WE PREDICT COMMERCIAL PERFORMANCE OF FIFTY MILLION DOLLARS-WORTH OF NEW MUSIC EVERY WEEK.

 
We predicted commercial performance of each single released from Ariana Grande’s latest album,   thank u, next  .  By analysing the musical content of each track, we were able to predict its actual performance with over 90% accuracy .  Predicted and actual values represent artist-standardised Spotify streams .

We predicted commercial performance of each single released from Ariana Grande’s latest album, thank u, next. By analysing the musical content of each track, we were able to predict its actual performance with over 90% accuracy. Predicted and actual values represent artist-standardised Spotify streams.

 
Hyperlive Stats Background.jpeg
 

MODELLING MUSICAL ENGAGEMENT

Using an algorithmic approach, we model a range of neurological, physiological and behavioural 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.

 
 
We predicted how many streams + sales each track released as a single from Beyoncé's third solo album —   I Am... Sasha Fierce   — was likely to have amassed.  By analysing each track's audio signature, we were able to predict its actual commercial performance with 83% accuracy  . Streams and sales figures represent combined Spotify and YouTube streams plus stream-equivalent single sales .

We predicted how many streams + sales each track released as a single from Beyoncé's third solo album — I Am... Sasha Fierce — was likely to have amassed. By analysing each track's audio signature, we were able to predict its actual commercial performance with 83% accuracy. Streams and sales figures represent combined Spotify and YouTube streams plus stream-equivalent single sales.

 
Hyperlive Stats Background.jpeg
 

QUANTIFYING EARNING POTENTIAL

A major application of our model is in predicting the earning potential of new songs. Because our algorithm models fundamental, mass-scale responses to music, we can predict how successful a song is likely to be before it’s released. This helps minimise risk and maximise return on what’s often a substantial investment. In the process, copyrights associated with these more successful tracks naturally increase in value.

 
 
To demonstrate how our algorithm might be used in different contexts, we predicted which entries in the 2019 Eurovision Song Contest would be most likely to progress to the final.  Based solely on an analysis of the musical content of each song, we were able to predict the results of the two semi-finals with 75% accuracy , correctly identifying 15 of the 20 tracks that progressed from the semi-finals to the final, including the eventual contest winner.

To demonstrate how our algorithm might be used in different contexts, we predicted which entries in the 2019 Eurovision Song Contest would be most likely to progress to the final. Based solely on an analysis of the musical content of each song, we were able to predict the results of the two semi-finals with 75% accuracy, correctly identifying 15 of the 20 tracks that progressed from the semi-finals to the final, including the eventual contest winner.

 
Hyperlive Stats Background.jpeg
 

ANY SONG, IN ANY STYLE, BY ANY ARTIST, IN ANY LANGUAGE, PLUS SYNC, PLAYLISTS, MOVIES AND MORE.

We’ve accurately predicted commercial performance of tracks spanning dozens of styles and genres, with lyrics sung in more than 20 languages, created and performed by thousands of artists from regional talent to global superstars. These tracks have amassed hundreds of billions of streams, and amassed many billions of dollars in the process. What’s more, our algorithm has a range of additional applications across the broader entertainment industry, from sync and branding to movies, gaming and copyright-valuation. Contact us to learn more.