AmaraSami/algerian-ecommerce-analysis

End-to-end data analysis of a simulated Algerian e-commerce dataset (2,000 orders) — data cleaning in Excel/Power Query, SQL analysis in PostgreSQL, and an interactive Power BI dashboard. # 🛒 Algerian E-Commerce Sales Analysis (2023–2024) A full data analysis project on a simulated Algerian e-commerce dataset (2,000 orders), covering data cleaning, SQL analysis, and an interactive Power BI dashboard. --- ## 📊 Dashboard Preview ### Page 1 — Revenue & Orders Overview ### Page 2 — Trends & Order Status --- ## 📈 Key Visuals ### Revenue by Wilaya ### Orders by Category ### Revenue Trend by Month ### Order Status Breakdown --- ## 🔍 Key Findings - **Alger** generates the highest revenue, followed by **Oran** and **Constantine** — reflecting Algeria's population distribution - **Electronics** is the most ordered category at **21.61%** of all orders - Revenue in **2024 is more stable** than 2023 — less volatility month to month, suggesting business maturity - **Cash on Delivery** is the dominant payment method, typical for the Algerian market - Only **8.65%** of orders were cancelled or returned — a healthy rate for e-commerce --- ## 🛠️ Tools Used | Tool | Purpose | |---|---| | Excel + Power Query | Data cleaning & transformation | | PostgreSQL | Data storage & SQL analysis | | Python (pandas, sqlalchemy) | Loading data into PostgreSQL | | Power BI | Interactive dashboard | | Jupyter Notebook (VS Code) | SQL queries & exploration | | Git + GitHub | Version control | --- ## 📁 Project Structure ``` ├── screenshots/ # Dashboard & chart exports │ ├── dashboard_page1.png │ ├── dashboard_page2.png │ ├── chart_revenue_wilaya.png │ ├── chart_orders_category.png │ ├── chart_revenue_trend.png │ └── chart_order_status.png ├── algerian_ecommerce_raw.xlsx # Raw + cleaned dataset (Power Query) ├── algerian_ecommerce_analysis.ipynb # SQL queries via Python ├── ecommerce.pbix # Power BI dashboard file └── README.md ``` --- ## 🗄️ Dataset - **Rows:** 2,000 orders (cleaned to 1,896 after removing dirty data) - **Period:** January 2023 – December 2024 - **Columns:** order_id, order_date, customer_i …