Food Delivery Analytics

Food Delivery Delay and Rating Analysis

This project analyzes 100,000 food delivery records to understand delay trends, customer ratings, refund behavior, and practical business actions.

Open Analytics Project Scope
End-to-end analysis includes data cleaning, KPI extraction, visualization, and baseline ML.
Delay Intelligence Customer Ratings Refund Risk

100,000

Orders Analyzed

29.54 min

Avg Delivery Time

3.24 / 5

Avg Service Rating

46.16%

Late Orders Share

45.94%

Refund Rate for Late Orders

Project Scope

  • Data cleaning and feature engineering from raw ecommerce delivery records.
  • Delay, rating, category, value, and refund behavior analysis.
  • Time-based analysis with 5-minute fallback because source time is minute-second formatted.
  • Baseline machine learning model (Linear Regression) to predict rating using delivery duration.

Key Findings

  • Platform with highest average delivery time: JioMart (29.63 min).
  • Category with highest delay: Grocery (29.58 min).
  • Category with lowest rating: Grocery (3.22).
  • Order value vs rating correlation is nearly zero (-0.0027).
  • Late vs on-time refund rate is almost identical in this dataset (1.00x).

Business Recommendations

  • If delivery exceeds 30 minutes, trigger apology coupon workflow.
  • Optimize high-delay categories with better dispatch and packaging flow.
  • Increase delivery partner availability during peak delay windows.
  • Use refund-risk flags for faster support intervention.
  • Use model output as an early signal for customer satisfaction drop.
Open Chart Analysis to view all generated charts with explanations.

Installation

  1. Clone or download the repository.
  2. Create and activate a Python virtual environment.
  3. Install dependencies with pip install -r requirements.txt.
  4. Run python food_delivery_analysis.py to regenerate the cleaned data, summary, and charts.
  5. Open templates/index.html locally, or use GitHub Pages deployment from this repository.