Artificial intelligence may dominate boardroom discussions, but payments companies face a more basic challenge.
They already possess enormous amounts of financial data. The harder task is turning that information into something finance teams can trust.
That theme anchored the latest installment of the PYMNTS Summer School series, which featured Ben Catterall, global head of solutions engineering at the finance ERP Fynapse. Firms that gain the strategic advantages will not necessarily collect more data than their competitors, Catterall said. They will build stronger financial foundations that preserve the meaning behind every transactional event.
“Every payment company has tons of data,” Catterall said. “But data without context is not particularly useful.”
It’s crucial to understand what the data represents, along with “what happened in the real world to generate that data,” he said.
Cards, digital wallets, buy now, pay later (BNPL) products, subscriptions and app store purchases each generate different financial records. Cross-border commerce adds currencies, settlement differences and multiple payment gateways, while transaction volumes continue to climb.
Regardless of payment type, finance departments must still reconcile transactions accurately and move them into the general ledger without delaying the business. Legacy finance systems, built around batch processing and summarized data for a pre-cloud era, often struggle to keep pace with that complexity.
The consequences extend beyond operational efficiency. Catterall described one multinational payments client that processed transactions across nearly 18 countries. At a summary level, the books appeared balanced. A closer examination of transaction-level records uncovered foreign exchange spreads that averaged about 2%, creating millions of dollars in lost revenue across roughly $100 million in cross-border payments volume.
“If you apply that to $100 million of cross-border payments being processed, that could be $2 million a year that’s lost,” Catterall said.
For finance leaders, examples like that reinforce why transaction-level visibility is more than an accounting exercise. It provides a way to understand where revenue disappears, where payment costs accumulate, and where operational changes can improve financial performance.
Build the Foundation Before Deploying AI
The same principle applies to AI.
Many financial institutions have launched AI proofs-of-concept during the past two years, but relatively few have progressed into production. The obstacle often lies beneath the AI models themselves, Catterall said.
“If you’re relying on batched, aggregated, summarized data, and you put an AI tool on top of that, all that AI tool can learn from is the summary view,” Catterall said. “It doesn’t have enough to go on.”
Research shows that 46% of AI proofs of acceptance fail to reach production because poor data quality limits their effectiveness, Catterall said.
That philosophy shapes Aptitude’s platform, which captures financial events as they occur instead of reconstructing them at the end of the reporting period.
The objective is not simply a faster month-end close. It is to give finance teams continuous visibility into margins, payment costs, and business performance while activity is taking place in real time, enabling finance teams to drive real-time business decisions.
Finance organizations should treat financial data as core infrastructure rather than a byproduct of payments, Catterall said. He added, “Making that available to the business is a differentiator.”
Companies that preserve detailed transaction records and make them available across treasury, pricing, forecasting and risk are better positioned to support growth without sacrificing financial control. This is what financial truth for payments really means and is what Fynapse delivers for the likes of T-Mobile, which now processes 200 million journal lines an hour in real time.
As payment methods multiply and AI assumes a larger operational role, that discipline may prove to be finance’s most durable competitive advantage.
The firms that modernize their finance data architecture today will be better equipped to understand tomorrow’s transactions rather than simply record them.
What Finance-Grade Data Unlocks
A modernized approach to data means three things. First, the data shows individual transactions, not just totals. Second, it keeps the details behind each transaction. Those details include the payment method, the currency, the fees and who was involved. Finally, the data is available right away, not pieced together later during a close.
Watch the complete PYMNTS Summer School interview to learn:
- Why real-time accounting changes the relationship between finance and auditors.
- How agentic AI could reshape financial governance and reporting.
- What finance teams should demand before approving another AI initiative.
For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.
