Returns data is the most honest feedback your customers will ever give you. Most stores treat it as a financial problem. The ones that treat it as a product intelligence system perform differently.
Your refund rate has a story to tell.
Not the headline number — although that matters. The story is in the distribution. Which products. Which reason codes. Which traffic sources. Which time windows. Which customer segments.
An overall refund rate of 3.2% tells you: something is wrong, or possibly nothing is wrong depending on your category. That is not a story. That is a data point.
A refund rate of 3.2% driven primarily by one SKU with an 8.4% refund rate, where the most common reason code is “not as described,” combined with a secondary cluster of returns from a specific traffic source where customers are converting at high rates but returning at 2× the store average — that is a story. And it is a story with clear, specific actions attached.
“A refund spike is never random. It is always pointing at something specific. Most stores find out too late — after the margin is gone and the reviews have landed.”
Three Things Your Returns Data Is Actually Telling You
Product description quality.
“Not as described” is the most common return reason code on most e-commerce stores, and it is the most actionable. When a SKU has a “not as described” refund rate above 5%, it means the product page is creating false expectations. The image does not match the product. The dimensions are wrong. The material is misrepresented. The compatibility claim is incorrect. Every one of those is a fixable catalog issue that will reduce returns and improve conversion simultaneously.
Traffic source quality.
When orders from a specific campaign or traffic source have a refund rate that is double the store average, the audience is mismatched to the product. The ad is reaching people who want something slightly different from what you sell. The conversion looks fine — they are clicking and buying. But the product experience doesn’t match the expectation set by the ad. This is a creative and targeting problem, not a product problem. But you only see it when you segment refund rates by traffic source.
Product quality signals.
A sudden spike in refunds for a SKU that previously had a low return rate — with reason codes like “product defective” or “quality below expectation” — is a manufacturing or supplier quality signal. Something changed in the supply chain. This is the kind of intelligence that a buyer needs immediately, not at the end-of-month reporting cycle.
The Chargeback Threshold
Refund rate monitoring has a hard threshold that makes it operationally urgent: the chargeback limit.
Visa’s chargeback threshold is 1% of transaction volume. Mastercard’s is 1%. Above these thresholds, payment processors impose additional monitoring, increased processing fees, and potentially account restrictions.
Most stores have no real-time visibility into their rolling chargeback rate. They find out they are approaching the threshold when their payment processor sends a letter.
At StoreSignals, the returns area monitors chargeback rate continuously against the 30-day rolling window. When the rate approaches 0.5% — the safe zone limit before entering warning territory — an alert fires. This is not a financial reporting function. It is an operational signal that says: something is systematically wrong in how orders are being processed or products are being represented, and you need to find it before your payment processing is at risk.
Turning Margin Loss Into Product Insight
The StoreSignals tagline for this area is “Turn margin loss into product insight.”
The framing matters because it repositions returns from a cost centre to an intelligence asset. Every return is an expensive piece of product feedback. The question is whether you read it.
A weekly returns intelligence report — top SKUs by refund rate, top reason codes, trend versus prior 4 weeks, revenue impact of the current rate — takes 30 minutes to review and typically surfaces 2–3 specific, fixable actions:
- Update the product description on this SKU to accurately reflect dimensions.
- Review this traffic source’s ad creative — the expectation it is setting does not match what customers receive.
- Contact the supplier for this SKU — the defect rate has tripled in the last 3 weeks.
Those actions have quantified revenue impact: a product description fix that reduces the refund rate on a high-volume SKU from 8% to 3% recovers meaningful margin. The analysis required to identify it takes less time than any other margin-improvement initiative available to the operations team.
The returns data is there. In every store. Telling a specific, actionable story.
Most stores just aren’t reading it.
