🧾 Netflix Content Portfolio Analysis – Shows & Movies Optimization

Analyzing Netflix's global content library to optimize acquisitions, genre diversity, and viewer engagement using SQL, Python, and Tableu.

---

📌 Table of Contents

• Overview
• Business Problem
• Dataset
• Tools & Technologies
• Project Structure
• Data Cleaning & Preparation
• Exploratory Data Analysis (EDA)
• Research Questions & Key Findings
• Dashboard
• How to Run This Project
• Final Recommendations
• Author & Contact

---

Overview

This project evaluates Netflix's content portfolio across genres, age, geography, and ratings to drive data-informed content strategies. A full pipeline was built with SQL for ETL, Python for advanced analysis and stats testing, tableu and Power BI for interactive dashboards.

---

Business Problem

In the competitive streaming market, Netflix must optimize its library for retention and growth. This project addresses:

• Genre imbalances and diversity gaps
• Geographic content distribution for global expansion
• Content age and relevance for viewer churn reduction
• High-rated vs. underperforming titles
• Portfolio concentration risks by type/region

---

Dataset

• Netflix titles CSV (~8K+ shows/movies) from public sources
• Columns: title, type, director, cast, country, date_added, release_year, rating, duration, genres, description
• Enriched with geographic and age metrics

---

Tools & Technologies

• SQL (joins, Group by, Aggregations)
• Python (Pandas, Matplotlib, Seaborn, SciPy)
• Tableu (DAX, Interactive Visuals)
• GitHub

---

Project Structure

netflix-content-analysis/
│
├── README.md
├── .gitignore
├── requirements.txt
├── Netflix Portfolio Report.pdf
│
├── notebooks/                  # Jupyter notebooks
│   ├── exploratory_data_analysis.ipynb
│   ├── netflix_portfolio_analysis.ipynb
│
├── scripts/                    # Python scripts for ETL
│   ├── data_ingestion.py
│   └── content_summary.py
│
├── dashboard/                  # Tableu dashboards
Tableu files
│   └── netflix_portfolio.twbx
│
└── data/                       # Raw & processed CSVs
    ├── netflix_titles.csv
    └── content_summary.csv

---

Data Cleaning & Preparation

• Handled missing values: ~5% in country/director (imputed/mode-filled)
• Parsed multi-genres into lists; exploded for analysis
• Calculated content age: (2025 - release_year)
• Filtered invalid ratings/durations; standardized countries (ISO codes)
• Created summary tables: genre counts, geo aggregates

---

Exploratory Data Analysis (EDA)

Key Distributions:

• Content Age: 70% released post-2010; oldest ~1920s classics
• Ratings: Skewed high (mean 7.0/10); TV-MA dominant
• Geography: US (45%), India (20%), UK (10%)

Outliers:

• Longest: 5+ hour docs/epics
• Genre Overlaps: Dramas in 60% titles

Correlations:

• Release Year & Ratings (0.12 weak positive)
• Duration & Ratings (-0.08 for movies)
• Genres & Countries (strong regional ties, e.g., Bollywood in India)

---

Research Questions & Key Findings

1. Geographic Coverage: US/India dominate (65%); only 20% from Africa/LatAm → expansion opportunities
2. Portfolio Size by Type: Movies (70%), TV Shows (30%) → diversify originals
3. Content Age Breakdown: 40% <5 years old; 25% >20 years → refresh legacy library
4. Top Genres: Dramas (25%), Comedies (15%), Thrillers (12%) → over-reliance risk
5. Genre Diversity: Treemap shows thrillers/dramas 50% share; niches like Anime <2%
6. Ratings Distribution: 35% rated 8+; highest in International (mean 7.4)
7. Hypothesis Testing: Significant difference in ratings by region (ANOVA p<0.01) → localize content

---

Dashboard

Interactive Tableu dashboard highlights:

• Global geo map (content by country)
• Portfolio size/content age bars
• Genre diversity treemap/pie
• Top performers & ratings histograms

---

How to Run This Project

1. Clone the repository:

git clone https://github.com/yourusername/netflix-content-analysis.git

2. Install dependencies:

pip install -r requirements.txt

3. Ingest & process data:

python scripts/data_ingestion.py
python scripts/content_summary.py

4. Run notebooks:
   • notebooks/exploratory_data_analysis.ipynb
   • notebooks/netflix_portfolio_analysis.ipynb
5. Open dashboard:
   • dashboard/netflix_portfolio.twbx

---

Final Recommendations

• Acquire Regional Content: Prioritize Africa/LatAm (20% gap) for 15% retention boost
• Balance Genres: Invest in niches (Anime, Docs) to hit 30% diversity
• Content Refresh: Phase out >20yr old low-raters; focus on 2015+ originals
• Geo-Targeting: India/US strongholds → tailor promos; test thrillers in thrill-heavy regions
• Originals Strategy: High-rated internationals → scale for global hits
• Monitor Ratings: Threshold 7.5+ for acquisitions

---

Author & Contact

PRANAY JHA
Data Analyst
📧 Email: Pranayjha535@gmail.com
🔗 LinkedIn