Newsletter

How the rise of Authorless Music will bring Authorful Music

Posted on December 9, 2019 | by Bas Grasmayer

Forty thousand. That’s the number of songs being added to Spotify every day. Per year, that’s nearly 15 million. With AI, we are approaching a world where we could easily create 15 million songs per day. Per hour even. What might that look like?

*Can music experiences performed by robots be Authorful? (photo: **Compressorhead**)*

The music trend we can most linearly extrapolate into the AI age is that of utilitarian music: instead of putting on an album, we put on workout music playlists, jazz for cooking, coffee time Sunday, music for long drives.

Artists have become good at creating music specifically for contexts like this. It often forms a big consideration in marketing music, but for also the creation process itself. But an artist can’t be everywhere at once. AI can and will be. Meaning that for utilitarian music, artificial intelligence will have an unfair advantage: it can work directly with the listener to shape much more gratifying, functional music experiences.

This will lead to the rise of Authorless Music. Music without a specific author, besides perhaps a company or algorithm name. It may be trained by the music of thousands of artists, but for the listener it will be hard to pinpoint the origins back to all or any of those artists.

Do we want Authorless Music? Well, not necessarily. However if you track music consumption, it becomes obvious that the author of music is not important at all for certain types of music listening. Yet we crave humanity, personality, stories, context.

Those familiar with trend watching and analysis, know to keep their eyes open for counter trends. When more of our time started being spent on social platforms and music became more anonymous due to its abundance, what happened? We started going to festivals in numbers never seen before. So what counters Authorless Music?

The counter trend to Authorless Music is Authorful Music. Although there will be a middle space, for the sake of brevity I’ll contrast the two.

	Authorless Music	Authorful Music
Origin	AI-created or obscure	Human-created (ish)
Focus	Specialised in function	Specialised in meaning
Relation	Little emotional involvement	Strong emotional involvement
Trait	Personalized	Socialized

Authorless Music: primarily driven by AI or the listener is unable to tell whether the listed artist is a real person or an algorithm. The music is specifically targeted towards augmenting certain activities, moods, and environments. Due to its obscure origin, the listener has little emotional involvement with the creator (although I’m looking forward to the days where we can see AI-algorithms fan bases argue with each other about who’s the real King / Queen of AI pop). In many cases it will be personalised to the listener’s music taste, environment, weather, mood, etc.

Authorful Music: primarily created and / or performed by tangible people or personalities. It will be focused in shaping meaning, as it is driven by human intent which embeds meaning by default. This type of music will maintain a strong emotional link between artists and their fans, as well as among fans themselves. This music exists in a social way – even music without lyrics, such as rave music, exists in a social context and can communicate that meaning, context, and intention.

With the increasing abundance of music (15 million tracks per year!), the gateway to Authorless Music has been opened. What about Authorful? What experiences will we craft in a mature streaming landscape?

Two important directions to pay attention to:

Socialising music experiences

It’s so easy to make and manipulate music on our smartphones now. Whether it’s music as a standalone or accompanying something on Instagram or TikTok. One reason for this massive amount of music being added to streaming services is because it’s easier than ever to make music. With apps that make it easy for people to jam around with each other, we’ll see a space emerge which produces fun tools and basically treats music as communication. This happens on smartphones but is strongly complemented by the virtual reality and gaming space.

See: JAM, Jambl, Endlesss, Figure, Smule, Pacemaker.

Contextualising music experiences

There is a lot of information around music. What experiences can be created by exposing it? What happens when the listeners start to enter the space between creator and listener and find their own creative place in the music through interaction? (I previously explored this in a piece called The future of music, inspired by a cheap Vietnamese restaurant in Berlin)

Examples of this trend: lyrics annotation community Genius, classical music streaming service IDAGIO, and projects like Song Sommelier.

Special thanks to Data Natives, The Venue Berlin, and Rory Kenny of JAM for an inspiring discussion on AI music recently. You’ve helped inspire some of these thoughts.

New to MUSIC x? Subscribe to the free newsletter for regular updates about innovation in music. Thousands of music professionals around the world have gone before you.

