Tools: Python, Excel, Tableau Dataset: Tableau Sample Superstore (rebranded for simulation) Prepared by: Arsen Tagibekov Date: March, 2025
This project explores sales, customer behavior, and discounting strategy at UrbanMart using the public Superstore dataset. The goal was to identify profit leaks, segment customers, and recommend actionable improvements. Deliverables include data cleaning workflows, exploratory analysis, KPI definitions, and an interactive Tableau dashboard.
- Gross Revenue: ~$2.3M | Profit Margin: ~12.5% | Average Order Value: ~$458
- Unprofitable sub-categories: Tables, Bookcases
- Home Office segment = most profitable; Corporate = high volume, lower margin
- West region leads in sales; South region suffers from discount-heavy, low-margin orders
- Clear negative correlation between discount levels and profitability
-
Cut or review low-margin products (e.g., Machines, Tables)
-
Promote Phones, Binders, Accessories
-
Limit discounts > 20% on weak-margin items
-
Focus marketing on Home Office segment
-
Replicate West region strategy in weaker zones
| File | Description |
|---|---|
| 'Superstore_Cleaned.xlsx/.csv' | Cleaned dataset |
| 'Data_Cleaning.ipynb' & 'EDA.ipynb' & 'KPI_Calculation.ipynb' | Python notebooks for data preprocessing & data analysis |
| 'EDA_UrbanMart.xlsx' | Pivot-based EDA in Excel |
| 'Tableau_Dashboard_UrbanMart' | Interactive KPI dashboard |
| 'UrbanMart Analytical Summary.pdf' | Deep-dive technical analysis |
| 'UrbanMart Business Report.pdf' | Executive summary with recommendations |
| 'UrbanMart Project Presentation.pdf' | Final stakeholder presentation |
This project was conducted independently with curated support from ChatGPT for process structuring, review, and business communication.