Re-audited the raw files and restored 7,795 rows incorrectly removed in the earlier analysis
VERIFIED CASE STUDY · E-COMMERCE OPERATIONS
Amazon.in Apparel Operations & Stock Review
A rebuilt analysis of one Amazon.in apparel seller, connecting 128,975 order lines with stock data to explain falling revenue, fulfilment cancellations and inventory risk.
PythonSQLPower BIDAX
01 · CONTEXT
From a broad concern to testable questions.
Business problem
Should the seller ship orders directly or use Amazon fulfilment—and is working capital tied up in the wrong products?
Questions explored
- Why did net revenue per day fall between April and June?
- Are seller-shipped orders cancelled more often than Amazon-fulfilled orders?
- Which best-selling products are at risk from thin stock?
- Where is excess inventory absorbing working capital?
Dataset
Amazon.in order report with 128,975 order lines from 31 March to 29 June 2022, joined to a separate stock report covering 9,170 SKUs.
02 · ANALYSIS
A transparent path from preparation to interpretation.
Standardised 13 order statuses and merged 69 state spellings into 36 states and territories
Joined sales to stock by SKU, achieving a 94% match, and calculated weeks of stock cover
Used SQL, a two-proportion z-test and Power BI to turn operational risks into quantified actions
03 · DASHBOARD
POWER BI REPORT GALLERY3 decision views
Select any view to inspect the dashboard at full size.
04 · FINDINGS
What the analysis surfaced
01Seller-shipped orders were cancelled at 17.5% versus 12.8% for Amazon-fulfilled orders; the 4.7-point gap held across every major category and was statistically significant (p < 0.001).
02Net revenue per day fell 13% from April to June as units per day declined 18%; Sets lost ₹114K per day while Western Dresses gained ₹33K.
03The top 10% of products generated 54% of revenue, yet 247 best-selling SKUs were out of stock or had under two weeks of cover.
04Seventy percent of stock units sat in products with more than a year of cover or no Amazon sales, while 25% of revenue depended on thin stock.
Recommended next moves
01Reorder the 247 at-risk best-sellers first, prioritising the highest-revenue Western Dress styles.
02Move more volume to Amazon fulfilment or correct the seller dispatch process; matching Amazon's cancellation rate could save about 1,840 orders and ₹1.2M per quarter.
03Clear slow stock through markdowns, bundles or other channels and redirect working capital toward proven products.
04Report net revenue consistently, record promotions at order time and capture returns for Amazon-fulfilled orders.
LIMITATIONS
What the analysis does not establish.
- The analysis covers thirteen weeks from one seller, so the April–June decline may be partly seasonal.
- The stock report is undated and was treated as an end-of-period snapshot, making stock-cover estimates indicative.
- Stock may sell through other channels, and the source data does not identify who initiated a cancellation.
- The fulfilment gap is an association rather than a causal experiment, and cost data is unavailable for profit analysis.
NEXT CASE STUDY
Continue exploring