What playing around with AI lyrics generation taught me about the future of music

Posted on December 2, 2019 | by Bas Grasmayer

Will AI replace human artists? What would the implications be? These questions grip many in the music business and outside of it. This weekend I decided to explore some lyric generation apps and see what I could get out of them – learning a thing or two about the future of music along the way.

Below I’ve posted the most coherent lyrics I managed to get out of one AI tool. I’m dubbing the song Purple Sun.

Image with a purple sun — *What I imagine the song’s artwork to look like.*

You can make the sun turn purple
You can make the sea into a turtle
You can turn wine into water
Turn sadness into laughter
Let the stars fall down
Let the leaves turn brown
Let the rainwoods die
Let wells run dry

I love the turtle line. I guess the algorithm struggled with rhyming purple.

Two lines down is a wine / water line. Initially I was impressed by having a western cultural reference. But hold up… turning wine into water? That’s just evil.

Read it over once more. Or twice. By reading it over more, I became convinced that obviously humans are the superior songwriters.

But you know what, I’ve been lying to you.

The origins of the above lyrics are actually human, from a 90s rave song called Love U More by DJ Paul Elstak.

And they carry meaning. A lot of meaning to a whole generation of people in The Netherlands and other parts of Europe. Myself included. The meaning comes not necessarily from what the intent of the lyrics is. It comes from the music, nostalgia, memories, associations.

This is listener-assigned meaning. As soon as you release music, you give over control of the narrative to an audience. Artistic intent may have a lot of sway, but sometimes a song that’s a diatribe against fame turns into something stadiums full of drunk people chant.

A few statements to consider:

AI has a role as a tool to be used by people to apply their creativity.
Not all successful human created art objectively requires a lot of skill.
Creativity doesn’t end with the creator. The creator sets intent, the listener assigns meaning.

Let’s pair #1 and #3. In the first statement I talk about people, rather than mention specific roles as in the thrid statement. That’s because AI allows more people to be creative, either as listener, creator, or the space in between.

It’s this space in between that will be impacted and shaped by AI. Think of the dadabots projects, such as their infinite neural network generated death metal stream, apps like JAM, Jambl, and Endlesss which allow people to express themselves musically in easy ways, or technologies that turn music into something more adaptive like Bronze and FLUENT (disclaimer: I’m an advisor to the latter). Not all of the above use AI, but all cater to this space in between listener and creator.

The reason why I added statement #2 is because AI-created music doesn’t necessarily have to be objectively good. Music is subjective. Its sucess depends on how well it can involve the listener. That’s why AI is destined to be the most important force for the future of music in a more creative world.

Credits for the lyrics above: Lucia Holm / Paul Carnell. Thank you for the wondrous energy, the memories, the music.

Image via Rising Sun.

New to MUSIC x TECH x FUTURE? Subscribe to the free newsletter for weekly updates about innovation in music. Thousands of music professionals around the world have gone before you.

Reducing music’s climate impact through innovation

Posted on November 25, 2019 | by Bas Grasmayer

Coldplay announced this week that they are not going to tour until they can figure out a climate neutral or climate positive way to do so. Touring has a massive CO2 output. U2‘s 2009 tour is said to have produced the CO2 equivalent of flying to Mars and back (or the annual waste produced by 6,500 British people, or the same as leaving a lightbulb running for 159,000 years, or flying 90,000 people at one of their stadium shows from London to Dublin – pick your favourite).

What are the implications? A single flight from London to Dublin produces about 100kg of CO2 per passenger, depending on the airline. Now consider this: “Each kg of CO2 ultimately melts about 650 kg of glacial ice.” Times 100, times 90,000. And that’s just for this one band, in perhaps the most polluting tour ever.

However, it’s audience travel that is the largest contributor to greenhouse gas emissions in music. Oft-quoted research by Julie’s Bicycle, published in 2008, estimates audience travel accounts for 43% of the UK music industry’s greenhouse gas output.

The Guardian article about U2 that I linked above, ends with a quote of a review by Mark Reed:

“The carbon footprint of this might be quite large, but the spiritual rewards to the audience of this are those that enhance a life. If all life were bread and water, then there would be nothing to lift mankind above the amoeba.”

