Pizza Sales and Performance Analytics

Problem Statement

The pizza chain had large volumes of transactional sales data but lacked a consolidated view of customer behavior, product performance, and sales patterns. Without clear insights, it was difficult for management to optimize inventory, identify best-selling products, and align marketing with peak demand periods.

Technical Solution

1. Step 1 – Architecture & Approach:
   • Built an integrated pipeline where SQL Server served as the data source and Python handled extraction, transformation, and visualization.
   • Structured queries captured time-based trends, product-level sales, and category-level performance.
2. Step 2 – Key Algorithms, Libraries & Highlights:
   • SQL: Advanced queries using joins, CTEs, subqueries, and aggregations to analyze revenue, customer preferences, and ingredient usage.
   • Python Libraries: pandas (ETL & transformations), matplotlib & seaborn (visualizations).
   • Highlights: Generated insights on peak order times, high-performing categories, and seasonal demand.
3. Step 3 – Deployment & Reporting:
   • Reports generated as Python-driven charts and Jupyter notebooks.
   • Final outputs included revenue trends, top products, and customer segmentation.
   • Findings shared with business teams to support marketing, operations, and menu optimization.

Results & Impact

• Identified top 10 best-selling pizzas contributing ~60% of total revenue.
• Pinpointed peak sales hours (12–2 PM & 6–9 PM) for targeted marketing.
• Improved decision-making for inventory planning and promotions, reducing waste and boosting margins.

Technologies Used