A portfolio-ready exploratory data analysis project examining pharmaceutical sales across daily, weekly, monthly, and hourly datasets. The project uses Python to clean data, engineer time-based features, compare drug-category performance, identify seasonality, and translate analytical findings into practical business insights.
- 2,106 daily records covering 2014-01-02 to 2019-10-08
- 50,532 hourly records
- 8 pharmaceutical categories
- Time-series analysis at daily, weekly, monthly, annual, weekday, and hourly levels
- Reproducible Jupyter Notebook with documented analytical steps
- Cleaned dataset and high-resolution charts included
| Finding | Result |
|---|---|
| Leading category | N02BE — Paracetamol / Analgesics |
| Share of total sales | 49.4% |
| Strongest year | 2016 |
| Peak month | January |
| Lowest-demand month | July |
| Strongest weekday | Saturday |
| Peak hour | 19:00 |
- Which pharmaceutical categories generate the highest sales?
- How did total sales change across the available years?
- Which months show the strongest and weakest demand?
- Are there clear weekday or hourly purchasing patterns?
- Which categories exhibit distinct seasonal behavior?
- How can daily performance be classified into low, medium, and high bands?
- Which drug categories move together?
- Python
- Pandas
- NumPy
- Matplotlib
- Jupyter Notebook
- Data cleaning and validation
- Feature engineering
- Exploratory data analysis
- Time-series aggregation
- Correlation analysis
- Data storytelling
Pharmacy-Sales-Analysis-Python/
├── data/
│ ├── salesdaily.csv
│ ├── salesdaily_cleaned.csv
│ ├── salesweekly.csv
│ ├── salesmonthly.csv
│ └── saleshourly.csv
├── images/
│ ├── repository-cover.png
│ ├── 01_total_sales_by_category.png
│ ├── 02_annual_sales_trend.png
│ ├── 03_monthly_seasonality.png
│ ├── 04_weekday_sales_profile.png
│ ├── 05_hourly_sales_pattern.png
│ ├── 06_monthly_category_patterns.png
│ ├── 07_performance_bands.png
│ └── 08_category_correlation_matrix.png
├── notebook/
│ └── pharmacy_sales_analysis.ipynb
├── DATA_DICTIONARY.md
├── requirements.txt
├── .gitignore
└── README.md
- Loaded and inspected four time-granularity datasets.
- Validated data types, missing values, and analytical fields.
- Converted date fields and derived year, month, quarter, and weekday attributes.
- Calculated total sales across eight pharmaceutical categories.
- Built category contribution, annual trend, monthly seasonality, weekday, and hourly analyses.
- Classified daily sales into performance bands using the 33rd and 66th percentiles.
- Examined correlations between pharmaceutical categories.
- Exported a cleaned dataset for reuse.
git clone https://github.com/HussieniGamal/Pharmacy-Sales-Analysis-Python.git
cd Pharmacy-Sales-Analysis-Python
pip install -r requirements.txt
jupyter notebook notebook/pharmacy_sales_analysis.ipynbThe raw files correspond to the public Pharma Sales Data dataset. Drug categories follow ATC classification codes.
Hussieni Gamal
Data Analyst | Power BI | SQL | Excel | Python