Mohammed Ashraf Hassan

The question

A retail business could see its revenue but not what was quietly draining it — cancelled orders, slow shipping, and discount-heavy sales reps.

What I built

A three-page Power BI report over a cleaned RetailSales dataset, modelled as a star schema joined on surrogate keys: an Executive page for revenue and margin, an Operations page for fulfilment health, and a Sales Force page for rep and product performance. All pages share date, channel, brand, and category filters.

What it revealed

• 29.62% net profit margin on 4.34M in sales, with margin dipping to 26.15% in Q3 — a seasonal squeeze worth planning around.
• 14.80% of orders were cancelled, equal to 625.61K in lost revenue: the single biggest recoverable number in the business.
• Average shipping lag of 3.92 days showed no strong link to cancellations by city, ruling out delivery speed as the main cause.
• Product-level margin flags exposed items selling in volume while earning below-target margin, and discount rates varying by rep up to 9%.

The question

With 300K flight records, which airlines, cabin classes, and booking windows actually drive ticket price — and where does price stop being about distance?

What I built

A three-page Power BI report on a star-schema model with airline, city, class, departure-time, and arrival-time dimensions. Pages cover an executive summary, a pricing and booking deep dive with correlation analysis, and a route-level view.

What it revealed

• Business class is 11.12% of tickets sold but a far higher share of value, with average business fares several times economy.
• Duration barely predicts price: the correlation plot flattens after the first few hours, so scheduling — not distance — sets the fare.
• Fares stay flat until roughly 10 days before departure, then spike sharply — a clear last-minute pricing window.
• Vistara and Air India dominate both flight volume and premium pricing; the six busiest city pairs cluster around Chennai, Bangalore, and Kolkata.

The question

Across four years of orders, which segments, categories, and states actually carry profit — and is the trend improving or just noisier?

What I built

A single-page executive sales dashboard over a star-schema model with customer, product, geography, date, and ship-mode dimensions, filterable by year, region, and shipping method.

What it revealed

• 2.90M in sales produced 360.13K in profit, with quarterly profit growing from 4K to a 62K peak in Q3 2019.
• The Consumer segment drives 53% of sales (1.55M), while Home Office contributes only 17% — a clear targeting signal.
• Technology leads category sales, with Office Supplies close behind on volume but thinner returns.
• Sales concentrate heavily in California and New York, exposing a geographic dependency worth diversifying.