SOLUTIONS-> Hit Song analysis methodology



Disclaimer - To protect the integrity of our HSS technology this is a simplified explanation. Certain aspects of the technology have been omitted and other information has been expressed in simplified laymen's terms. It is intended only to give the user a basic understanding of how our technology works.

To have a hit song you should ideally be able to answer "yes" to the following three questions:
" 1. Does the song sound like a hit?
" 2. Does the song have encouraging mathematical patterns?
" 3. Does the songs have the right kind of promotion for the current market?
We can only help you answer the second question. But our research has shown that if the answer to that question is "no" your song most likely will not see commercial success and music labels are taking this research very seriously. You should too.

Through our partnership with Loudeye Technologies, a Seattle-based company we have been able to analyze approximately 3.5 million songs. This includes almost everything that has been released by the music labels since the 1950's until the present time. The database is updated weekly with new releases. The analysis application is able to "listen to" any CD and isolate patterns in many musical events, some of which are melody, harmony, tempo, pitch, octave, beat, rhythm, fullness of sound, noise, brilliance, and chord progression. This is a process called Spectral Deconvolution. Each song is then mapped onto a grid we call the music universe and is positioned according to its mathematical characteristics. Each song is represented by a dot on the universe and the songs on one end of the universe are vastly different from songs on the other end of the universe. Songs with mathematical similarities are positioned very close to one another.

What do hit songs have in common with each other?
We were so astounded by how songs with similar mathematical patterns related to each other that we asked ourselves a few questions:
Is music appreciation learned? Is it genetic? Is it tied to culture?
In the course of trying to answer those questions there was one question that also begged an answer. That question was:
What do hit songs have in common with each other?
In order to answer that question we removed all of the songs from the music universe that had not been hits in the past five years. An amazing thing happened. The songs that were left were grouped into a limited number of small clusters all over the universe but with vast spaces between them.

Hit songs seemed to have common characteristics. It was rare to see a song that fell outside of the clusters and had become a hit. But every time we spotted one we identified it and when we listened to a few of those songs it was immediately apparent why this was so. Most of the time it was due to their lyrical content which our application does not analyze nor understand. For example, some hip hop songs have become popular, not because of the music but rather because of the lyrics. A patriotic song that was released shortly after the September 11th, 2001 terrorist attacks in the US became popular, again, not because of the music but because of the power of the message in the lyrics

It then become obvious that many factors influence whether a song becomes a hit or not. Who the artist is. Where they are in their career. Who they are dating. What they wore to the Grammys and not least importantly how much money and effort was put into promoting a given song.

We realized that we could help music labels predict the hit potential of new music by analyzing new tracks and superimposing them on the already existing clusters.

Our technology helps labels see which songs may or may not have the mathematical characteristics that help predict success or failure in today's market. There are currently an overwhelming majority of misses promoted rather than "hits". Our technology is helping labels turn this ratio in their favor.

As we said at the beginning of this explanation:
To have a hit song you should ideally be able to answer "yes" to the following three questions:
" 1. Does the song sound like a hit?
" 2. Does the song have encouraging mathematical patterns?
" 3. Does the songs have the right kind of promotion for the current market?
We can only help you answer the second question. But our research has shown that if the answer to that question is "no" your song most likely will not see commercial success and music labels are taking this research very seriously. You should too.