Predict Song Popularity: Insights from Spotify Data

Predicting Song Popularity: Insights from Spotify Data

📍 Project Overview

This project applies machine learning and statistical analysis to predict song popularity on Spotify, helping Universal Music optimize its $30,000 per-track promotional budget. By analyzing streaming data, audio features, and genre trends, the study identifies key predictors of hit songs and provides insights to maximize ROI on music investments.

🚀 Solution

✅ Market Trends Analysis – Identified evolving genre-specific trends influencing hit songs.
✅ Machine Learning Models – Built Decision Trees, Neural Networks, and Random Forest models to predict song success.
✅ Feature Engineering & SHAP Analysis – Determined Loudness, Energy, and Danceability as top predictors of song popularity.
✅ ROI Optimization Strategy – Developed a profit-maximizing model for Universal Music’s $30K/song investment strategy.
✅ Delayed Popularity Adjustment – Incorporated a 20% probability factor for songs that gain popularity later, improving forecast accuracy.

More works 📁

View All

Predicting Song Popularity: Insights from Spotify Data

📍 Project Overview

📈 Key Outcomes

⚡ Problem Statement

🚀 Solution

🔍 Tools & Methods Used

More works 📁