New Music Friday Nordic Special: Sweden

We predicted commercial performance of over 200 new Swedish tracks with a high degree of accuracy over both the short- and medium-term.

We predicted commercial performance of over 200 new Swedish tracks with a high degree of accuracy over both the short- and medium-term.


While the USA dominates much of the pop music scene, it’s a well-known fact that Sweden consistently punches above its weight.

Home of Spotify, the world’s largest music streaming platform (by number of paying subscribers), it has also produced some of the biggest names in songwriting and a host of internationally-dominant artists over the last few decades. It has a thriving domestic music industry, too.

So, in the first of a series of New Music Friday Nordic Specials, we’re going to predict the commercial potential of some recent new music from Sweden.

Specifically, we’re going to focus on tracks that have appeared on the Swedish version of Spotify’s New Music Friday playlist from the beginning of March to the first week of May 2019.

This also gives us a chance to showcase our ability to predict commercial performance of music sung in languages other than English.


Our test corpus was comprised of 209 tracks that appeared on the Swedish version of New Music Friday throughout March, April and into early May 2019.

  • The corpus included tracks sung in both Swedish and English, performed by both domestic and international artists such as Zara Larsson.

  • The 209 tracks analysed were performed by 180 unique lead artists.

  • Forty tracks included at least 1 featured artist, 5 of which included 2 or more featured artists.

Using our unique, music-focused algorithm, we predicted how many steams each track was likely to amass on Spotify 1 month after release.

We also predicted how big a hit each song was likely to be for the lead artist 3-6 months after release. Here are our results…


In terms of our ability to predict streaming performance, 8 out of 10 of our predictions fell within 10% of the respective artist’s typical performance range. Average error between our predictions and actual number of streams amassed was just 8%.

We were also able to classify two-thirds of all tracks into one of 6 performance ranges, e.g., <100K streams, 100K-1M streams, 1-10M streams, and so on.

Our predictions thus closely matched actual short-term performance.


In terms of predicting medium-term hit potential, we we correctly predicted 7/10 songs that would go on to be MAJOR HITS for the lead artist, and 8/10 tracks that would UNDERPERFORM for them.

For songs with featured artist, we correctly identified 9/10 MAJOR HITS for the first featured artist, and 100% of MAJOR HITS for the second featured artist.

Our predictions thus closely matched actual medium-term performance, too.


In sum, we were able to predict commercial performance of a diverse range of music from Sweden with a high degree of accuracy. In fact, the level of accuracy we achieved here is very much on par with our typical levels of precision, exemplified for example by our predictions for the biggest songs of Summer 2019 and for 1000 new tracks released in Q1 2019.

In real-terms, our algorithm allows artists, labels, and other rights-holders to predict how commercially successful (and financially valuable) a song is likely to be before it’s released. Even in Sweden!

Coming soon: New Music Friday Nordic Special - Finland

Want to know more about these results, or learn more about how we calculated them? Perhaps you’d like us to analyse some tracks for you? We'd love to hear from you

For more general info about how Hyperlive quantifies listener engagement and forecasts future performance, check out our Home or About pages.

About Hyperlive. Music isn’t something we just listen to — music is something we experience with our heart and soul. Hyperlive captures this experience by modelling a range of neurobiobehavioural responses to music as well as the psychological processes that underpin them. This gives us a deep understanding of what drives musical engagement on a fundamental level, allowing us to quantify, model and predict that engagement — and the musical features that motivate it — with unmatched levels of precision.