Data has become a decisive force in payments, but most firms still struggle to use it with precision.

That distinction, according to Max Spivakovsky, senior director of global payments risk management at Galileo, is separating firms that merely collect information from those that convert it into sustained advantage.

“Data is one of the biggest plays right now in the market,” he said. “The lagging organizations treat the data as a storage problem while the leading organizations actually treat it as a decisioning system.”

Spivakovsky pointed to decision quality as the central metric governing success, arguing that leading organizations are not simply reacting faster but are measuring whether those reactions produce better results.

Firms that integrate customer, operational and financial data into a single view are better positioned to act with speed and context. Spivakovsky said breaking down silos allows companies to “be much more proactive,” particularly when responding to customer needs or emerging risks.

Anecdotal evidence illustrates the point. In one case, analysis of negative balances enabled a program to extend minimal overdraft flexibility, which in turn strengthened engagement among established users. The improvement was modest in dollar terms but meaningful in behavior, suggesting that precise data application can influence both retention and revenue.

AI Reshapes Payments and Fraud Decisioning

The role of artificial intelligence (AI) has intensified that evolution. In a “What’s Next in Payments” series discussion focused on the “data game,” Spivakovsky described how AI is altering both the speed and quality of decision-making across payments and fraud management.

Within payments, AI is already optimizing routing, reconciliation and exception handling, which reduces unnecessary declines and shortens onboarding timelines. “It helps to optimize routing, reconciliation and exception handling,” he told PYMNTS, resulting in “fewer necessary declines from the customer perspective and quicker onboarding.”

In fraud, the shift is more structural. Traditional rule-based systems are proving insufficient against increasingly complex attack patterns. AI-driven models, by contrast, can adapt continuously. The traditional fraud system relied on fixed rules, Spivakovsky said, noting that “modern fraud is moving too quickly for purely static approaches.”

The result is a transition toward real-time, adaptive decisioning that integrates new data as it emerges, rather than relying on preset thresholds.

Unified Views and the Feedback Loop

Beyond speed, the quality of customer engagement is also changing. Spivakovsky emphasized the importance of unified data environments that consolidate disparate information into a single perspective. These environments enable firms to respond more precisely to customer behavior while maintaining consistency.

Equally important is the feedback loop that such systems create. Organizations that learn from each interaction, rather than treating transactions as isolated events, develop insights that competitors find difficult to replicate.

This iterative process supports more tailored engagement strategies, particularly in segments with varying risk profiles. Firms serving gig workers, for example, cannot apply identical controls to higher-income customers without introducing friction or misjudging risk.

End-User Experience Improves With Precision

For end users, the practical effect of these changes is reduced friction and more consistent approvals. Data-driven personalization allows institutions to distinguish between low-risk and high-risk behaviors with greater accuracy.

Spivakovsky framed the objective in operational terms. “The best companies use data … to make good customers move faster, while at the same time reserving the scrutiny only for the high risk situations,” he said.

That balance, he suggested, improves both customer satisfaction and approval rates by minimizing unnecessary intervention while maintaining vigilance where it matters.

Fine-Tuning Fraud Responses

The refinement of fraud controls is another area where data is having meaningful impact. Rather than applying uniform rules, firms are beginning to tailor responses based on broader context, including customer history and transaction patterns.

Spivakovsky cited internal scoring approaches that allow institutions to calibrate fraud responses in line with both risk tolerance and brand posture. The objective is not to increase the volume of data, but to ensure that relevant data is delivered at the right moment to inform decisions.

“It’s not about flooding the teams … it’s actually about giving them the right data,” he said.

Who Wins the Data Game

Looking ahead 12 to 24 months, Spivakovsky expects the advantage to accrue to firms that combine broad data visibility with disciplined execution. The ability to make rapid decisions is necessary but insufficient without governance and transparency.

Firms that treat data as an evolving system rather than a static asset are building capabilities that extend beyond immediate gains. As Spivakovsky put it, “the companies that win will treat trust, identity and security as core part of their data strategy overall.”