A governance-driven deep dive into Reseller & Internet Sales performance using Microsoft Fabric.
This project demonstrates:
This project reimagines the Microsoft AdventureWorks dataset as a modern analytics case study on commercial performance, governance, and ethical decision-making.
Built end-to-end on Microsoft Fabric, it exposes how pricing strategy, reseller participation, and customer growth dynamics influence long-term business sustainability — and what stronger governance could have prevented.
A fully integrated Microsoft data stack powers the workflow.
| Layer | Platform | Description |
|---|---|---|
| Data Source | SQL Server | Restored from AdventureWorks.bak (extended version) |
| Ingestion | Power BI Gateway + Fabric Data Pipelines | Secure import and refresh automation |
| Storage | Fabric Lakehouse (OneLake + Delta) | ACID-compliant, versioned storage |
| Modeling | Power BI Semantic Model | Measures, DAX logic, and relationship design |
| Visualization | Power BI Reports | Internet + Reseller perspectives |
Microsoft Fabric was chosen as the platform for this analysis because it provides:
- Enterprise-scale data management: Lakehouse architecture supports large transactional datasets like AdventureWorks.
- Seamless integration: Connects easily with on-prem SQL Server via Power BI Gateway.
- Advanced analytics pipelines: Enables historical and scenario-based analysis with minimal infrastructure setup.
- Centralized governance & security: Ideal for enterprise reporting and sharing insights with stakeholders.
- Power BI synergy: Dashboards, visualizations, and storytelling are fully integrated with Fabric datasets.
Get started with a Fabric trial:
Microsoft Fabric Trial Capacity
Why this analysis matters:
- Enterprise Data Integration: Harness 10 years of transactional data in a single, accessible lakehouse.
- Historical Analysis: Track trends in sales, discounts, margins, and units sold.
- Insights & Strategy: Understand the impact of discounts on revenue, identify growth opportunities, and spot profitability trends.
- Stakeholder Communication: Present actionable insights in a boardroom-ready, investor-focused format.
flowchart TD
%% Left: SQL Setup
subgraph SQL["💾 SQL Setup"]
A[💾 Download AdventureWorks206.bak<br>from Microsoft Learn] --> B[💾 Restore backup<br>to local SQL Server]
end
%% Right: Fabric Setup
subgraph Fabric["🏗️ MS Fabric Setup"]
C[🏗️ Setup MS Fabric<br>trial capacity] --> D[🏗️ Create new Fabric workspace<br>for AdventureWorks]
D --> E[🏗️ Create new Lakehouse<br>in Fabric workspace]
E --> F[🏗️ Create new pipeline<br>to connect to local SQL<br>via Power BI Gateway]
end
%% Power BI Gateway (common step)
B -->|Data access| G[⚙️ Download, install & configure<br>Power BI Gateway] -->|Connects SQL → Fabric| F
%% Data Import & Reporting (converging)
F -->|Validated tables| H[📥 Import tables into Lakehouse<br>& validate]
H -->|Processed data| I[📥 Download Power BI Desktop<br>& import tables from OneLake]
I -->|Build semantic model| J[📊 Build semantic model,<br>measures,<br>tables,<br>and reports]
J -->|Publish reports| K[🚀 Publish reports to<br>AdventureWorks Fabric workspace]
%% Styling
classDef sqlBox fill:#cce5ff,stroke:#333,stroke-width:2px;
classDef gatewayBox fill:#d9ccff,stroke:#333,stroke-width:2px;
classDef fabricBox fill:#ffd9b3,stroke:#333,stroke-width:2px;
classDef importBox fill:#ccf2f2,stroke:#333,stroke-width:2px;
classDef reportBox fill:#d6f5d6,stroke:#333,stroke-width:2px;
classDef publishBox fill:#e6ffcc,stroke:#333,stroke-width:2px;
class A,B sqlBox;
class C,D,E,F fabricBox;
class G gatewayBox;
class H,I importBox;
class J reportBox;
class K publishBox;
- Source: AdventureWorks .bak file
- Pipeline: Fabric connects securely to SQL Server via Gateway
- Destination: Lakehouse in Fabric
- Visualization: Power BI dashboards built on imported tables
Top-line growth masked critical margin risks across both Reseller and Internet Sales channels — driven not by discounting, but by uncontrolled pricing decisions lacking elasticity insight and governance discipline.
| 📦 Reseller Sales | 💻 Internet Sales |
|---|---|
|
|
Strong analytics mean little without strong governance.
This project highlights the risks of chasing growth through reactive pricing and unmonitored channel dependencies.
What should have been done:
- Strong Oversight: Embed regulatory and ethical governance to enforce disciplined pricing strategies.
- Balanced Growth: Align commercial ambitions with sustainable profitability and risk control.
