End-to-End Business Case Study
This case study brings together all Power BI concepts covered in the course — from data connection to deployment — in a realistic business scenario.
Scenario: Contoso Retail Analytics Platform
Company: Contoso Retail — a multi-channel retailer with 50 stores, an e-commerce platform, and 500+ employees.
Challenge: Leadership lacks visibility into cross-channel performance, inventory efficiency, and customer behavior. They need a unified analytics platform.
Phase 1: Requirements Gathering
| Stakeholder | Needs |
|---|---|
| CEO | Executive summary: revenue, profit, growth trends |
| Sales VP | Regional performance, rep rankings, target tracking |
| Operations | Inventory health, supply chain, warehouse utilization |
| Marketing | Customer segmentation, campaign ROI, channel performance |
| Finance | Budget vs actual, P&L, cash flow |
| HR | Headcount, attrition, satisfaction |
Phase 2: Data Architecture
Data Sources: • SQL Server — Transaction data (POS, e-commerce orders) • Excel files — Budget and targets • SharePoint — Employee data and HR surveys • Web API — Marketing campaign metrics • Azure Blob — Historical CSV archives
Data Model (Star Schema):
┌── Products (Dimension)
│
Dates ── Sales ─────┼── Customers (Dimension)
│ (Fact) │
│ ├── Stores (Dimension)
│ │
│ └── SalesReps (Dimension)
│
├── Inventory (Fact) ── Warehouses (Dimension)
│
├── Budget (Fact) ── Departments (Dimension)
│
└── HRData (Fact) ── Employees (Dimension)
Phase 3: Data Preparation
Power Query Transformations: • Clean transaction data: remove nulls, fix data types, trim text • Merge customer table with loyalty program data • Append multiple CSV files from Azure Blob • Create calculated columns: fiscal year, customer segment • Generate a complete date table with fiscal periods
Phase 4: Data Modeling & DAX
Key Measures Created: • Revenue, COGS, Gross Profit, Net Profit • YoY Growth, MoM Growth • Customer Lifetime Value (CLV) • Inventory Turnover, Days of Stock • Employee Attrition Rate • Budget Variance
Time Intelligence: • YTD, QTD, MTD calculations • Same Period Last Year comparisons • Rolling 12-month averages
Phase 5: Report Development
Reports Created:
- Executive Dashboard (1 page)
- Sales Analytics (3 pages + drill-through)
- Inventory Management (2 pages)
- Financial Overview (2 pages)
- HR Insights (2 pages)
- Marketing & Customers (2 pages)
Design Standards: • Consistent color theme (brand colors) • Navigation bar on every page • Dynamic titles reflecting slicer selections • Tooltip pages for contextual detail • Bookmarks for view toggling
Phase 6: Security Implementation
RLS Roles:
| Role | Filter Logic | Assigned To |
|---|---|---|
| Regional Manager | Dynamic: [Region] = UserRegion | Regional managers |
| Store Manager | Dynamic: [StoreID] = UserStore | Store managers |
| Department Head | Static: [Department] = "X" | Department heads |
| Executive | No filter (all data) | C-suite |
OLS: Salary and cost columns hidden from non-finance roles
Phase 7: Deployment
Pipeline:
Development Workspace → Test Workspace → Production Workspace
Deployment rules: Different database connections per stage Scheduled refresh: Every 4 hours during business hours Gateway: On-premises gateway for SQL Server data
Phase 8: Governance
• Sensitivity labels applied: Confidential for HR and Finance, General for Sales • Certification applied to production reports • Audit logging enabled • Monthly review of access permissions • Documentation: Data dictionary, DAX measure catalog, RLS design document • Training: Conducted user training for all stakeholder groups
Lessons Learned
| Challenge | Solution |
|---|---|
| Slow report loading | Optimized DAX, reduced columns in model, used aggregations |
| Users confused by navigation | Added clear page navigation buttons and tooltips |
| Data quality issues | Implemented validation rules in Power Query |
| Too many ad-hoc requests | Created self-service reports with comprehensive slicers |
| Stale data complaints | Increased refresh frequency, added "Last Updated" label |
Skills Applied
Every concept from Modules 1-14: data connection, Power Query, data modeling, DAX (basic + advanced), time intelligence, visualizations, drill-through, slicers, bookmarks, RLS/OLS, Power BI Service, deployment pipelines, governance, and collaboration