Home›Query Library›Amazon

Which of your Amazon SKUs are being returned the most?

A high return rate on a single SKU can silently erode your margins, damage your seller metrics, and attract negative reviews. This query surfaces it automatically after every data sync.

📊 Amazon Seller Central👥 Amazon FBA sellers🔤 Plain-English → SQL

The question

Show me return rate by SKU for the last 90 days, ordered by highest return rate first

↓

The SQL Taptic generates

Amazon Return Rate by SKU — generated by Taptic Data AI

SELECT
  i.sku,
  MAX(i.product_name)                                      AS product_name,
  COUNT(DISTINCT o.amazon_order_id)                        AS total_orders,
  COUNT(DISTINCT r.return_quantity)                        AS total_returns,
  ROUND(
    COUNT(DISTINCT r.return_quantity)::numeric * 100.0
    / NULLIF(COUNT(DISTINCT o.amazon_order_id), 0), 2
  )                                                        AS return_rate_pct
FROM amazon_order_items i
JOIN amazon_orders o
  ON o.amazon_order_id = i.amazon_order_id
LEFT JOIN amazon_returns r
  ON r.sku = i.sku
WHERE o.order_status NOT IN ('Canceled', 'Cancelled')
  AND i.item_price IS NOT NULL
  AND o.purchase_date::date >= CURRENT_DATE - INTERVAL '90 days'
GROUP BY i.sku
HAVING COUNT(DISTINCT o.amazon_order_id) > 10
ORDER BY return_rate_pct DESC
LIMIT 25

Schema-aware SQL generated from plain English21 lines

This query was generated by Taptic Data from plain English against a real Amazon Seller Central schema. In Taptic, you type the question — the AI writes the SQL, runs it, and returns the result. You can edit the SQL, ask for explanations, and save it as a refreshable report.

How this query works

Breaking it down line by line

Joins order_items to orders to get SKU-level order counts, filtered to non-cancelled orders with valid prices
Left-joins the returns table so SKUs with zero returns still appear (with a 0% rate)
The HAVING clause filters out SKUs with fewer than 10 orders — low-volume SKUs have noisy return rates
NULLIF prevents division-by-zero on SKUs with no orders in the period
Orders by return rate descending so your worst performers appear first

What it returns

Result description

Returns a ranked table of SKUs with their order count, return count, and return rate percentage — filtered to the last 90 days and sorted worst-first.

Why it matters

The business impact

Amazon's A9 algorithm factors return rate into seller health metrics. A return rate above 8% on a product can trigger listing suppression. Catching it at 4% gives you time to investigate and fix the root cause — product quality, misleading images, or inaccurate descriptions — before it escalates.

Generate this automatically

Skip the SQL. Ask the question.

In Taptic Data, you type "Show me return rate by SKU for the last 90 days, ordered by ..." and this SQL runs automatically against your real Amazon Seller Central data.

Try Taptic Free — $29.99/mo

No credit card required. Connect your data source in under 5 minutes.