Math or Magic? New Software Predicts the Hits

What makes a good song? Is it A) topping music charts B) having dances conjured up based on the hook or C) having the perfect break-up lyrics you can blast after receiving your first “dear John" letter?

None of the above, according to CEO of Music Intelligence Solutions David Meredith.


“[It’s] a series of algorithms that we use to look at what’s the potential of a song to be sticky with a listener,” Meredith told NPR. “To have those patterns in the music that would correspond with what human brain waves would find pleasing.”

So, according to Mereidth, what makes a good song is more of a math problem than the stars aligning.

His software, Hit Song Science, rates songs based on common patterns of rhythm, harmony, chord progression, length and lyrics. And when Harvard Business School tested out the software, it found Hit Song Science was accurate eight out of 10 times. (For the romantics who still believe in magic, don’t fret, this means there’s room for error.)

Music Intelligence Solutions launched Uplaya, a website for songwriters who want to manufacture their next No.1.But not everyone likes to reduce their success in the industry to an equation.

Independent singer-songwriter Kim Tuvim told NPR the new software was “horrifying” to her and that she couldn’t stand the star-making machine behind popular songs.

“You’ll find a decreasing amount of any kind of surprises in music,” said Tuvim. “This just becomes a tool to make that narrowing of the field more accessible.”

As for me, I’ll take my music with a strong dose of magic and leave the algorithms for math class.

--Ashley Thompson


Currently listening to Dub Conscious
This content is solely the property of The Athens Blur Magazine and may not be re-produced without expressed written consent. The Athens Blur Magazine is an 8 issue/year music & variety publication proud to be based in the Classic City of Athens, Ga. For more information, please contact editorial@athensblur.com