What Makes a Hit Song: Analyzing Data from the Billboard Hot 100 Chart
Level up your digital marketing at: http://joshdviner.com/
Thank you to my team members Tanvir Singh and Nigel Phillip. Also, a big thank you to Jason Joven for helping us obtain some of the data used in this analysis.
I recently completed a 10-week Certification of Data Analytics at Brainstation (BrainStation). For the final project my team analyzed a dataset of the Billboard Hot 100 charts from the past 20 years along with their corresponding Spotify Song Attributes.
Our goal was to analyze the age-old question of what makes a hit song? By pairing Spotify data with the Billboard Hot 100 charts we thought we’d be able to identify the key attributes of hit songs. But it got a little complicated…
Definitions
First, let’s make sure we’re on the same page in terms of what’s what.
- The Billboard Hot 100 is the music industry standard record chart in the United States for songs, published weekly by Billboard magazine. Chart rankings are based on sales, radio play, and online streaming in the United States.
2. Spotify classifies every song on its platform with certain “song attributes.” It uses these attributes to build its powerful playlists and through its API it provides access to these attributes. Here are the ones we chose to analyze…
Challenges
- The initial dataset was fairly long. 80,422 rows to be exact. It actually only had 5,976 unique songs in it, which immediately told us how little turnover there tends to be on the chart.
- Tastes have changed. What made a hit song in 1999 is different from today. For example, Ricky Martin’s “Livin La Vida Loca” spent 20 weeks on the charts in 1999 while Lil Baby and Gunna’s “Drip Too Hard” spent 35 weeks on the chart between 2018–2019. Listen to the two songs and you’ll see our issue.
- We had to define popular. Was simply being on the Billboard Hot 100 “popular” or should we define it further?
Solutions
- We chose to only analyze the past 10 years. This narrowed down the dataset to 46,880 rows and 4,001 unique rows. This solved both the first and second challenge.
- We defined popular by songs that spent at least 13 weeks on the chart AND had an average weekly ranking of 50 or below. This decreased our dataset to 510 songs.
Key Attributes of “Popular” Songs
After refining our dataset we found the averages (and medians) for each key attribute.
The data told us that “popular” songs had:
- High danceability and energy
- Low acousticness and speechiness
- Tempo around 117 BPM
Additional Analysis
Interestingly, Kings of Leon’s “Use Somebody” spent the most number of weeks on the chart in our dataset at 53 weeks (in total, it spent 57 weeks but those additional four weeks went beyond our dataset). “Use Somebody” was followed by Imagine Dragons’ “Radioactive” (44 weeks) and Queen’s “Bohemian Rhapsody” (43 weeks — the song re-entered the chart after the release of the movie, Bohemian Rhapsody).
Additionally, Drake appeared the most on Hot 100 with 13 different songs. He was followed by Taylor Swift (10 songs), Maroon 5 (9 songs) and Rihanna (9 songs).
We also discovered a difference between genres for songs that spent the most weeks on the chart versus songs that charted highly. Below you can see how bands like Kings of Leon, Imagine Dragons, and Lady A all have songs that spent 40+ weeks on the chart, while more traditional pop artists like Halsey, Post Malone, Adele, and Drake all have songs with high weekly rankings. (Although part of it may be because our dataset ended in 2019 — Halsey’s “Without Me” went on to spend 38 weeks on the chart).
So What Makes a Hit Song?
The answer to the above question is complicated and a lot more data is needed to try to answer it. As well, there are a number of issues and questions that need to be addressed. For example, are Spotify’s song attributes and the Billboard Hot 100 accurate datasets to properly answer the question? How about the role of external factors (e.g., Prince’s songs rose up the charts when he passed away)? Do lyrics play a role? As a marketer, I believe much of what makes a hit song has to do with marketing — how do we account for that aspect? You get the point.
All that being said, based on our analysis there are certain song attributes that tend to be higher or lower for the most popular songs. Additionally, pop artists seem to chart higher on the Hot 100 and have more songs on it at any one time, while less traditional pop artists appear to have less songs on the Hot 100 but perhaps more staying power for the songs that do make the chart.
Going through the process of examining all the data and considering these questions demonstrate the complexity of answering this question. Furthermore, the framework of hit songs is constantly evolving, especially with the rise of TikTok and its influence on how hit songs are made. Attempting to answer “what makes a hit song” is the pursuit of a moving target. On that note, it only feels rights to end the article with this Drake quote,
“sometimes it’s the journey that teaches you a lot about your destination.”
Clap 👏 👏 if you liked this article to help others find it
Comment 💬 if you want to discuss (or just say hello! 👋)