I personally dislike how dismissive the quote sounds, especially when it’s used to conclude an article about an important problem. Mankind has a much worse chance of surviving catastrophic climate change than amoeba after all. However, the author does touch upon something important. The experiences that the music business provides are important (so important that it’s moved someone to describe a U2 concert as a spiritual experience). So what drives me are the following two questions:

How can we continue to provide these experiences?
How can we use these experiences to inspire a better world?

What comes to mind is a quote by Sammy Bananas, founder of DJs for Climate Action. Talking to Stuart Swift in an article on Stamp the Wax, the example of transitioning from plastic straws to paper comes up:

“While Sammy admits its impact on the climate is negligible, it “may have a much greater effect on making people wonder why the venue took the effort to make the switch.” This creates a snowball effect where “individuals want to learn more and engage.””

Alright. I hope the doom and gloom in the paragraphs above is enough of a motivator. Let’s look at actionable steps players in the industry can take to reduce our impact and inspire a better world.

Measure

Whether it’s understanding energy use of a venue or festival, waste produced, or the audience’s footprint, the first step to tackling a problem is to start mapping it. This makes it easier to research and identify areas of high impact and not get distracted by working on things which have negligible impact.

Few venues have an accurate idea of the energy consumption of everything inside. Sure, stage lighting and audio is often well-considered and engineered, but what about cooling systems, lighting in other spaces of the venue, heating, etc.

Do artists know their CO2 footprints? Understanding better what the environmental cost of a tour or a gig is, can help identify ways to address or mitigate that cost.

Then there’s festivals and the audience that travels to them. In considering audience travel, the main question is often a logistical one: how do we get everyone on-site in a decent timeframe? Is there enough road and public transport capacity? As audience travel constitutes a majority of a festival’s greenhouse gas footprint (Dr Jillian Anable & Julie’s Bicycle put it at two thirds), it’s worth tracking the problem and mapping it out.

Audience travel

The problem of audience travel emissions is important to approach within its specific contexts. What type of event are people traveling for? What are their modes of transportation? What type of (public transport) infrastructure exists?

Festivals may consider offering discounted combined entrance & public transport tickets, organising events closer to urban areas, and offering camping equipment rental and supply sales on-site. The latter is an important why people choose to travel by car to multi-day festivals.

In general, organisers should make sure public transportation services are mentioned in event communications, as audiences are often not aware of these options.

Energy use at events

*The UK’s **Shambala Festival** is often mentioned as one of the greenest festivals and managed to reduce its carbon footprint by 80% and is free of meat, fish, and disposable plastics.*

Paul Schurink, co-founder of Green Events Netherlands, is an expert in the field of temporary energy supply and as such has worked with countless festivals to improve their energy use. In an interview with Clubbing TV, he explains some of the basics. Some takeaways:

Smart power plans for festivals take about 3 years to build. The first year you go for the quick wins, and in following years you get a feedback loop of expertise and new practices. After 3 years, festivals can save 40% of their power per edition.
If a festival uses forty thousand liters of fuel for generating power, they’d need fifteen thousand liters less. Financially, that’s at least fifteen thousand euros saved. It also means less generators to rent, less generators and fuel to transport, less transport costs.
If you use less power, you’re more sustainable. Using less power also makes it easier to make use of sustainable energy sources like solar panels.

Live events as testing grounds

Duncan Stutterheim, founder of dance event organisation ID&T and the legendary Thunderdome events, set up Open House a few years ago. Akin to a startup incubator, they helped partner innovative young companies with events and NGOs. Together, they could find out more about topics like how the same innovative energy solutions used at festivals can be used for humanitarian aid, and also looked into circular use of waste.

