We now have a full look at 2025 data from Sift’s Fraud Industry Benchmarking Resource (FIBR), giving fraud and risk teams a clearer picture of how attack patterns and merchant behavior evolved over the past year.
With a complete year of data, we can spot seasonal spikes and understand where fraud pressure eased, where costs quietly increased, and how operational decisions shaped outcomes just as much as fraudster tactics.
Here’s what the 2025 data reveals.
The State of Fraud in 2025
Payment Fraud: Reducing Risk Without Rejecting Revenue
Across industries, the average payment fraud attack rate was 3.15% in 2025, 37% lower than the MRC’s reported 5% order rejection rate. The gap between these benchmarks highlights a meaningful difference in how fraud is being detected and managed. Sift customers consistently operate below broader industry averages, reflecting more precise fraud prevention. This precision allows merchants to prevent fraud without running the risk of blocking legitimate transactions and sacrificing customer trust or revenue.
Manual review data reinforces this trend. While the MRC reports that merchants manually screen approximately 23% of orders, Sift’s average manual review rate in Q4 sat at just 2.7%, signaling a significant shift toward automation. For many businesses, this gap represents both an operational challenge and an opportunity to reduce cost and friction while maintaining control.
Despite continued growth in alternative payment methods, credit and debit cards still account for 72.7% of fraudulent payments, largely due to their widespread use. At the same time, financing, including buy now, pay later, showed the highest risk at a 6% fraud rate, emphasizing the need for tighter controls and better visibility as these payment methods continue to scale.
Account Takeover (ATO): The Quiet Persistent Threat
Account takeover attacks averaged 0.99% in 2025 and showed little seasonality compared to payment fraud. Instead, ATO spikes were more closely tied to external events such as large-scale credential leaks. Attackers increasingly use ATO to access stored payment details or personal data, often blending into higher traffic periods like Q4 to avoid detection.
At the same time, 2FA adoption averaged 8.2%, helping keep ATO rates relatively contained. While adoption continues to rise, the data reinforces a consistent takeaway: layered authentication is most effective when teams are prepared to respond quickly to sudden shifts in attacker behavior.
Chargebacks: The Cost Curve That Keeps Climbing
In 2025, the average general chargeback rate reached 0.22%, while fraudulent chargebacks remained lower at 0.11%, indicating that economic pressure and first-party misuse are driving a growing share of disputes. General chargeback rates increased 41% year over year, reinforcing that chargebacks remain a rising cost even when fraud itself is relatively stable.
Industry Fraud Patterns
Across verticals, one theme stood out in 2025: operational choices materially shape fraud outcomes.
In digital commerce, both payment fraud attack rates and manual review rates declined toward the end of the year, reflecting a common holiday strategy where merchants increase risk tolerance to approve more transactions during peak demand.
Internet and software companies saw an average payment fraud attack rate of 3.4% in 2025, slightly above the overall benchmark. This reflects a higher risk tolerance paired with lower manual review rates, which many companies offset by increasing reliance on controls like two-factor authentication.
The online gambling industry reported the lowest payment fraud rate at 0.16%, driven by risk monitoring that extends beyond upfront checks. Ongoing coordination across fraud, compliance, and responsible gambling throughout the player lifecycle makes context essential when interpreting benchmarks in regulated environments.
Looking Ahead: Cleaner Systems, Fewer Tradeoffs
In an environment where fraud patterns are increasingly hard to predict, the most effective teams focus on precision. That starts with reviewing and cleaning up data so teams can operate with confidence, even as conditions shift.
Risk tolerance also can’t be static. Fraud teams need to continuously assess risk in proportion to seasonal demand, holiday traffic, business goals, and merchant context. By aligning controls to these realities, teams can protect revenue and trust without relying on rigid rules that break when conditions change.
How to Use FIBR
FIBR is designed to provide context, not conclusions. It helps teams understand what’s typical for their industry, where they diverge, and which questions are worth investigating further.
The public version of FIBR is available to anyone, offering high-level benchmarks across industries. Sift customers can access deeper insights directly in the Sift Console and through the Sifters community, including additional breakdowns by payment method, order value, and chargeback reason.
Explore FIBR to see where you stand and where to go next.
