⏱ 4-minute read
Most teams manage returns purely as an ops problem: reverse logistics, refund volume, support load. It shows up on a report, gets treated as a cost of doing business, and never goes any further.
That's a mistake. Return behavior is a customer segment. And when growth teams run blanket promos without accounting for it, you're not just discounting, you're subsidizing the customers who hurt you most.
The Executive Shift: Stop Managing Returns as an Average
If you're an executive looking at your business's average return rate, you're looking at the wrong number. Returns aren't evenly distributed, they're concentrated behavior.
So run a simple exercise: pull your top 20% of returners and look at what you've spent marketing to them in the last 120 days. Discounts triggered. Retargeting served. Then look at their net contribution after returns and fulfillment costs.
Returns live in operations but promos live in marketing, and too often, they don’t speak. That separation is why this persists. Marketing drives conversion. Finance flags margin erosion. Ops absorbs the volume. Nobody connects the behavior to the incentive that caused it.
When you segment by return history and cross-reference with promo response, the cause and effect becomes impossible to ignore. A chronic returner who converts on your next promo isn't a win, it's a discount, two shipping legs, a refund, and a support ticket.
Once you accept that, the strategy becomes obvious: stop spending promotional margin on the customers who destroy it.
Five Principles To Change How You Think About High Returners
Return behavior is a customer segment, not an ops metric. It belongs next to LTV, CM1, and repeat rate, because it's a direct input into all three. Keeping it in operations keeps it invisible to marketing.
Chronic returners turn discounts into negative CM1. A discount is painful when the customer keeps the product. When they're likely to return it, you're not driving revenue, you're buying a refund with extra steps.
Promos don't just drive orders. They shape your database. Every discount email is a training signal. Keep promoting to return-heavy customers and you don't just lose margin today, you manufacture a worse customer mix tomorrow.
You can't retention-market your way out of structural return behavior. Some customers return because of a one-off issue. A chronic returner is telling you: "I shop like this." No subject line changes that. The lever is exposure and incentives, reduce both.
Protect your clean customers from the mess. Clean customers don't need constant vouchers. When you spam them with discounts, you're teaching them to wait, and eroding the segment that actually funds your growth.
What To Do This Week
Segment by return behavior. Use a rolling 12-month window with a minimum order count. Three buckets: Clean (rarely returns), Mixed (occasional), Chronic (consistently high return share or net-negative after returns). Keep the threshold specific to your business.
Suppress chronic returners from all discount activity. Remove them from sitewide promo emails, winback automations, and offer-based retargeting. If you still want to reach them, shift the message to sizing help, product education, or store credit. Stop paying them to repeat the same loss pattern.
Measure promo profit, not promo revenue. Add CM1 per promo-driven order, return rate by segment, and refund and shipping drag to your promo reporting. Revenue tells a story. Contribution margin tells the truth.
Fix the mixed segment separately. Better sizing guidance, clearer PDPs, smarter next-best-offer logic. This is where real behavioral improvement lives, don't conflate it with the chronic segment. They're different problems with different levers.
The Takeaway
Returns aren't a brand problem first. They're a portfolio problem. Segment return behavior, stop incentivizing the worst patterns, and you protect CM1, improve LTV, and build a customer file that actually compounds.
If you want a sanity check on your own chronic returner problem, reply “RETURNS” and I'll tell you what I'd look at in your data first.