An organisation called Innofest matches innovative startups with festivals to test their solutions out in the real world. Since their website is completely in Dutch, I’ll highlight some of their cases:

Ditching single use rain ponchos (15 minutes of use, 500 years to decompose): a startup called Weather Underground did a test at Noorderzon festival with ponchos that can be repurposed as a bag and are biodegradable (video in English).
Building tables of festival waste: Futuretable made tables from recycled waste at Welcome to The Village festival. By communicating that these tables were made of waste, they successfully encouraged people to recycle more because they could see and try out what their waste would turn into.
E-waste Arcade tested better waste separation by making it fun through sound-producing garbage disposal units at Eurosonic Noorderslag.
Plantjebandje is a compostable festival wristband that’s biodegrable and filled with plant seeds. Take the wristband off at home, plant it and see what sprouts. (article in English)
&Cricket tested what it takes to get people to eat insect-based food as a sustainable alternative to meat (a major contributor to greenhouse gases). Their cricket fries sold out.

Connect & stay up to date

Sustainability is becoming a major topic of my music & innovation newsletter MUSIC x TECH x FUTURE – followed by thousands of people in creative industries around the globe. If you’re not on the list, please consider subscribing.

If you are professionally active in the music industry or sustainability and collaborate with creative sectors, please drop me a short email introducing yourself — I’d like to invite you to the Music Tech Network Slack. I’ve set up a #sustainability channel and I want it to be a place to exchange knowledge, experience, and build connections. Reach me at bas@musicxtechxfuture.com

The listening time trap

Posted on November 11, 2019 | by Bas Grasmayer

My biggest gripe with most music business conferences is that I hardly ever bump into engineers, designers, or product managers. If they’re there, they’re presenting their company rather than talk about the work they do, like many other music professionals. There’s marketing talks, A&R talks, talks about bookings, management… Where are the design talks?

Part of this frustration is personal: when going to a conference, I want to learn from people. While music conferences help me build perspective, they hardly help me develop as a professional.

More importantly, I believe the people who have some of the biggest impact on the modern music landscape often aren’t actually part of the conference. It would help the whole business to better understand them, their goals, their motivations.

They’re the people who decide upon the interfaces through which we experience music and connect with creators. They decide upon algorithms. What feature goes in, what doesn’t. And they’re largely invisible.

How decisions get made in tech companies

Tech companies set themselves up for rapid growth. Either in terms of users, staff, or both. In order to do so, it’s important people inside an organisation have a framework for autonomy. They have to be able to understand the company’s goals at the top-level, and what that means for their team specifically. They should be able to derive goals from the top-level goals themselves.

This type of grass-roots level autonomy helps the velocity and quality of decision-making compared to old school top-down chains of command and approval.

One of the most popular frameworks at the moment is OKR: objectives and key results. A team decides what they want to accomplish in a certain timeframe (objective: “shape a delightful social experience around music”) and then defines ways in which they want to measure their performance on the objective (key result: “active users share music 5 times a week”).

Once everything is set, the time period is kicked off and the team works together to try to accomplish their objective. They might use data about the service, in-person interviews with users or potential users, and the advice from stakeholders around the company.

There is one metric so important, that you will almost always encounter it when spending some time in a digital entertainment company. Either as a “key result” or as a “health metric” to see how well the company is doing.

The most important metric in music

One popular concept in optimizing a company for growth is “One Metric That Matters“. It means giving company one metric to focus on improving during a given stage. This may be “customer lifetime value” (CLV): how much revenue does a user bring in as long as they use our service?

For many music streaming services, CLV will be composed of various factors. Does a user upgrade to premium or do they stay on the ads platform? How much are advertisers paying for ads? How long does a user stay with the service before ‘churning’ (leaving, and not returning)?

There is one metric that has arguably had more influence on music than any of the above: how do you make sure you get more ad revenue per user on average? How can you tell that a person is enjoying their subscription and are unlikely to unsubscribe (churn)?

Listening time.

How many hours per day, week, month, does a user spend listening to music on our service?

It can be a good health metric, and it can have a rather direct relation on revenue growth when applied to the ad-based free tiers of services.

So designers, engineers, and product managers get to work and try to figure out how to optimize the amount of time people spend listening to music on their service.

The never ending push for listening time

In come tools for curators to optimize their playlists: and out go songs that lead to skips. How many skips away from the pause button are we? Let’s not risk it.

Out come the algorithms which continue audio playback after an album or playlist finishes playing, which populate users’ home feeds with music they’re most likely to listen to at this time of day, which create context on artist profiles by showing the ‘related artists’ users are most likely to click on and listen to next.