- Fabric + Power BI Synergy:
Transforms complex sales data into strategic intelligence — uncovering margin risks, governance gaps, and price-performance trade-offs. - Ethical, Data-Backed Strategy:
Drives sustainable, compliant, and insight-led decision-making across both sales ecosystems. - 🧩 Establish price-change governance with board-level accountability.
- 🌐 Enforce reseller diversification thresholds per region.
- 💹 Use profitability simulations to test the real impact before approval.
- 🔍 Maintain data lineage and ethical oversight within Fabric workloads.
| Theme | Observation | Strategic Response |
|---|---|---|
| 📈 Growth Surge | Top-line increase across channels | Reinforce data-driven pricing models to sustain gains |
| Erosion from unmanaged price reductions | Introduce elasticity-based decision frameworks | |
| 🧠 Governance Fix | Lack of pricing accountability and oversight | Deploy Fabric + Power BI governance dashboards |
💡 Outcome: A data-driven, ethically grounded growth strategy — balancing commercial ambition with sustainable profitability and strong accountability. .
This project was designed as an end-to-end commercial intelligence framework showcasing how Microsoft Fabric and Power BI can transform raw AdventureWorks sales data into governance-ready business insights.
- Data Sources: AdventureWorks fact and dimension tables representing Reseller and Internet Sales channels.
- Data Ingestion & Transformation: Leveraged Microsoft Fabric Dataflows and Lakehouse integration for structured ingestion, cleansing, and harmonization.
- Semantic Modeling: Built unified Data Model with relationships, hierarchies, and calculated measures (e.g., Margin %, Unit Price Variance, Elasticity Index).
- Developed Power BI semantic model exposing dual perspectives — Reseller and Internet — for comparative analytics.
- Implemented dynamic DAX measures to uncover pricing elasticity, margin volatility, and revenue–cost interactions.
- Created cross-channel governance views aligning profitability insights with business accountability.
- Embedded governance metrics to track compliance with pricing thresholds and ethical standards.
- Unified insights under Power BI dashboards, delivering executive-level visibility across growth, margin, and governance metrics.
- Clarity over complexity: Every visual and metric connects directly to a strategic decision driver.
- Governance-first mindset: Designed to highlight not just “what happened,” but “how responsibly it happened.”
- Ethical AI & Data Integrity: Built on transparency, interpretability, and traceable pricing vs. sales intelligence.
💡 Result: A modern analytics ecosystem — powered by Fabric and Power BI — where data, governance, and strategy converge to deliver sustained, ethical profitability.
- Sales vs. Margins: Trend over 10 years by product category.
- Discount Impact Analysis: Show how discounts affected revenue and gross margins.
- Units Sold Trends: Correlation with pricing and promotions.
- Executive Summary Dashboard: Ready for boardroom presentations with clear KPIs.
| Document | Description |
|---|---|
| 📊 Full Analysis Flow | End-to-end data prep, modeling & regression workflow |
| 🎯 Executive 1-Pager | High-impact overview for senior stakeholders |
| 🧩 Presentation Deck | Visual storytelling & business interpretation |
| Tool | Purpose |
|---|---|
| Microsoft Fabric | Unified lakehouse, pipelines & semantic modeling |
| Power BI | Data storytelling & commercial dashboards |
| SQL Server | Source system for AdventureWorks data |
| Delta Lake / OneLake | Reliable, versioned data storage |
| GitHub | Documentation, transparency & version control |
- Clone this repository.
- Download & Restore
AdventureWorks2016_EXT.bakto local SQL Server. - Configure Power BI Gateway to connect to your local SQL Server instance.
- In Microsoft Fabric, create a new workspace and Lakehouse.
- Run the data pipeline → validate tables in Lakehouse.
- Open Power BI Desktop, connect via OneLake → publish reports to Fabric.
- AdventureWorks 2016 Extended: Microsoft Sample Database.
- Power BI Desktop: Free/Pro version for dashboard creation.
- Microsoft Fabric Trial: Used for lakehouse and pipeline creation.
- Icons & Emojis: Public domain / Unicode.
- Implement Advanced Pricing Analytics Develop models to study price elasticity and demand sensitivity, ensuring pricing strategies optimize both volume and margin sustainably.
- Integrate Predictive and Prescriptive Analytics Use Python or Azure ML integration to build forecasting and scenario simulation models that guide future pricing and sales strategies.
- Enhance Governance Monitoring Dashboards Build Power BI dashboards that track compliance to pricing policies and margin thresholds, flagging risks in real time.
- *Develop Interactive Scenario Simulations Enable stakeholders to simulate various pricing, discount, and volume scenarios to visualize business impact through what-if analysis.
- Extend Storytelling for Different Audiences Tailor storytelling dashboards and slides for executives, sales teams, and governance committees to support data-driven decision making.
- Document Governance Framework and Ethical Guidelines Explicitly outline governance principles and ethical sales standards informed by data insights, to embed accountability in commercial operations.