Project case study

Business Insights 360: AtliQ Hardware

Multi-view Power BI dashboard that turns 1.5M+ rows of sales, cost and forecast data into one source of truth for Finance, Sales, Marketing, Supply Chain and Executive teams.

Problem

The company relied on manual Excel reporting, with limited visibility across global markets and poor coordination between departments.

Approach

Bring sales, cost and forecast data into a star-schema model and build department-specific Power BI views. Use DAX measures and interactive navigation to support detailed analysis from a shared source of truth.

  • Star schema model with DAX measures for net sales, gross margin % and forecast accuracy
  • Drill-through pages, bookmarks, dynamic titles and custom tooltips
  • P&L breakdown, top/bottom customers and products, regional performance

Tools

Power BIDAXPower QuerySQLStar SchemaDAX Studio

Key Insights

  • Net sales grew 353.5% to 3.74B and gross margin improved to 38.08% (+1.59 pts), but net profit fell to -522.43M (-13.98%) from -54.65M (-6.63%) last year.
  • Operating expense of 1,945M equals 52% of net sales and grew +447.5% vs +353.5% for sales, wiping out the entire 1,423M gross margin.
  • Pre- and post-invoice deductions take 49.3% of gross sales (3,634M of 7,370M), so only about half of gross sales becomes net sales.
  • India is the largest market (945M, 25.3% of sales) but generates 41.6% of the total loss (-217M, -23.0% net profit %), with gross margin of 35.75% vs 44.48% in the USA.
  • Forecast accuracy is 81.17%, but Peripherals (68.17%) and Storage (71.5%) are flagged out-of-stock risk, and Peripherals alone drives 92% of the net forecast error. Accessories and Desktop are flagged excess inventory.

Recommendations

  • Cap operating expense growth below revenue growth, and break the 1,945M into its main cost lines to find the biggest drivers.
  • Tighten discount policy: recovering just 1 point of gross sales (about 74M, assuming volume holds) would cut the net loss by roughly 14%.
  • Fix India's pricing, discounts and cost-to-serve before pushing more volume there.
  • Raise forecast buffers for Peripherals and Storage, reduce stock for Accessories and Desktop, and track forecast accuracy at customer level.

Links