What it also does is strip music of context. It removes music from circulation that is not optimal for performing on this metric. It values art based on metrics.

What happens when people listen to more music?

One could do academic studies on the above subject (and if you have done so, please get in touch with me), so for the sake of this article I’ll give a few examples of what happens.

Theme-based playlists and other features that make the friction of choosing something to listen to smaller. Indecisiveness = lower chance of playback = less listening time.
Decreased familiarity with the artists one listens to. Listening to a higher number of artists means that on average people will be less familiar with each individual artist and their music. This does not mean that people’s familiarity with their absolute favourite artists is necessarily affected. However when they don’t know which of these artists to tune into, they might go for option 1 and just pick something theme-based, put it in the background, and listen to hours of music from anonymous artists, because the user was never confronted with their names.
Decreased importance and awareness of context. Think of a feature like Spotify’s Discover Weekly. A great tool to get people to come back to the app every week and listen to something, perhaps even explore some new music. The challenge is that it presents music stripped of any context. It’s just a list of tracks based on what you’ve been listening to. Recently, that’s sent me into 80s dark wave and industrial, but I honestly have no idea about the landscape. Who were the important artists? Where did they come from? Who inspired them? What subgenres, microgenres, and adjacent scenes exist? What does the subculture look like? All sacrificed for convenience. (I actually think there are interesting business opportunities here, now that the music streaming landscape has matured in many countries)

All that, because of a business decision to focus on a metric, and hundreds of thousands of small decisions by thousands of designers, engineers, and product managers that then influence the future of music styles, scenes, and the way artists connect to fans.

Why focusing on listening time is inevitable

I love this age of music and although the last section may sound pessimistic, I’m actually excited by the ease of access of music and all the experimentation that exists now. I’m excited by how easy it is, relatively speaking, to build up a listener base these days.

The hard thing about the world we’ve created is that with infinite free media (which I consider a good thing, inherently) we’ve opened up a massive competition for attention. The amount of music people would listen to used to be as large as people’s disposable budget for music. Now, for $10 / month or even $0, we can listen to music 24/7 and never hear the same song twice.

This is the landscape in which companies have to build their business models, and the landscape in which the music industry has to identify business models. With advertising-based models it’s simple: you lose attention, you lose the revenue. With subscription-based models, it’s similar.

Music competes with podcasts, video game streamers, tv shows, cat videos, and unfunny pranks. Either on one platform like YouTube, or spread out over various platforms (Apple Music competing for attention with Netflix, for example). This competition for attention, unfortunately, has become a rule of the web.

The part on which we can work together is the how: how do we hold people’s attention? How do we connect them to what they care about? How do we generate revenue around that?

What do you think?

I’m curious to collect more perspectives. Add on by penning your thoughts on Medium, LinkedIn, your blog, or as a Twitter thread. Email me or ping me on Twitter (@basgras) with a link and I’ll include it in next week’s newsletter (sent out 18 Nov, 2019 – 4pm Berlin time).

Bonus

You made it to the end. Here’s a video of every time Mark Zuckerberg said “more”, “growth”, or mentioned a growth metric.

Free competes with paid and abundant competes with scarce

Posted on December 11, 2017 | by Bas Grasmayer

Facebook recently launched a sound library including tracks you can use for free on videos. People criticized the concept in a music business discussion group (also on Facebook, ironically). I would hear the same rhetoric that people have when they say bands shouldn’t perform for free, because it’s not just a bad practice, it is also bad for your peers.

But let’s look at the reality that people in music are complaining about.

1. There are many different types of artists

There are always going to be people who find it awesome to see their music used by other people: even if they don’t get any money for it. They may be college students who are just happy to see their music travel. They may be people working full time jobs who do a little music on the side and don’t depend on the income. They may be professional producers who put out these tracks to libraries as a type of calling card.

Either way: there is always going to be free music and you will always have to compete with it.

