📊 A/B Testing: Fast-Food Promotion Campaign Analysis

Overview

This project analyzes a controlled experiment run by a fast-food chain testing three marketing campaigns for a new menu item. Using A/B testing methodology, we identify which promotion generates the highest sales uplift.

Dataset

• Weekly sales for 4 weeks across multiple markets.
• Covariates: Market size, store age, and location.
• Dataset source.

Tools

• Python (pandas, numpy, statsmodels, scipy, matplotlib, seaborn, scikit-posthocs)
• Bootstrapping methods for robust inference
• ANOVA, regression adjustment, power analysis

Key Results

• Promotion 2 significantly outperformed both Promotion 1 and Promotion 3.
• Uplift: ~25% increase in sales over Promotion 1.
• Recommendation: Roll out Promotion 2 chain-wide.

Deliverables

• 📓 Analysis Notebook
• 📑 Executive Summary

Resume Bullet

“Performed A/B test analysis on fast-food marketing campaigns (design checks, ANOVA, regression adjustment, bootstrapped confidence intervals) and delivered actionable business recommendation.”