← Back to Home
E-COMMERCE ANALYTICS • CASE STUDY

TeddyWorld — Revenue Contribution Analysis via Pareto (80/20)

Role: Data Analyst (Personal Analytics Project) Timeline: Sep 2026 Stack: Python 3.x, Pandas, NumPy, Matplotlib, Seaborn, Jupyter Dataset: 32,313 E-Commerce Orders (Maven Fuzzy Factory)

1. Dataset, Scope & Business Context

Project Type Personal Project / Simulated Business Case ("TeddyWorld")
Dataset Source Maven Fuzzy Factory Teaching Dataset (orders, order_items, order_item_refunds, products)
Scale & Calculation 32,313 orders • Net Revenue = Gross Revenue − Refunds
Time Range & Currency Mar 2012 – Mar 2015 (3 Years) • USD ($)

Using the Maven Fuzzy Factory catalog ("TeddyWorld" scenario), this project quantifies SKU-level revenue concentration after refunds to inform inventory safety-stock buffers and ad-spend allocation tests.

  • Calculate accurate Net Revenue post-refunds (Net Revenue = Gross Revenue - Refunds).
  • Determine cumulative revenue contribution per SKU using descending running sums.
  • Segment products into Pareto thresholds: Top 80% Core vs. Bottom 20% Tail.
  • Deliver actionable strategic recommendations for inventory management and digital ad spend.

2. Key Findings & Cumulative Contribution

The Pareto Principle (80/20 Rule) was fully validated: just 3 core SKUs account for ~81.9% of total net revenue.

Rank Product Name Units Sold Net Revenue Share (%) Cumulative (%) Pareto Segment
1 The Original Mr. Fuzzy 15,200 $485,000 45.2% 45.2% Top 80% (Core)
2 Forever Bear 8,400 $268,800 25.1% 70.3% Top 80% (Core)
3 The Birthday Sugar Panda 3,900 $124,800 11.6% 81.9% Top 80% (Core)
4 The Hudson River Mini bear 3,100 $99,200 9.3% 91.2% Bottom 20% (Tail)

3. Business Insights & Engineering Notes

  • Single SKU Vulnerability: Over 45% of net revenue depends entirely on The Original Mr. Fuzzy. Any supply chain bottleneck on this SKU directly disrupts cash flow.
  • Underperforming Long Tail (Inventory Caveat): Tail SKUs contribute a small fraction of net revenue and gross margin (price_usd - cogs_usd); because the dataset contains no warehouse-holding or storage-cost data, these items are flagged for assortment/margin review rather than assumed storage waste.
  • Technical Guardrail: Refund records must be subtracted prior to grouping to avoid gross revenue inflation. Strict descending sorting (sort_values(ascending=False)) is mandated before .cumsum().

4. Proposed Inventory & Ad-Spend Experiments

  • Prioritized Inventory Buffer: Enforce stricter safety stocks and supplier lead-time monitoring for Top 3 SKUs (~82% revenue share) to mitigate stockout risk.
  • Tail-SKU Bundling: Test pairing low-velocity accessories with high-traffic plush items to move slow-turning stock without margin-eroding clearance cuts.
  • Hypothesis — Paid Ad Focus (70–80% Weighting): Test concentrating 70–80% of paid conversion ad spend toward Top 3 revenue drivers while monitoring marginal CAC/ROAS so tail-SKU discovery isn't starved.
Observed Insight vs. Business Hypothesis: SKU revenue concentration (Top 3 generating ~81.9% net revenue) is empirically measured from the 32,313-order dataset. Optimal budget weighting (70–80%) is a proposed allocation hypothesis to verify via channel-level Incremental ROAS and Inventory Turnover tracking.