2. Giving your music away for free can actually work

You have to have a monetization strategy at the end of this, but the easiest way to win attention online is to make great ‘content’ (in this case music). This content should be available with as few barriers as possible: which means making sure it’s available for free. The second part of your strategy should include steps on 1) how to hold people’s attention after you capture it, and 2) how to identify opportunities to monetize your fanbase (I wrote about it in detail in this thesis).

But sometimes you don’t need a strategy for monetization. It’s not easy to get signed to big labels nowadays and it usually requires you to show that you can build up your own audience. One of my favourite examples of someone who successfully leveraged free is Alan Walker. An EDM artist with tracks that have more plays than some of the most popular tracks from stars like Kendrick Lamar. How? He released his somewhat odd music through NoCopyrightSounds, which specialised in providing YouTubers and Twitch streamers with music they could use for free, without fear that their videos would get taken down. Eventually, they soundtracked the whole subculture and put a new sound in EDM on the map (read more).

3. AI is going to one up everyone

We are seeing amazing developments in AI. The most recent example is Google DeepMind‘s AlphaZero, which beat the world’s best bot in chess after spending just 4 hours practicing. Startups from Jukedeck, to Amper, to Popgun, to Scored are all trying to make music generation easier.

We already see more music being released than ever before, but so far it has still depended on human output. Through AI, music is already being untethered from human productivity. Standing out in abundance is a minuscule problem compared to what it will be 5 years from now.

Free music libraries are the least of your problem

There is no singular music business or industry. Everyone is playing by different rules and all those rules will be upended every time there’s a big shift in technology. From the record player, to the music video, to the internet, to AI and blockchain, music is the canary in the coal mine and you have to have a pioneer mentality or else you are falling behind every day.

The people who are one step ahead may be underground today, but some are the stars of tomorrow.

By all means, let us discuss the ethics. But be careful not to let your opposition blind you to the point where you cannot see how a new generation of music is thriving and leaving you behind. Because then it’s too late. For you.

How I got over a quarter million plays on my Soundcloud

Posted on November 25, 2017 | by Bas Grasmayer

Building up a following as a DJ in the social web’s early days: a how-to for time travellers. 💫

Back when I was in college, my friend and I would go to a lot of parties. We also used to rap in a band together. Up until then, I had always been writing a lot of lyrics and would visit every hiphop gig in my city. When there was nothing better on, we’d go to student parties in a local club that gathered around 800 people every week, and in between dancing and chatting, we’d be rapping our lyrics over the beats of popular songs.

Then one day we stumbled upon the drum ‘n bass scene (with regular parties in my hometown being hosted by the renowned Black Sun Empire). I always thought electronic music was not for me, but it changed the way I looked at electronic music. Instead of trying to make beats on FL Studio, I started playing around with making electronic music. Then, one day, I stumbled upon a simpler tool that allowed me to mix tracks together. It carried the tacky name Mixmeister, but it is still my all-time favourite tool for making mixes from the comfort of (what was then) my bedroom.

I still wish a company like Native Instruments or Ableton would buy this firm, and release a better and renewed version of their software that hasn’t worked on my Mac for years. But I digress.

Up until then, I had been writing lyrics. Lots of them. Daily. I was involved in the “textcee” scene, which is how people participated in online rap back when it was still a little tricky to record and upload tracks. I participated in battles, topical challenges, wrote about complex (and often silly) subject matter, and really got my creativity out — all in text format. It was easy to distribute, light-weight, and it had its communities and forums.

Pre-Soundcloud

For DJs, it was harder. Bandwidth was not great, and back in 2006 or so, when I started, there were no good online communities. There was no Soundcloud, there was no Mixcloud, and YouTube only allowed videos of up to 10 minutes. My tools of choice, for hosting DJ sets, were YouSendIt, uploaded.to and MegaUpload. They were iffy and you always had to monitor that your files were not taken down, but they would do.

I thought a lot about the format. I never mixed over 80 minutes, because I wanted to make sure that fans (if I had any, and it was hard to tell pre-Facebook & Twitter) would be able to burn it to CDs and listen to it from their cars or home stereos.

I would write detailed information about my tracklists, for a number of reasons:

It’s only fair that the creators of the music get acknowledged – especially since I was sharing their music without permission;
If one of my listeners liked a track, I wanted them to be able to know what it was (there was no such thing as Shazam);
I put detailed time markings, so that people would be able to identify the transitions and the amount of work I’d put into blending tracks together.

