Fraud rate in DTC averages 0.5 to 2 percent of orders, costing chargebacks plus product loss plus shipping. Aggressive blocking hits real customers, gentle blocking lets fraud through. This SOP finds the right balance.
Every DTC brand processing payment cards. Most acute above $1M ARR where fraud volume becomes worth dedicated process.
A single fraud-flagged real customer becomes a public complaint. A single missed fraudulent order becomes a chargeback plus product loss. Both are bad. The middle path requires deliberate process.
Free tier blocks obvious patterns. Adequate baseline.
Orders that aren't obvious fraud but show 2+ risk signals. Manual review before shipment.
Email domain mismatch with billing name, shipping address different from billing in different country, multiple cards tried, high-value first order from new customer.
CVV failed = automatic cancel or manual review. Never ship without CVV pass for cards.
5+ orders same email same day = block. Genuine customers don't do this.
3D Secure shifts liability to issuer. Enable for medium-risk countries or high-value first orders.
Why blocked, why approved on review. Builds team learning.
Each chargeback teaches you a pattern. Update detection rules accordingly.
Confirmed-fraud emails, addresses, phones. Block automatically on future attempts.
Record your screen while running this procedure. ReccordSOP captures every step with screenshots, then alerts you when the actual process drifts from documentation.
Try ReccordSOP free0.5 to 2 percent of orders. Above 3 percent and you have a problem. Below 0.5 might mean you're too aggressive and blocking real customers.
Above $5M ARR, yes. NoFraud, Signifyd worth the cost. Below, Shopify built-in plus manual review is fine.
Process within 4 hours during business hours. Customers expect orders to ship. Long fraud review delays drive cancellations.
I built ReccordSOP after watching too many DTC ops teams lose months to undocumented workflows. These SOPs are battle-tested with Shopify operators running $1M to $50M brands.
Last reviewed June 1, 2026