Declining Fraud Rates Don’t Mean Declining Fraud Risk

Why It Matters

Relying solely on lower attack rates can mask rising high‑value fraud, threatening revenue, reputation, and long‑term growth.

Key Takeaways

• Fraud attack rates fell 14% YoY, but chargebacks rose 56%.
• 21% of consumers experienced account takeover in past year.
• Manual review volume down 17% while high‑value fraud increases.
• Subscription models amplify chargeback exposure despite low attack rates.
• 52% of shoppers would abandon platform after a fraud incident.

Pulse Analysis

The headline drop in payment‑fraud attack rates is tempting to celebrate, but the underlying data tells a more nuanced story. AI‑powered detection tools have trimmed the sheer volume of low‑value attempts, allowing merchants to automate many decisions and reduce friction for legitimate shoppers. However, fraudsters are adapting, focusing on accounts with stored credentials and high balances where a single breach yields outsized returns. This shift from quantity to quality means that traditional metrics—such as attack‑rate percentages—no longer capture the true risk landscape, and organizations that ignore the rise in chargebacks risk under‑investing in critical defenses.

Operational efficiency gains, evidenced by a 17% dip in manual review volume, are real and beneficial. Yet when that efficiency coincides with a 56% jump in chargeback rates, it signals that decision thresholds may be tuned for speed rather than depth. High‑value, sophisticated attacks often require human judgment to identify subtle behavioural anomalies that automated models miss. Companies with subscription‑based or credential‑heavy models—online gaming, SaaS, and digital media—face disproportionate exposure because the same friction‑reducing features that boost conversion also lower the barrier for account takeover. Segment‑specific benchmarking, rather than industry‑wide averages, is essential to allocate review resources where they matter most.

Beyond the financial hit, fraud erodes consumer trust. Over half of shoppers say they would abandon a platform after a fraud incident, and nearly three‑quarters abandon purchases over security concerns. This translates into lost lifetime value and reputational damage that far exceeds a single chargeback. To stay ahead, fraud teams should track longitudinal account risk, align chargeback trends with attack‑rate data, and ensure that automation frees analysts for complex investigations rather than masking emerging threats. By measuring the right signals, merchants can protect both their bottom line and brand equity.