Skip to content

arsentag/UrbanMart_Retail_Optimization

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

50 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

UrbanMart_Retail_Optimization

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.

Key Insights

  • 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

Recommendations

  • 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

    Deliverables

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.

Releases

No releases published

Packages

 
 
 

Contributors