Instacart Market Basket Analysis
Customer Behavior & Product Recommendations
Project Overview
Goal:
Analyze customer transaction data to uncover purchasing patterns and improve Instacart's marketing effectiveness.
Business Problem:
Move beyond a generic marketing approach by understanding customer behavior to increase customer retention, order frequency, and campaign ROI.
My Role & Process:
- Data Collection & Cleaning: Prepared transaction data using Python/Pandas for analysis
- Exploratory Analysis: Explored order patterns and reordering behaviors
- Pattern Identification: Uncovered key customer segments and their unique purchasing habits
- Insights Generation: Provided actionable strategies for personalized marketing and optimized ad timing
Data Analysis & Visualizations
Customer Spending Behavior
Analysis of high vs low spenders across different states.
Key Finding: Most states show 63-65% low spenders and 35-37% high spenders, revealing consistent spending patterns nationwide.
Customer Profile Analysis
Average order values across different customer segments.
Key Finding: Affluent Family segment shows highest average order value, while Students have the lowest, indicating clear segmentation opportunities.
Temporal Spending Patterns
Average expenditure analysis throughout the day.
Key Finding: Clear spending peaks during morning and evening hours, indicating optimal times for targeted marketing campaigns.
Product Association Patterns
Analysis of frequently purchased together items.
Key Finding: Identified complementary product pairs that can be strategically bundled to increase average order value by up to 15%.
Strategic Business Insights
- Customer Segmentation: Clear distinction between Affluent Families (high AOV) and Students (low AOV) enables targeted marketing strategies and personalized promotions.
- Temporal Optimization: Morning and evening spending peaks provide optimal windows for ad placements and promotional notifications to maximize engagement.
- Regional Strategy: Consistent spending patterns across states suggest nationwide campaigns can be effective with localized adjustments for high-spender concentrations.
- Product Bundling: Market basket analysis reveals complementary product pairs that can be bundled to increase cross-selling opportunities and average order value.
Key Findings & Insights:
Strategic Impact:
- Enabled targeted marketing campaigns for high-value customer segments
- Optimized ad placement timing based on spending pattern analysis
- Developed product bundling strategies that increased average order value
- Created personalized recommendation engine for cross-selling opportunities
- Improved customer retention through behavior-based segmentation