Welcome to the code behind the chips! This repository houses the analytics and visualizations used to optimize the chip category for a major grocery retailer. Buckle up for a data-driven adventure filled with customer segmentation, statistical insights, and snack-sized recommendations.
To understand chip purchasing behavior, identify key customer segments, and provide actionable recommendations for boosting chip sales while keeping snackers happy.
- Data Wrangling: Cleaning, preprocessing, and exploring transaction and customer data to unlock its hidden potential.
- Customer Segmentation: Identify top performance customer segments based on purchase behavior and preferences.
- Control Store Selection: Employing correlation analysis and magnitude distance to find the perfect control stores for testing the impact of new chip layouts, ensuring fair and statistically sound comparisons.
- Impact Assessment: Comparing sales and customer data of trial and control stores to measure the effectiveness of the new layout. Did the chips fly off the shelves or flop?
- Visualization: Translating complex data into engaging charts and graphs to tell the story of customer behavior and layout impact.
- Actionable Insights: Delivering recommendations tailored to specific customer segments and backed by data-driven evidence.
Python (pandas, numpy, scikit-learn) Presentation (Powerpoint)
- Statistical hypothesis testing to check the insights before recommendation
- Correlation analysis and magnitude distance for control store selection
- Statistical hypothesis testing for assessing layout impact
Explore the code and scripts to understand the analysis process. Fork the repository and build upon the findings to further optimize the chip category. Discuss and share your own insights in the issues section. Let's keep the chip love flowing! By combining data analysis, clear communication, and actionable recommendations, we can unlock the secrets of snacking success and keep shelves stocked with smiles.