Pricing & Revenue Optimization Diagnostic

A consulting-style data diagnostic: given a retail product portfolio (9 categories, 52 SKUs, 15 months of monthly price/quantity/competitor data), identify where the business can adjust pricing to improve revenue without hurting demand.

Method

1. Data cleaning & feature engineering (scripts/01_clean_and_feature.py) — parsed dates, computed competitor price gap, log-transformed price/quantity.
2. Price elasticity estimation (scripts/02_elasticity_analysis.py) — for each category, ran a log-log OLS regression of quantity on price with product-level fixed effects, to isolate the true price effect from product popularity.
3. Scenario modeling (scripts/03_scenario_analysis.py) — quantified the revenue impact of realigning price to the competitor benchmark, for categories with statistically significant elasticity.

Key finding

No category showed a confident opportunity to raise prices — every statistically significant category was highly price-elastic. The real opportunity was Watches & Gifts, priced 12.8% above competitors despite being elastic: realigning its price down is projected to lift category revenue ~21%.

Files

• clean_pricing_data.csv — cleaned transaction-level dataset
• elasticity_by_category.csv — elasticity estimates and significance by category
• pricing_scenario_impact.csv — projected revenue impact of price realignment
• Pricing_Optimization_Memo.docx — one-page executive recommendation memo
• scripts/ — full analysis pipeline in Python (pandas, statsmodels)

Data source

Adapted from a public retail pricing dataset (Olist e-commerce, Brazil), originally sourced via Kaggle.

Tools

Python (pandas, statsmodels) for analysis · Power BI for visualization