I would post them to the forums where I was already going (as well as my MySpace), where I already had my fans because of my texts, together with the links. Here’s an example of such a tracklist:

Then I started a blog on Blogspot to post all the mixes. People would subscribe via RSS and get the posts through their RSS reader. I even added a way to get email updates when the RSS feed would be updated, by using a popular tool at the time called FeedBurner. When posting my mixes to forums, I would also always include download links but also a link to the blogpost, so I could build up my followers there, too.

I didn’t know it at the time, but what I was doing really helped with SEO. If people were Googling those tracks, they’d often find my blog, because not everything was on YouTube, today’s major streaming platforms were non-existent, and the underground was not represented well on iTunes. By sharing my mixes everywhere, I was also generating a lot of backlinks. I was publishing multiple mixes per month. Throughout 2007 I published as many as 35.

Then Soundcloud arrived on the scene

I’m not sure how or when I discovered Soundcloud, but it must have been in its early days back in 2008. I managed to register my first name as my username, which I have held on to ever since, despite people trying to hack my account and even being hit by a trademark claim by an American rapper (after I rejected offers to buy it).

This is where things really started taking off. Now I was able to collect streams instead of downloads. It was so incredibly convenient. No wonder DJs flocked to the platform. All fans had to do now was hit play, but the option to download and listen in high quality was there too. On top of all that, I was able to timestamp my mixes in a much more interesting way: by commenting the tracks.

Something else happened too. By tagging my mixes, it was possible for others to find my work. And by browsing tags, I was able to find other DJs. This was a first. Never before had there been as big a community of DJs. Never before had it been so easy to connect to others. Never before had it been so easy for producers and DJs to connect from the comfort of their bedrooms.

I started listening to other DJs. Commenting everywhere. I continued the same strategy of tracklists and tagging, which maybe also helped my SEO on Soundcloud. But I also didn’t give up on my website until many years later when Facebook was more established and it was getting hard to get people to visit websites. Owning your audience was important, and I always knew this. I needed to have my own place to keep the people who are interested in what I do connected to me.

Then in 2009, Soundcloud changed the rules of the game for DJs.

The first big DJ revolt on Soundcloud

When Soundcloud started, they allowed everyone to upload 4 tracks every month. Tracks could be of any length, or at least long enough to fit a DJ set, but if you wanted to upload more than 4 in a month, you would have to get a paid account. This was great for DJs, but it didn’t last.

In October 2009, Soundcloud switched over to a model with a maximum amount of minutes per account. Even if you’d upgrade to the most expensive monthly package there was no way to get rid of the maximum. It caused an uproar (link to discussion with participation of the founders – but layout is messed up, because it’s a cached page). I participated and tried to be understanding. The model made sense for producers, who were more likely to spend money on Soundcloud. It sucked for DJs though. I wanted DJs to think about what kind of model would allow for Soundcloud to monetize them and very actively participated in the discussion.

The people who participated in that discussion got lucky, and it’s really a token of how user-centric Soundcloud was in those days. A link was shared with the participants, where they could list their accounts, and they were given 30 extra hours. For me, that was about 30 extra DJ sets and it has lasted me to this day (I never matched my 2007 streak) — and I should have probably mentioned this in my ‘Benefits of Being an Early Adopter‘ piece. And props to David Noel, who was Soundcloud’s community lead. The email exchanges (and exchanges on Soundcloud’s support community) that I had with him stuck with me. I was writing my thesis at the time and when I graduated and got into music startups those exchanges were a big inspiration for my early career.

Life goes on

As Soundcloud grew into the giant it is today, I grew along with it. My taste grew, my following grew, my tactics and strategies evolved, and I saw new genres flourish on Soundcloud, such as moombahton.

Before all the download-gate bullshit, that make you jump through hoops, follow random accounts, like Facebook Pages, etc., it was pretty convenient to get free downloads from Soundcloud. I actually set up an IFTTT script that would automatically download tracks I favourited to my Dropbox. This way I could discover new music while I was working at Zvooq by day, in passive mode, and then by night play around with the files in my mixes.

