Welcome to the QueBox Profit Intelligence Dashboard project repository! 🚀
This end-to-end data warehousing and analytics solution was designed to help QueBox, a global retail and distribution brand, gain deeper insights into product profitability and drive smarter strategic decisions. From raw CSV files to structured SQL data warehouse and an interactive Power BI dashboard, this solution follows modern data architecture and analytics best practices to support real-time, data-driven decision-making.
QueBox, a global retail and distribution brand, has experienced steady revenue growth in recent years; however, actual profit margins have not met expectations. The leadership suspects that certain products and product categories contribute more significantly to profitability than others. Ufortunately, there’s currently no clear insight into which products are driving profit and which are dragging it down. Without this insight, QueBox risks continuously investing time, money, and resources into low-performing products and categories that deliver minimal impact on overall business performance.
To enable QueBox executives, particularly the Chief Revenue Officer with real-time insights into product performance and profitability across regions, categories, and time.
Data was sourced from two systems: CRM and ERP, each containing 3 CSV files. CRM data covered customers, products, and sales information WHILE ERP on the other hand provided extended info like customer demographics, sales location, and product categorization.
Using SQL, a layered ETL pipeline was built, with all workflow documented in the scripts folder of this repository. The process began by ingesting raw files into the Bronze layer, which stores the unprocessed data in its original form. This was then followed by the Silver layer, where key transformation steps were applied, including:
- Standardization of gender and country columns to ensure uniformity
- Validation of the arithmetic relationship between sales amount, price, and quantity sold
- Removal of duplicates across both primary and foreign keys using window functions
- and all other forms of data harmonization to resolve inconsistencies between the CRM and ERP sources.
The final step of the pipeline involved aggregating and integrating into star schema model the data within the Gold layer for analytical use
The final model consists of:
-
gold.fact_sales – transactional data
-
gold.dim_products – product details
-
gold.dim_customers – customer demographics
ETL scripts are in the scripts folder. Architecture diagrams (data flow, integration logic, star schema) are located in documents
To bring the data to life and support data-driven decision-making, an interactive Power BI dashboard was developed using the cleaned, business-ready dataset from the data warehouse. The dashboard is structured across two pages.
Page 1: Shows gross sales, total profit, and profit margin (filterable by year), monthly profit trends, top-performing countries, and breakdowns by product categories.
Page 2: A product-level view showing each product’s total sales, profit, margin, and classification as high, moderate, or low-margin.
This design empowers business stakeholders to monitor performance, identify trends, and take action—all in real time.
Here’s what the dashboard reveals at a glance:
- 📈 High-margin vs. low-performing products, grouped by product category
- 📅 Monthly and Yearly Profit trends tracked over time
- 🌍Profit contributions by country, revealing top-performing regions
- 🧾 Real-time product profitability breakdown
⚠️ Inactive products identified, those occupying space, costing QueBox but not driving value
🚀 Actionable Steps for Stakeholders With these insights, business leaders can:
- 💰 Reallocate marketing budgets to focus on high-margin products
- 🧹 Phase out or reprice underperforming inventory to protect profit
- 📦 Optimize inventory decisions based on profit classification
- 🔍 Spot emerging profit trends early and act before they become problems
- SQL Server
- Power BI