<aside>
💡
📄 YT Music Analytics (summary)
├── 🧭 Overview
├── 🧾Summary
└── 🎯 Product Vision
└── 🧱 Architecture
└── 🗺 Roadmap & Delivery
└── 🎼 Epics A1–A4 : Library Analytics (Sprint 1)
└── 🎧 Epics B1–B4 : Listening History (Sprint 2)
└── 🔁 Epics C1–C4 : ETL Automation (Sprint 3)
└── 🔓 Epics DX : Non-Proprietary Database Migration & API Connectivity (Early Concept)
└── 📊 Dashboards
└── 🧩 KPI Table Global
└── 🔗 Links (GitHub / Power BI)
</aside>
🧭 Overview — YT Music Analytics
A production-oriented analytics project turning YouTube Music Google Takeout into analytics-ready models and dashboards, enriched with Spotify metadata, built with modern data-stack practices.
What this project demonstrates
- End-to-end delivery: raw files → warehouse → transformed models → BI dashboard
- Analytics engineering mindset: staging/intermediate/marts, tests, documentation, reproducible runs
- Product thinking: epics, incremental roadmap, clear deliverables, MVP-first approach
- Realistic data: personal dataset, but treated like a production source (quality + governance)
Scope
- ✅ Sprint 1 — Library Analytics (A1–A4): load + enrich + model + dashboard MVP
- ✅ Sprint 2 — Listening History (B1–B4): parse + load + enrich + model + Dashboard MVP (on going)
- ✅ Sprint 3 — ETL Automation (C1–C4): orchestrate + schedule + CI/tests + monitoring + documentation (Next phase)
Tech stack (simple)
- Source: Google Takeout (YouTube Music library + watch history + Spotify API)
- Ingestion: Python
- Warehouse: BigQuery
- Transformations: dbt (staging → intermediate → marts + tests)
- BI: Power Bi
- Orchestration : n8n / scheduled runs
How to review in 2 minutes