A self-built SQL + Tableau project simulating a Pakistani retail chain (PakMart) with realistic product categories, customer names, store locations, and promotional events — built entirely from scratch by M. Ammar.
Note: Sales data is randomly generated via a stored procedure (pakmart_sales.sql) and does not reflect real business insights. All other data (products, customers, stores, promotions) uses realistic Pakistani names and locations.
Live Dashboard: View on Tableau Public
| # | Business Question | Technique Used |
|---|---|---|
| 1 | What is the total revenue per promotion? | GROUP BY, SUM |
| 2 | What is the revenue per product category? | JOIN, GROUP BY |
| 3 | Which product ranks highest in each category? | Window Function — RANK() |
| 4 | What is the revenue by city during promotions? | Multi-table JOIN, CASE |
| 5 | What is the monthly revenue trend? | DATE functions, GROUP BY |
| 6 | What is the month-over-month revenue change? | Window Function — LAG() |
| 7 | Which store ranks highest per city? | Window Function — DENSE_RANK() |
| 8 | How does revenue compare: promotion vs non-promotion? | CASE, GROUP BY |
| 9 | Who are the top 5 customers by spending? | ROUND(), ORDER BY, LIMIT |
- Products, stores, promotions inserted manually via
PakMart_Retail_Schema&Queries.sql - 1000 customers imported via
customer.csv - 7000 sales rows generated using a stored procedure (
pakmart_sales.sql) which auto-assigns promotions based on Pakistani calendar events and calculates real prices from category tables
M. Ammar — BSCS Data Science, KSBL Karachi
GitHub • Tableau Public • LinkedIn