DAD-220 · Structured Database Environments

SQL Return Rate Analysis

This project uses SQL against the QuantigrationUpdates database to analyze product return patterns for a product manager. Using multi-table JOIN queries, it surfaces return frequency by state and return percentage by SKU — then interprets those findings, identifies data limitations, and communicates conclusions clearly to a non-technical stakeholder.

Database schema

The analysis draws on three related tables: Orders, RMA (return merchandise authorizations), and Collaborators. The queries use INNER JOIN to connect return records to geographic data and product identifiers.

Query 1: returns by state

SELECT Collaborators.State AS STATE,
       COUNT(*) AS RETURN_FREQUENCY
FROM Orders
INNER JOIN RMA
  ON Orders.OrderID = RMA.OrderID
INNER JOIN Collaborators
  ON Collaborators.CollaboratorID = Orders.CollaboratorID
GROUP BY STATE
ORDER BY RETURN_FREQUENCY DESC;

Findings

Massachusetts shows the highest return frequency in the dataset, followed by Arkansas. This geographic concentration warrants further investigation — it may reflect regional distribution patterns, a high-volume partner concentrated in those states, or a logistics or quality issue specific to that region.

Query 2: returns by product SKU

SELECT Orders.SKU AS SKU,
       COUNT(*) * 100 / (SELECT COUNT(*) FROM RMA) AS PERCENTAGE
FROM RMA
INNER JOIN Orders
  ON Orders.OrderID = RMA.OrderID
GROUP BY SKU
ORDER BY PERCENTAGE DESC;

Findings

Stakeholder summary

The data gives the product manager two actionable signals. First, Massachusetts is a geographic concentration point for returns and deserves a closer look at whether a single large partner or regional logistics issue is driving volume. Second, BAS-48-1C and the ENT product line are disproportionately represented in returns — this warrants a quality review, customer feedback analysis, and potentially a conversation with the relevant supplier.

Limitations and next steps

Skills demonstrated