a higher price tag. Something between £40 and £60 sounds about right, but I’m not backing this up with anything other than some very quick back-of-envelope maths. Having said this, it’s a fact that it will be incredibly difficult to find a one-size fits all model, so perhaps there would have to be multiple fixed-fee tiers based on usage, a la Pulselocker.

Pay-per-play

For more occasional or bedroom DJs who can’t justify a significant monthly fee, a pay-per-play/pay-as-you-go model could work very well. A ‘play’ would constitute anything above two minutes, and the amount a DJ pays is based directly on how many tracks they played in that month.

Power, responsibility, and revenue

Software manufacturers will now be providing much more than just the core DJing functionality – they will be providing the entire toolchain for producers, DJs and labels. This means they will be entitled to revenue streams that they have never had access to previously. To the DJs, platform fees will have to be built in to the financial models listed above, and this will force these packages to move to a more SaaS business model. Instead of charging a one-off fee, future DJ packages could effectively adopt the Freemium business model, whereby the software is free to download and use with locally stored content, with the paid plans only being necessary for all of the extra features associated with streaming.

Many of the other features could also be monetised on a sliding scale. The analytics reporting for labels, for example, could be sold in a tiered system, by which all labels receive basic reporting for free, with a premium being paid for more in-depth data.

Other thoughts

Piracy

Just like the music listening industry, piracy features in the DJing community. Granted to a lesser extent, but it still exists. Interestingly, it seems that much of the piracy that occurs is not due to not wanting to pay for a record, but out of convenience. If a DJ wants to share a released track with some of his or her peers, they’re more likely to send them the file itself rather than send them a link to the track on Beatport. With a streaming system, this loss would be mitigated.

Yet more democratisation, or is it…

Again, it seems clear that the world of streamed music and DJing are on a collision course, and that the benefits will be huge for everyone involved. Detractors will no doubt berate the idea as they did the advent of CDJs, software-only DJ solutions, the Sync button and most other revolutionary technologies, and that is par for the course. Interestingly, this is one technological advance that I feel doesn’t actually make the art of DJing any easier – only more enjoyable by the removal of annoyances.

What remains true is the following: as the trend of music listening marches further into permission versus ownership, the upcoming generations of musicians who want to become DJs will ‘own’ fewer and fewer records. I myself am an example of an early adopter of Spotify, consequently having bought very few CDs that were released since 2007. Future generations will grow up in an entirely content-streaming world, so it follows that professions such as DJing will start transitioning, too.

Where to from here?

The best placed companies to turn this vision into a reality are undoubtedly the manufacturers of the DJing software programs, i.e. Native Instruments (who make Traktor), Serato and VirtualDJ. Their collective piece of the puzzle is by far the most difficult to replicate – they have decades of experience in the professional market between them and have the complete and trust of their customers.

Spotify, Deezer and Omnifone obviously have instrumental parts to play in this, but which of those three (or any others) get in to this game first is not of critical importance to the end user. A particularly dedicated company (or even the software manufacturers) could feasibly go and obtain new license agreements with labels, themselves. For the extra features e.g. promo distribution, analytics, social plug-ins etc., the software company could make a few very tactical acquisitions and be well on their way. That said, it’s probably more likely that a solid partnership between a software manufacturer and an existing streaming service will form.

We will continue to research this field by talking to DJs, label owners, and technology companies and post more on this blog as interesting opinions and view points surface. Do follow us on Twitter (@reactify) or sign-up to our mailing list if you’d like to keep informed on the what we find…