Did you attend 88rising’s Head in the Cloud concert? Eskwelabs uses data science to analyze 88rising artists.

Basty’s Notebook

Hi, everyone! My name is Basty, and I’m a data scientist and educator from Metro Manila. If there’s something I value so much, that is education. Education has allowed me to not only dive into the amazing world of code and data, but also to encourage and inspire others to do the same. Read more about me here.

Outside of work and school, I love playing video games like Valorant and League of Legends. I also love listening to Broadway musicals (HAMILTON, DEH, TICK TICK BOOM ALL THE WAY!). Lastly, I LOVE watching Friends, New Girl, HIMYM, and The Big Bang Theory.

Now, let’s take a look at my notebook!

December 2022 Notebook entry

It is now time to dive into the music industry, and see how data science can be a tool to empower our artists. Are you looking for a record label that represents a fresh, diverse sound? Then 88rising is the label for you. From pushing the boundaries of modern hip hop to showcasing Asian-American culture, 88rising has quickly become one of the most important and influential labels in the music industry. 

88rising is an influential music and media company that seeks to celebrate and give a voice to Asian and Asian-American artists. Founded in 2015 by CEO Sean Miyashiro, 88rising has released numerous successful singles, collaborated with famous musicians, and created original content such as documentaries and music videos. Some of their artists include NIKI, Joji, Jackson Wang, and MANILA GREY, who is actually the main topic of this blog post. 

So in this blog post, we’re gonna explore an analysis done by a team of data scientists named Dan, Emer, JB, and Ran with the overall objective of helping Manila Grey land a hit in the PH top 200 daily charts. Let’s get started!

Who is Manila Grey?

The stars of this analysis revolve mainly on the Filipino-Canadian singer-rapper duo MANILA GREY, who are composed of childhood friends Soliven and Neeko. The two have been making music for a while, but have only been active as the group MANILA GREY since 2016. 

In Spotify, they currently hold a 56 Spotify popularity score with 246,000 monthly listeners, and 67,077 total followers. And so with this information, the team wanted to help MANILA GREY land a hit in the PH top 200 daily charts in Spotify. 

Problem Formulation

They first started by identifying the questions they want answered in order to help MANILA GREY, and they came up with these:

1. What does the market look like?
   
   
2. What can we learn from similar artists?
   
   
3. What did the top songs do right?
   
   
4. What are the external factors? 

By identifying these questions, they were able to guide the flow of their analysis on how they can help MANILA GREY.

Data Gathering

In order to answer the questions they came up with, they first had to gather the necessary data needed for the analysis to continue. 

First, they used the Spotify App to download the PH Daily Top 200 charts from 2018-2020 with the help of Python. They then extracted and merged audio features and artist data for the top 200 songs using the Spotify API. 

The next set of data they extracted was the top 100 R&B playlists using the Spotify API, and also extracted the audio features for songs in the said playlists. 

Exploratory Data Analysis

After getting the data needed, they now did some EDA on it to explore the datasets and answer the questions they had in mind. Let’s go through them one-by-one!

1. What does the market look like?

Using the top 200 dataset, they were able to see that 1.6% of the songs were classified as R&B songs. With this information, they then compared the R&B playlist with MANILA GREY’s songs based on audio features where they found that the best genre that describes MANILA GREY’s music is in fact R&B.

2. What can we learn from similar artists?

For this question, they looked at similar artists that produce the same type of music, such as KIYO, FLOW G, and XXXTENTACION. By using a time-series graph, they looked into the different times that these three artists peaked in the Spotify Daily Charts to see how well they performed.

Next, they then compared the three artists to MANILA GREY based on the danceability, energy, and loudness of their songs. Their insight based on these graphs is that there doesn’t seem to be an obvious formula on how R&B songs can be successful. These graphs suggest that despite the difference on each of the artists performance in the Spotify Daily Charts, there isn’t much of a difference on the characteristics of their songs.

3. What did the top songs do right?

In order to answer this question, the team utilized two supervised learning algorithms, Logistic Regression and K-Nearest Neighbors, to see and classify which audio features belonged to the top songs in the daily charts. After getting an accuracy score of 67% for Logistic Regression and 78% for KNN, the team found out that the top songs had high tempo, low acousticness, and low valence.

4. What are the external factors?

Other than the audio features that contributed to the success of the top songs, the team also looked into different external factors that contributed to the success of other artists and songs. These external factors include:

‍
- Popularity

- Follower count on Facebook and Instagram

- Locations that people listen in

Insights

With all their guide questions being answered, the team was able to come up with recommendations that will answer their main question: How can we help MANILA GREY?

The team has identified two main recommendations to help MANILA GREY:

1. Compose songs with high tempo, low acousticness, and low valence
   
   
2. Collaborate with local artists

Based on the data and the analysis done by the team, MANILA GREY should be more formulaic about their song composition and establish themselves more in the Philippines to help boost their music careers.

Improvements

For every analysis being done, there will always be room for improvement. For this team, they’ve identified 5 points for improvement to come up with better insights with what they were able to gather.

1. Incorporate other sources of data such as social media activity
   
   
2. Recommend the best local artists to collaborate with
   
   
3. Try other machine learning algorithms that may give better results
   
   
4. Analyze song lyrics if we want to focus on audio features
   
   
5. Find data with genres attached to tracks and not artists

Succeeding Back Home

With the analysis done by Dan, Emer, JB, and Ran, they were able to showcase how data science can be used to help music artists better understand their market and audience, so that they can make informed decisions when it comes to their music.

Data science can also help other artists develop targeted strategies to effectively market their music, analyze their performance and revenue, and even optimize their fan engagement through social media.

With analyses such as this one, we can see that by fully leveraging the power of data, music artists can make decisions that will help them succeed in their career in the music industry.

Speaking of music and artists, the Head in the Clouds concert was held last weekend at the SM Festival Grounds Parañaque City! Did you know that MANILA GREY together with other artists from 88rising brought their hit singles and special covers on stage? I hope you got lost in the music last weekend and made new friends while you were at it!

Never stop learning!

‍