This project delivers a comprehensive analytical study of Netflix's global content catalogue spanning over 8,800 titles from 1925 to 2021. Working with a raw CSV dataset, the workflow covers every stage of the data analytics lifecycle: raw data ingestion, systematic cleaning and feature engineering in both Python and SQL, exploratory data analysis (EDA) to surface content trends, and a structured set of 30+ business queries designed to answer real strategic questions around content performance, regional distribution, and audience targeting. The result is a reproducible, insight-driven pipeline that maps the shape of Netflix's streaming library and reveals actionable opportunities for content strategy and growth.
- Clean and preprocess raw Netflix data — handling nulls, duplicates, and type inconsistencies
- Engineer derived features: primary genre, content duration value/unit, and date components
- Perform univariate and bivariate EDA to identify content distribution patterns
- Answer 30+ strategic business questions on content performance, regional trends, and growth
- Deliver a reproducible dual-language pipeline using both Python (Pandas) and SQL (PostgreSQL)
The project employs a dual-language analytics stack — Python for programmatic data wrangling and visualisation, and SQL for structured querying and business intelligence:
- Raw data ingestion and exploratory profiling of a real-world streaming dataset
- Multi-strategy null handling: mode imputation, row-dropping, and placeholder substitution
- Feature engineering — extracting primary genre, date components, and numeric duration
- Univariate and bivariate EDA with distribution analysis across content type, rating, and region
- Authoring 30+ SQL business queries covering content performance, regional segmentation, and risk
- Data storytelling and strategic recommendation framing for content and marketing teams
📁 Netflix-Data-Analytics/
├── 📂 Dataset/
│ ├── Raw_Data.csv
│ └── Cleaned_Data.csv
├── 📂 Python/
│ ├── Data_Cleaning.ipynb
│ ├── EDA.ipynb
│ ├── Data_Visualization.ipynb
│ ├── data_cleaning.py
│ ├── eda_analysis.py
│ ├── data_understanding.py
│ └── visualization_script.py
└── 📂 SQL/
├── Data_Cleaning.sql
├── EDA.sql
└── Business_Queries.sql
A visual walkthrough of the analysis outputs — from data cleaning notebooks to EDA charts and SQL business query results exploring Netflix content patterns.
Across 8,797 cleaned titles, the analysis found that Movies constitute roughly 70% of the Netflix catalogue (6,131 titles) while TV Shows make up the remaining 30% (2,666 titles). The United States is by far the largest content producer, followed by India and the United Kingdom. Dramas dominate genre output (1,600 titles), with Comedies and Action & Adventure close behind. The TV-MA rating is the most prevalent across the library, signalling a platform leaning heavily toward adult audiences. Content additions peaked in 2019 (2,016 new titles), with the 30+ SQL business queries surfacing actionable recommendations around release scheduling, genre investment, regional partnerships, and audience segmentation.
This project demonstrates how a structured analytics pipeline — from raw CSV to strategic business intelligence — can unlock meaningful insight from a large real-world dataset. By combining Python's data wrangling power with SQL's querying precision, the project covers the full spectrum of a data analyst's workflow: cleaning messy inputs, engineering informative features, conducting rigorous exploratory analysis, and translating findings into concrete business recommendations. The Netflix dataset proved a rich canvas for understanding content strategy, regional market dynamics, and audience behaviour at scale, making this project a strong showcase of end-to-end data analytics capability.