This repository contains the code, documentation, and resources for the Spotify Top Hits Analysis Project, which explores trends, insights, and recommendations based on the most popular songs from 1990 to 2020. Using Power BI and Tableau, the project visualizes data to uncover key patterns in music popularity and provides actionable recommendations for stakeholders in the music industry.
By creating this dashboard, we are essentially building an encyclopedia for artists—newcomers or veterans—to reference song trends, understand shifting audience preferences, and identify the factors driving popularity over the years. This project offers valuable insights into what makes a song successful, empowering artists to make data-driven decisions in their creative and marketing strategies.
The project focuses on analyzing Spotify’s top hits from 1990 to 2020.
Key Features:
- Data Analysis: Investigate song features such as danceability, energy, tempo, duration,etc.
- Visualizations: Create dashboards to identify trends in genre, artist dominance, and regional streaming growth.
- Recommendations: Provide actionable insights for aspiring artists and music producers.
Use Case: Understand the factors driving song popularity and emerging music trends over several decades.
The project follows a structured workflow for streamlined analysis.
High-Level Process Flow:
- Team Discussion: Collaborated to identify interests and finalize the Spotify Top Hits topic.
- Data Preprocessing: Cleaned and prepared the dataset for analysis.
- Visualization: Built interactive dashboards in Power BI and Tableau to derive insights.
- Recommendations: Synthesized findings into actionable insights for stakeholders.
- Source: Kaggle - Spotify Top Hits Dataset
- Details:
- Metadata: Track title, artist name, release year.
- Audio Features: Danceability, energy, loudness, tempo, duration, etc.
- Popularity Metrics: Spotify popularity score.
- Size: Contains 2M+ records of top hits over several decades.
-
Data Preparation:
- Handled missing or null values to ensure dataset integrity.
- Consolidated artist names for consistency and clarity.
- Converted song durations from milliseconds to minutes for better interpretability.
- Grouped years into decades to analyze temporal trends effectively.
-
Dashboards:
- Power BI:
- Visualized trends in audio features, artist dominance, and popularity metrics.
- Enabled interactive filtering by artist, genre, and year for detailed exploration.
- Tableau:
- Focused on regional streaming trends through geospatial analysis.
- Highlighted genre evolution and regional preferences using dynamic visuals.
- Power BI:
-
Insights:
- High-energy, danceable songs with 3-4 minute durations dominate the charts.
- Pop and hip hop are consistently the most popular genres over the analyzed period.
- While the US leads in streaming activity, Asian markets are rapidly growing in significance.
-
Recommendations:
- Create songs with high energy and danceability to align with listener preferences.
- Blend pop and hip hop genres to appeal to a broader audience.
- Focus marketing strategies on regions with increasing streaming activity, such as Asia.
spotify_top_hits/
├── data/
│ ├── Merged_dataset.csv
├── dashboards/
│ ├── spotify_dashboard_PowerBI.pbix
│ ├── spotify_dashboard_Tableau.twb
├── insights/
│ ├── DATA230-Team8-ProjectPresentation.pdf
├── README.md
- San Jose State University (SJSU): For supporting this academic project.
- Professor Venkata Duvvuri: For guidance and support throughout the class and lab.
- Spotify and Kaggle: For providing access to the data.
- Team Members: Sreenidhi Hayagreevan, Saumya Varshney, Victor Dumaslan, Bence Danko.