It is a Exploratory Data Analysis Challenge
📊 The EDA Mission Briefing This repository contains our team's submission for The Manhattan Project's Exploratory Data Analysis (EDA) Group Challenge. Our mission was to analyze the provided dataset to uncover actionable insights for OmniMart Retailers, focusing on customer behavior, product performance, and operational efficiency.
Team Members:-
Nayanika Chatterjee - Lead Analyst, Data Cleaner
Ankan Sadhu - Visualization Specialist, Insight Writer
Asmit Dey - Presentation Designer
🚀 Install Dependencies and rerun the notebook :
pandas numpy seaborn matplotlib plotly scipy tabulate networkx community
🔍 Key Insights & Findings:
Our analysis of the OmniMart dataset led to the following key insights:
Overall Business Performance: Sales show strong seasonality — January peaks at $6.54M while February dips to $6.15M, and mid-week (Wed–Thu) drives the highest daily revenue.
Product & Category Performance: Electronics + Grocery dominate both volume and spend, with Electronics acting as the central “hub” product, strongly cross-linked to other categories like Books and Clothing. Electronics is the undisputed top category by purchase volume (70,944), followed by Grocery as a strong second (66,552).
Customer Insights: Churn risk is concentrated in at-risk loyalists (high frequency but inactive for 200+ days), signaling an urgent need for win-back campaigns and loyalty fixes.
For a more detailed breakdown, please refer to our full analysis in the Jupyter Notebook and the accompanying presentation.
📁 Repository Structure
README.md (This file)
retail_data.csv (The raw dataset used for the challenge. Its uploaded in zip file due to file upload size limitations.)
manhattan_analysis.ipynb (The main Jupyter Notebook with all our code and analysis)
manhattan_presentation.pdf (A summary of our findings and recommendations)