Instacart Basket Analysis Portfolio

This repository contains Python scripts and Jupyter notebooks for the Instacart Grocery Basket Analysis project. The project focuses on customer segmentation, sales trends, and purchasing behavior using data analytics. The project covers data cleaning, exploration, visualization, and reporting using Python.

Repository Structure

• 01 Project Management: Includes project documentation, task breakdowns, and requirements.
• 03 Scripts: Contains Python scripts used for data analysis and visualization.
• 04 Analysis: Includes Jupyter Notebooks and reports detailing insights from the data.
• 05 Sent to Client: Final reports, presentations, or deliverables shared with stakeholders.

---

Project Overview

This project analyzes over 3 million grocery orders from Instacart’s open dataset to uncover patterns in customer purchasing behavior. The objective is to provide actionable insights into product demand, reordering habits, and user segmentation to support strategic decision-making in retail operations and marketing.

Developed as part of the CareerFoundry Advanced Data Analytics Program, the project demonstrates a full analytical workflow—from problem scoping and data preparation to reporting and stakeholder delivery.

---

Data Sources

The primary data source used for this analysis is the Instacart Market Basket Analysis dataset, which includes:

• orders.csv – Metadata on individual orders, including user ID and order timing
• products.csv – Product-level information
• order_products__prior.csv – Product details for all previous orders
• order_products__train.csv – Product details for most recent orders
• aisles.csv – Aisle-level categorization
• departments.csv – Department-level categorization

All datasets are available publicly via Kaggle.

---

Methodology

The project followed a structured, task-driven workflow:

1. Project Management

   • Defined objectives, stakeholders, deliverables, and timeline
   • Documented planning materials in /01 Project Management/

2. Data Cleaning and Preprocessing

   • Loaded and merged datasets using pandas
   • Checked data consistency, handled missing values, optimized memory usage
   • Conducted exploratory data analysis (EDA) using matplotlib and seaborn

3. Customer Segmentation and Feature Engineering

   • Analyzed ordering frequency, basket size, and reorder behavior
   • Generated new variables to classify customers by loyalty and habits

4. Data Visualization and Interpretation

   • Created clear visual summaries of trends, patterns, and behaviors
   • Built a Tableau dashboard for stakeholder-friendly presentation

5. Final Reporting

   • Prepared a summary slide deck and business report
   • Delivered findings and recommendations to simulated stakeholders

---

Findings

• Order Timing

  • Most orders are placed on weekends, especially Saturday and Sunday mornings.
  • Peak order hours are between 9 a.m. and 12 p.m.

• Product Trends

  • Organic products and dairy items are most frequently reordered.
  • A small number of products dominate order frequency.

• Customer Behavior

  • Over 60% of products are reordered.
  • Customers can be segmented into:
    • Frequent Repeaters: Consistent, large baskets with high reorder rates
    • Occasional Buyers: Infrequent orders with more variation
    • Trial Users: Minimal engagement or single-order behavior

---

Conclusion

This project highlights the power of behavioral data in online grocery retail. By analyzing reorder habits and segmentation profiles, businesses can:

• Improve inventory forecasting
• Tailor marketing campaigns to different customer segments
• Promote high-retention products with targeted offers