🛒 Online Retail Analytics & Interactive Dashboard Portfolio

🚀 Live Interactive Dashboards

No setup needed - open in any browser for full interactivity!

• Super Interactive Dashboard - Full dashboard with hover tooltips, zoom, and filters
• Realistic Sales Forecast - Interactive forecast for next 12 months

📊 Key Visualizations

Full Interactive Dashboard

Sales Forecast with Capped Growth

Customer Cohort Retention Heatmap

📈 Project Overview

End-to-end analysis of an online retail dataset (2009-2011, ~780k transactions):

• Data cleaning & preparation (handling cancellations, missing values)
• Exploratory Data Analysis (EDA)
• Customer Segmentation using RFM analysis
• Customer Retention with Cohort heatmap
• Time Series Forecasting using Prophet (with capped growth for realistic predictions)
• Fully interactive dashboards built with Plotly

Key Business Insights:

• UK accounts for ~90% of total sales
• Strong seasonal peak in November (holiday season)
• Most products priced under $5
• Most orders contain small quantities (1-10 items)
• Top customers drive significant revenue
• Forecast shows continued growth with realistic limits

📁 Repository Files

• online-retail-analytics-portfolio.ipynb - Complete code, cleaning, RFM, cohort, forecasting
• super_interactive_retail_dashboard.html - Main interactive dashboard (recommended)
• realistic_sales_forecast.html - Interactive sales forecast
• online_retail_dashboard.png - Dashboard screenshot
• sales_forecast.png - Forecast visualization
• cohort_heatmap.png - Customer retention heatmap

🛠️ Tech Stack

• Python (pandas, sqlite3)
• Forecasting: Prophet
• Visualization: Plotly (interactive), Matplotlib/Seaborn
• Database: SQLite (normalized schema)

⚡ How to Explore

1. Click any .html file for interactive experience (hover, zoom, explore)
2. View PNG images for quick visual insights
3. Run the notebook in Kaggle or Colab for full code details

Built for clarity - technical and non-technical users can discover insights easily.

---

By Omneya Saeid
GitHub: @Omneya21

⭐ Star the repo if you found it useful!