Streaming has become an essential part of the way we listen to music: over 400 million people pay for various streaming services (see report at musically) and the number is growing. With the increasing importance of streaming, the relevance of playlists also on the rise. But as an artist or promoter, how do I find the right playlist that best suits my song?
Finding the right playlist for your own songs means a lot of work. There are no current figures on how many playlists there are worldwide. Nevertheless, they play a big role in the cycle of a release. In addition to the very popular editorial playlists, there is also a whole range of so-called "user-generated playlists" that can attract a lot of attention just like the music editorial playlists.
To make the search for the perfect playlist easier, we have developed a playlist recommendation engine. The engine is based on our musicube AI, the core of our self-developed software. It helps to categorise and tag the audio files, but also to compare the songs for similarities and to establish connections between the data. With the similarity factor calculated in this way, our recommendation engine is therefore able to predict in which playlist you can put your new song in the running.
Let's take "Rock and Roll Shoes" by Johnny Cash, for example. A song that stands out because of its melodic-harmonic and very pleasing qualities. For this song, our playlist recommendation gives five popular playlists on Spotify as a result, such as "Old Slow Country Love Songs" or "Old Western Cowboy Playlist". Both of them definitely hit the right genre as well. But interestingly, the playlist "Aquarium Drunkard Presents: Quietly: Winter Is Blue" is also suggested as fitting. There's not a lot of country here, but songs with similar characteristics to "Rock and Roll Shoes". Who would have thought that the song, although assigned to the "country" genre, would still fit well here?
The first test customer is MPN, a project of PHONONET GmbH, a sampling platform in which over 3,000 music creators participate. Would you also like to test our playlist recommendation engine? Then feel free to contact us directly.