In the ocean of over 120,000 new tracks uploaded to streaming services every day, emerging artists often feel like they're shouting into a void. With nearly a quarter of new tracks never being played, and 42% receiving fewer than 10 plays, the task of finding a fan base might feel like an insurmountable mountain to climb. This is sometimes referred to in music streaming as the cold start or “zero plays” problem.

Enter SoundCloud’s ‘First Fans’ initiative – a new tool powered by Musiio’s AI, designed to give artists a fair shot at being heard by the right audience. Initially built for creators with a Next Pro subscription, the 'First Fans' experiment serves new tracks to around 100 listeners in the critical window right after uploading.

Aron Pettersson, SoundCloud’s VP of AI and ML, along with his team, are the technical minds behind the product’s recommendation engine, working tirelessly to fine-tune the AI's music-matching gears. 

Giving Artists a Fair Chance to Get Heard

“First Fans is an attempt to give new music a fair chance,” says Aron. Imagine the untapped potential in those overlooked tracks – it’s a proverbial gold mine waiting to be discovered.

Addressing this "zero plays" problem required a new type of recommendation system – one that didn't merely focus on what was already popular, but gave equal attention to new uploads. This meant moving away from the traditional 'collaborative filtering' approach, which recommends content based on what similar users enjoy, to a ‘content-based’ recommendation system.

Technical Challenges

To bring this idea to life, there were 3 key technical challenges to overcome:

Knowing what music you have

Aron, along with lead engineers Vignesh Badrinath Krishna and Nikolai Tatarinov, and a wider team of over 20, used AI to construct a unique system to catalogue and tag new music as it’s being uploaded. The Musiio AI tech they built analyses the characteristics of every song, considering factors like recording quality, mood, genre, energy and more.

Optimising for the user experience

"Next, recommendations are performed by an AI that is figuring out what’s a good match for a user and what’s a bad match, and optimising for a good user experience," Aron explains. This process uses user information like listening history and followed artists, and inherently biases against recommending music to bots and fraudulent accounts. However, powering this process is no walk in the park. The AI needs to process thousands of recommendations every second - one of several challenges that took years of integration to overcome. 

Speed and Scalability 

"There's a lot of moving parts in it,” says Aron, alluding to the big-data challenges that SoundCloud had to contend with. One part of the solution was building cloud computing systems that balanced cost efficiency with speed and reliability, ensuring every track uploaded to SoundCloud is analysed in near real time. The system also includes a data pipeline designed for real-time recommendations, which was another non-trivial challenge.

Real-world Impact for Artists

Aron anticipates that the 'First Fans' initiative could exponentially increase the number of artists being discovered and making a living out of their music.

With new tracks being recommended to around 100 real people through the system, SoundCloud's curators gain valuable insights into the song's performance soon after uploading. Therefore, if a track resonates with listeners immediately, it can rapidly catapult the artist into a much larger audience.

What’s Next?

In the future, Aron and his team plan to explore further assisting undiscovered musicians and he believes that precise targeting is the key. He believes that no matter how niche or obscure a genre is, there's an audience for it. Interestingly, Aron adds that tracks in some niche genres have shown to perform better than mainstream ones due to less competition.

SoundCloud's 'First Fans' offers a promising solution to the discovery problem faced by emerging artists, demonstrating how AI can help musicians reclaim their careers. This experiment also brings the vision of Musiio by SoundCloud to life: proving we lead what's next in music and that we’re also making great music more discoverable.

To learn more about automated music tagging, and how descriptive AI can help your music business, drop a message to, or use the contact form. Alternatively, drop us a message on Linkedin or Twitter.

Share this story