Status: 🚀 Completed

💻 [Click here to view the Raw Data & Python ETL Scripts in my GitHub Repository ] 💻

https://github.com/Nimna404/spotify_personal_music_Insight_dashboard2025.git

🎯 The Analytical Goal & Project Overview

Spotify's recommendation algorithm claims to know our musical tastes perfectly. But what does that taste actually look like in raw data?

Most music dashboards simply visualize what a user listens to. The objective of this project was to build an analytics engine to investigate why a recommendation algorithm keeps a user engaged. Using an export of my top algorithmic Spotify tracks, I built an end-to-end data pipeline to reverse-engineer my personal "audio footprint." By analyzing metadata such as Danceability, Energy, Valence, and Tempo, I set out to uncover the exact acoustic patterns driving my listening habits and form data-backed hypotheses about user retention strategies.

🛠️ The Tech Stack

• Data Extraction: Exportify (Bypassing API rate limits)
• Data Cleaning & ETL: Python (Pandas)
• Visualization: Microsoft Power BI (DAX, Data Modeling, Custom UI Design)

📖 The Data Dictionary

The raw dataset includes 24 columns. Key metrics include:

• Danceability: How suitable a track is for dancing (0.0 to 1.0).
• Energy: Perceptual measure of intensity and activity.
• Valence: The musical positiveness conveyed by a track.
• Tempo: The overall estimated tempo of a track in beats per minute (BPM).
• Popularity: Spotify's algorithmic ranking.

⚙️ Phase 1: Data Engineering (The Pipeline)