This project represents a comprehensive Business Intelligence solution designed to analyze and optimize the operations of Instacart. By processing over 32 million transaction records, I have bridged the gap between raw Big Data and strategic decision-making.
As a Business Analyst, I didn't just build dashboards; I managed the entire Software Development Life Cycle (SDLC) using Jira, engineered an optimized data pipeline, and delivered actionable insights to drive customer retention and sales growth.
🔗 Data Source: Instacart Market Basket Analysis Dataset (Kaggle)
🔗 Power BI Interactive File: Download the .pbix file from Google Drive
- Strategic Planning: Jira (Agile/Kanban, ITSM, User Stories, Acceptance Criteria).
- Data Engineering: Python (Batch-processing methodology for 32M+ rows).
- Database Management: Microsoft SQL Server (OBT - One Big Table methodology, Feature Engineering).
- Business Intelligence: Power BI (DAX, Executive Dashboard Design, UI/UX Optimization).
To ensure alignment with business goals, the project was managed under a strict Agile framework.
- Requirement Gathering: Defined professional User Stories and Acceptance Criteria.
- Workflow Management: Implemented a complex enterprise workflow (To Do -> In Progress -> In Review -> UAT -> Done).
- ITSM Simulation: Managed critical performance bugs and feature requests using priority-based ticketing.
Caption: Professional Kanban Board showcasing task dependencies, labels, and team assignment.
Handling a dataset of this magnitude required a high-performance architectural approach:
- Python ETL: Developed a script to migrate data in chunks (50k rows/chunk) to ensure system stability.
- OBT Methodology: Engineered an OBT (One Big Table) view in SQL Server to optimize report performance.
- Feature Engineering: Extracted
Day_Type(Weekend vs Weekday) andTime_of_Day(Peak Hours) to analyze consumer behavior patterns.
The final deliverable is an executive-level dashboard focused on Customer Behavior and Product Performance.
Key Business Metrics (KPIs):
- 3.21 Million total orders processed.
- 11.10 Days - Average customer return cycle (The primary target for retention campaigns).
- Top Department: 'Produce' leading with 9.5M units sold.
Caption: Executive Dashboard utilizing advanced DAX for real-time business tracking.
Based on the data-driven evidence, I recommend the following:
- Inventory Priority: Heavily prioritize logistics for 'Produce' and 'Dairy' departments (Top Volume Drivers).
- Churn Prevention: Trigger automated marketing campaigns on Day 9 post-purchase to shorten the 11.1-day return cycle.
- Operational Efficiency: Shift marketing focus to "Off-Peak" hours to balance warehouse pressure during peak windows (10 AM - 3 PM).
Hassan Ali Business Information Systems (BIS) | Data Analyst | Project Manager