I participated actively in the new, emerging online scenes. Commenting on tracks and connecting to amazing new talent emerging from the internet, rather than from a particular network of DJs. This got me a lot of listeners. I started making mixes in which all tracks were available to download for free. This had value in different ways:

I knew for sure that all DJs would be ok with me uploading this;
People would listen to them, because they knew they can find and quickly download new tracks through there;
I would link to all the tracks and afterwards comment on them to let people know I had featured their work. Sometimes they would share my music on their social media (this is before the repost function on SC).

If you’re not communicating your music this way, if you’re not networking with your inspirations, you’re not doing it right. This is probably how I got most of my plays from 2012 to now. Tactics and landscapes change, but some principles are true forever. Participate!

Other tactics not listed above:

Make playlists on 8tracks with the tracks of my mixes in order to promote my mix;
Try to win followers via social listening platforms like turntable.fm;
Make short mixes and post them on YouTube in order to find new audiences;
Facebook & Twitter accounts through where I would connect to segments of my audience.

My demise as a net-DJ

Then things got harder. It wasn’t any particular issue, but a lot of factors combined to halt me.

I switched to a Traktor S4 controller with Traktor software, so now I had to do all my mixes live. I’m a perfectionist, so this decreased my output. Digging also got harder: the communal nature of Soundcloud changed and a lot of DJs stopped offering their tracks as downloads (even when they’re not selling them). Others would put their stuff behind download gates, which just made it a pain in the ass to collect tracks and way more time-consuming. This also decreased my output.

As the number of mixes I put out decreased, so did the growth of my followers and my exposure to my audiences that were not directly connected to me. Followers ‘churn’ even when they stay part of your follower count. This means that followers go inactive on the platform they follow you on, so the follower count no longer translates to playback or other forms of engagement. This doesn’t matter so much when you’re new, but if you’re working on something for over a decade, it matters.

All of this compounded. It’s been about 5 years since I had a mix that got ~5k plays. And 8 years for 15k. But the lesson here is: to rack up following & plays, you can get lucky with a hit or just be insanely productive.

I’m at peace with what happened and now that I’m in Berlin, with talented friends as producers, plus friends in companies like Ableton and Native Instruments, I’m slowly getting back into DJing and producing. I haven’t put out a track in a decade, and no mix in 2 years, but I’m surrounded by the right people to get back into it… and do things right with all the experience I’ve collected plus that surrounds me. (if I actually end up having enough time — the irony of working in music)

Key takeaways

If I had to distill this into key lessons (and I do, because I owe it to you after reading 2000 words), these would be my main takeaways:

GET THERE EARLY. I got really lucky with being early to Soundcloud, but it also helped that what I was doing back then was not as common as it is now. Stay on top of developments in sounds and genres, and be slightly ahead of the curve, so you can shine a spotlight on up & coming talent. It will pay off when someone blows up.
BUILD YOUR FOLLOWING. Don’t trust in platforms: own your following. Connect them to your presence in many places, get their email addresses. Make sure your following is loyal, build trust, be consistent. If you’re slightly ahead of the curve, they know they’ll always discover new artists through you.
ALWAYS CREDIT PEOPLE. Scenes are small. Help each other. If you play someone’s music: list it. Don’t have time to provide a tracklist? Then you don’t have time to be a DJ. Sorry.
BE HELPFUL. This is connected to crediting: help people to understand the music they’re listening to. They’ll connect to you for this.
BE CONSISTENT & PRODUCTIVE. My best days were when I was a student. I don’t know how I found the time in between college and 12-20 hours of side jobs per week, but often I’d get home and get to mixing. I’d be doing stuff with music almost every spare minute. That’s the only type of dedication that really works.

I’ve had my run. Maybe I’ll do it again, but in a different way. I still like DJing, but prefer to do it live now. Besides, I have other ways to enjoy music now, such as my day job at IDAGIO, as well as MUSIC x TECH x FUTURE.

But to the generation that’s out there, on the cyber highways, hustling: best of luck & I hope this piece helps you.