A data-driven Streamlit dashboard for Amazon vendors to analyze sales, product performance, regional trends, and inventory needs.
This dashboard connects directly to a MySQL database and provides an intuitive UI with filtering, forecasting, and visual analytics.
- 💰 Revenue Analysis: Track total sales, average order value, return/cancellation counts
- 📦 Product Performance: Top categories, best-selling sizes, monthly breakdowns
- 🌍 Regional Insights: Highest revenue states, category popularity by state
- 🔄 Inventory Planner: SKU-level restock recommendation with buffer %
- 🔍 Sidebar Filters: Filter data by garment category and selected month
- 📈 Interactive Visuals: Plotly and Seaborn-powered charts
- 🧠 Insight Section: Key trends extracted from EDA
Add your screenshots to a folder like
/assets/and rename accordingly.
Filters by product category and month
Key KPIs and monthly/daily revenue charts
Bar charts by product category and size
Top-performing states and categories
Restock simulator with recommended stock level
Key insights from notebooks/amazon-sales-data-analysis.ipynb:
- Western Dress and Set dominate revenue share
- Sizes S, M, and L drive most orders
- Business buyers spend more than regular customers
- April leads in revenue vs. May and June
- Cancellation rate of 14.22% is significant and actionable
| Layer | Tech Used |
|---|---|
| UI & Frontend | Streamlit, Plotly, Matplotlib, Seaborn |
| Backend | MySQL (via SQLAlchemy + PyMySQL) |
| Data Layer | Pandas, NumPy |
| Dev Utilities | dotenv, Jupyter Notebook |
- 🤖 AI-based SKU recommendation engine (OpenAI)
- 📈 Time-series forecasting (Prophet/LSTM)
- 🌐 Deploy publicly on Streamlit Cloud or Hugging Face Spaces
- Ishan Kumar – @IshankumarP
- Krithik Raman – PES1UG22AM087
- J.O. Shivnesh – PES1UG22AM073
- Rahul D – PES1UG22AM127