Resources

Return fraud costs U.S. retailers over $103 billion a year. Here's the framework high-volume merchants use to catch it - without hurting the customer experience.

QSR looks like one of the more fraud-cyber converged verticals in our 250-company study, with the highest structural integration, highest CISO ownership, most confident fraud team. But its own threat data tells a different story: bot/credential stuffing is QSR's #1 threat, yet ranks dead last in what QSR wants a unified platform to solve. This post breaks down the gap, and why QSR's median fraud performance looks nothing like its mean.

Security and fraud teams often measure the same problem differently. Here's the shared language: five core fraud metrics, plus the Precise Yes Score, that turn security events into numbers the whole business understands.

Choosing a fraud detection partner is a business-critical decision, not just a technology purchase. Here are the six questions enterprise buyers must answer before issuing an RFP, covering revenue alignment, lifecycle coverage, vendor incentive models, chargeback integration, vertical expertise, and data network strength.

Credential-stuffing attacks used to be easy to spot - datacenter IPs, repeated addresses, obvious patterns. Not anymore. Fraudsters are now renting residential proxy networks that make bad logins look identical to your best customers, and standard perimeter tools aren't built to catch it.

What is loyalty abuse in quick service restaurants, and how can QSR brands stop it? This guide covers the top abuse patterns, detection signals, and account takeover risks in loyalty programs.

In our research on fraud-cyber convergence, Retail leads every vertical on fraud metrics, yet it's also the one most concerned about what's coming next. Here's what 50 retail fraud and security leaders told us about their signature threat, where they're investing, and the gaps still open.

Fully converged organizations see 3.4x more dollars approved per dollar lost to fraud. New data from 250 risk leaders reveals which perceived "blockers" to convergence are real — and which aren't.

A 2026 Accertify + Liminal study of 250 fraud and risk leaders found organizations that fully converge fraud and cybersecurity score 3.4x higher on fraud precision.

We kept seeing fraud and cyber teams join forces across our client base and wondered if it actually improved performance. So we partnered with Liminal to find out - and the answer wasn't just "yes." It was $1,084 more approved per $1 lost to fraud, with a catch: doing it in the wrong order can leave you no better off than staying siloed.

Fraud decisioning is a series of tradeoffs. Every decision either protects against fraud or risks impacting a good customer, which means there is no perfect answer, only a balance. The real question in machine learning fraud detection is not simply whether a model is accurate. It is what the model is being optimized to achieve.

The reality is simple: if a business isn't discoverable, it effectively doesn't exist for a large and growing segment of consumers. That same dynamic is beginning to shape agentic commerce. What Is Agentic Commerce? Agentic commerce is the emerging shift in which AI agents research, compare, and — increasingly — act on a shopper's behalf, from narrowing options to initiating a purchase.