Superstore Sales Analysis – Power BI Project

A complete end‑to‑end analytics project built using Power BI, focused on understanding sales performance, returns, and customer profitability across regions and product segments. This project demonstrates practical skills in data modeling, DAX calculations, dashboard design, analytical storytelling, and insight generation.

Project Overview

This project analyzes the Global Superstore dataset to answer real business questions:

• Which regions drive the most sales and profit?

• How do sales differ by order size and product category?

• Where do returns concentrate and how do they impact margin?

• Which regions and customers generate healthy margins, and which operate at a loss?

• How can management prioritize improvement efforts?

The final output includes:

• A fully functional Power BI report with 3 analytical pages and drill‑throughs.

• A PDF summary highlighting the key insights.

• Screenshots of dashboards for quick review.

Dataset

Source: Global Superstore dataset (public Kaggle dataset).

Contains:

• ~50,000 orders

• Sales, Profit, Quantity, Discounts

• Customer, Product, Category, Regional attributes

• Return records

The dataset was cleaned and modeled in a star schema:

• 1 Fact table — Orders

• Multiple Dimensions — Customer, Product, Region, Date

Dashboard Pages

Page 1 – Sales Overview

Sales Are Highly Concentrated in One Region

The Central region dominates revenue, while several regions underperform. This indicates strong dependency on a single market and untapped growth opportunities elsewhere.

Page 2 – Returns & Margin

Specific Products Drive Majority of Returns

Returns are not systemic — they come from a few problematic SKUs. Fixing those would improve profitability more than broad strategy changes.

Page 3 – Region Customer Matrix

Customer Profitability Varies Strongly by Region

Southeast Asia and South show many unprofitable customers, while EMEA, North Asia, and Canada maintain stable margins.

Full detailed insights are in Insights_Summary.pdf.

What I Learned

This project strengthened my end-to-end BI development skills across the entire analytics workflow:

Data Exploration & Cleaning

• Performed initial data profiling (distribution, missing values, outliers)
• Cleaned and normalized fields, fixed data types, reduced unnecessary columns
• Validated dataset structure for analytical modeling

Data Modeling Using Best Practices

• Designed a clean star schema with one fact table and multiple dimension tables
• Defined relationships, cardinalities, and filter directions
• Applied Power BI modeling standards (naming conventions, hierarchies, sorting, metadata)

DAX Calculations & Analytical Logic

• Created core business metrics:
  • YOY growth using DATEADD and time intelligence
  • Margin %, Profit %, Return Rate
  • Top-N customers using Boolean flags + disconnected slicer
• Used essential DAX patterns:
  • CALCULATE, FILTER, DIVIDE
  • Context transition + row/column context handling
  • Dynamic calculations controlled by slicers

Interactive Reporting & UX

• Built fully interactive dashboards:
  • Slicers (years, categories, regions)
  • Drill-through navigation
  • Tooltips with customer-level detail
• Implemented advanced conditional formatting for profit margin evaluation
• Balanced layout, readability, and analytical storytelling

Business Insight Development

• Translated visuals into clear, actionable insights:
  • Regional performance patterns
  • Return-driven profit erosion
  • Customer-level profitability variance
• Learned to interpret metrics, detect patterns, and explain business meaning clearly

Complete BI Case Delivery

• Structured the entire workflow from raw CSV → star schema → visuals → insights
• Packaged the solution as a professional project (.pbix + insights PDF)
• Prepared the project for a public GitHub portfolio

This project helped me connect technical analytics with practical business reasoning, and strengthened my ability to deliver end-to-end BI solution with PowerBI.