Skip to content

README.md

Latest commit

 

History

History
74 lines (50 loc) · 3.46 KB

README.md

File metadata and controls

74 lines (50 loc) · 3.46 KB
Raw

Amazon India — Reducing Delivery Delays in Tier-2 & Tier-3 Cities

A Data-Driven Operations & Logistics Analytics Case Study Prepared by Yatharth · Business Analyst Portfolio Project

Cover

📌 Overview

Amazon India's on-time delivery (OTD) rate in Tier-2/3 cities fell from 89.2% → 79.1% over 12 months while Tier-1 metros held steady above 96%. This case study walks through the full analyst workflow — problem framing, root-cause analysis, pilot design, results, and a financial case for a national rollout — the way it would be presented to an Operations leadership team.

Metric Result
OTD decline diagnosed (Tier-2/3, 12 mo) -10.1 pp
Cities in diagnostic pilot 6
OTD lift achieved from pilot fixes +11.9 pp
Estimated annualized value at national scale ~$42.0M
Estimated payback period < 4 months

🧠 What this case study demonstrates

  • Problem framing — translating a vague "delivery is slow" complaint into a measurable, segmented business question
  • Root cause analysis — Pareto analysis to isolate the 3 causes driving 76% of delays
  • Hypothesis testing — correlation analysis to check (and partially reject) the "it's just a staffing problem" theory
  • Experiment design — a 6-city, 90-day before/after pilot bundling three targeted fixes
  • Financial modeling — a full ROI build (capex, opex, avoided cost, retained revenue, payback period)
  • Executive storytelling — structuring findings the way a Business Analyst would present to VP-level stakeholders

📂 Repository Contents

├── Cover.png                          # Case study cover page
├── Amazon_Delivery_Case_Study.pdf     # Full case study report
├── data/
│   ├── otd_trend.csv                  # 12-month OTD trend, Tier-1 vs Tier-2/3
│   └── pilot_ab_results.csv           # Before/after results, 6-city pilot
├── images/                            # All charts used in the report (PNG)
└── README.md

🔍 Key Findings

  1. The gap is structural, not seasonal. Tier-2/3 OTD declined nearly monotonically across all 12 months while Tier-1 stayed flat — ruling out a one-off seasonal spike as the explanation.
  2. Three causes drive 76% of delays: last-mile handoff delays (34%), address/geocoding errors (24%), and rural hub capacity gaps (18%).
  3. It's concentrated, not diffuse. The worst-performing 20% of hubs account for 58% of all delayed shipments — making a targeted fix far more cost-effective than a blanket one.
  4. The pilot worked, consistently. All 6 pilot cities improved by 10.7–12.5 percentage points, with the lowest-performing city improving the most.

🛠️ Tools Used

  • Python (pandas, matplotlib) for data aggregation and visualization
  • WeasyPrint for report generation
  • Methodology mirrors a production stack of SQL (Redshift/Athena) + a BI tool (QuickSight/Tableau)

⚠️ Disclosure

This is an independent, self-initiated portfolio project built to demonstrate business analyst skills — problem framing, root-cause analysis, experiment design, and ROI modeling. It uses a synthetic dataset designed to reflect realistic patterns in Indian e-commerce logistics. It is not affiliated with, endorsed by, or based on any real Amazon internal data.

📫 Contact

Feel free to connect if you'd like to discuss this case study or my other analytics work.