
The Sound of Success
Using data visualization techniques to map out the
Musical Landscape of Tiktok
My Role
Data analysis
Storytelling
Visual Design
Timeline
August 2024 - December 2024 (5 months)
Team
2 Designers
1 Engineer
1 Researcher
Tools
Figma
Tableau
Framer
Flourish
Excel
The *Interactive* Prototype
This is the final deliverable of the project, exploring the musical landscape of TikTok. We take an in-depth look at the acoustic and lyrical trends of the platform's most popular artists and tracks, uncovering the sonic elements that keep users coming back for more.
This interactive prototype allows users to manipulate values and uncover their own insights. You can select specific artists to analyze their trends, explore patterns by year, and much more.
Background
TikTok has transformed the music industry by making virality a key driver of success. However, this algorithm-driven popularity has also led to a homogenization of sound, with artists optimizing their music to fit TikTok trends. This project explores how data visualization can help uncover the patterns behind viral TikTok songs and provide insights for artists, producers, and industry professionals.
Data Sources and Characteristics
We analyzed datasets from Kaggle, containing information on popular TikTok songs from 2019 to 2022. These datasets were compiled using web scraping techniques, Spotify API data, and music charts. The data includes:
-
Song Details: Track name, artist, album, track popularity
-
Acoustic Features: Danceability, energy, loudness, tempo, speechiness, valence, etc.
-
Popularity Metrics: Artist popularity and track popularity scores
Key Stakeholders and Use Cases
-
Musicians & Artists: Identify viral trends while balancing creativity and TikTok-friendly elements.
-
Music Producers & Record Labels: Leverage insights to enhance production and marketing strategies.
-
Content Creators & Marketers: Optimize content strategies and predict upcoming music trends.
-
Music Analysts & Researchers: Study correlations between musical attributes and virality over time.
Project Goals
Identify Recurring Trends
Analyze the evolution of viral sounds and engagement patterns.
Highlight Outliers
Investigate songs that became viral despite deviating from trends.
Cluster Similar Songs
Group songs based on musical characteristics to uncover patterns.
Provide Actionable Insights
Help artists and marketers make informed creative decisions.
Exploration and execution
After gathering our data and defining our objectives, we initially explored it in Excel to identify emerging trends. Once we noticed patterns, we transitioned to Tableau for more advanced visualization and interactive analysis.
During our exploration, we discovered that Doja Cat’s breakthrough success was largely driven by the viral spread of her song Say So on TikTok, making her an ideal case study for our project.
​
To ensure compatibility with Tableau, we reformatted and structured the data accordingly. We then iteratively refined our visualizations, experimenting with different representations before finalizing the prototype.
For the overall structure, we followed a martini glass approach—starting with a brief exploratory phase, guiding the audience through key insights, and then allowing them to freely explore the data on their own.

Insights
Finally, we identified key insights to support users during the guided exploration phase. Using Doja Cat as an example, we demonstrated how our tool can be leveraged to explore different artists and trends, helping users navigate and interpret the data effectively.
We then built the prototype on Framer to effectively present our case study and showcase our findings.