A strategic data pipeline and visualization suite analyzing 500,000+ transaction logs to optimize revenue, inventory, and global expansion strategies.
This project transforms raw, noisy transactional data from a UK-based online retailer into actionable business intelligence. By integrating a Python ETL pipeline with a Tableau Command Center, the solution provides stakeholders with real-time visibility into market performance and operational bottlenecks.
Key Objectives:
- Cleanse and standardize raw log data (handling missing IDs, cancellations, and outliers).
- Identify high-value customer segments and geographic expansion targets.
- Optimize operational resource allocation based on peak trading times.
(Add a screenshot of your Tableau Dashboard here. Rename it 'dashboard.png' and upload it to your repo)
Based on the analysis of the FY2010-2011 dataset:
- 🌍 Market Strategy: The United Kingdom drives 80% of total revenue. However, data identifies Germany and France as the highest-ROI targets for localized marketing and expansion.
- 📅 Seasonality: Revenue exhibits a sharp, predictable spike in November, signaling a critical need for Q3 inventory ramp-up to meet pre-holiday demand.
- ⏰ Operational Efficiency: Transaction volume peaks consistently at 12:00 PM. Support staff and server capacity must be scaled during this window to prevent bottlenecks.
- 📦 Product Portfolio: Low-cost volume drivers (e.g., Party Bunting, Cake Cases) significantly outperform high-ticket luxury items, dictating a volume-based inventory strategy.
| File | Description |
|---|---|
eda.ipynb |
ETL & Analysis. Python notebook handling data cleaning (IQR outlier removal), feature engineering, and statistical analysis. |
\tableau dashboard\dashboard tableau.twbx |
Interactive Dashboard. Tableau workbook featuring a geospatial "Command Center" layout with dynamic global filtering. |
\presentation slides\dashboard slides.pptx |
Presentation Deck. High-level slide deck exporting key visualizations and narrative story points. |
- Data Processing: Python (Pandas, NumPy)
- Visualization: Tableau Desktop 2024.1, Seaborn, Plotly
- Data Source: UCI Machine Learning Repository - Online Retail
- Clone the Repo:
git clone https://github.com/yoursmaddyy/Retail-Sales-Performance-Dashboard.git
- Acquire Data:
Download
Online Retail.xlsxfrom the UCI Repository (Data not hosted here due to size constraints). - Run Pipeline:
Execute
eda.ipynbto process raw logs and generate the clean dataset. - Explore Dashboard:
Open
dashboard.twbxto interact with the visualizations.
Distributed under the MIT License. See LICENSE for more information.