For Treasury4, the AI-Driven Future of Finance Starts With Data

# AI

# Technology

# Sponsored Content

Ed Barrie and Randy DeVita of Treasury4 explain why companies must build a strong data foundation before AI can transform forecasting and reporting.

Artificial intelligence is everywhere in finance conversations this year, including at NeuGroup meetings, where members are eager to understand how the technology will shape treasury. But as Ed Barrie and Randy DeVita of Treasury4 emphasize in a new episode of the Strategic Finance Lab podcast, AI is only as good as the data behind it.

“Without clean, well-structured data, it’s impossible to unlock AI’s potential,” Mr. Barrie tells NeuGroup’s Justin Jones in the episode, available now on Apple and Spotify .

Mr. Barrie is the fintech’s co-founder and chief product officer. Before launching the company, he led treasury and finance teams at Microsoft, Salesforce, Tableau and Itron, earning multiple industry awards for innovation. Mr. DeVita, Treasury4’s vice president of customer success, has spent more than two decades implementing treasury technology at firms including GTreasury and Kyriba.

Together, they describe how Treasury4’s cloud-based analytics platform helps corporates turn fragmented data into structured insights—rapidly onboarding banking history, enriching transactions and improving forecasting, reporting and reconciliation.

Randy DeVita, VP of Customer Success at Treasury4

Ed Barrie, Chief Product Officer at Treasury4

Recent acquisition gives added weight. In September, Treasury4 acquired TreasuryGo —bringing on board George Zinn , co-founder of TreasuryGo and former longtime treasurer at Microsoft, as its chief strategy officer. The deal is intended to enhance Treasury4’s capabilities around bank account management, workflow and debt management—which the company says will deepen its “AI-ready data infrastructure.” The foundation for AI. The expansion underscores the point Mr. Barrie and Mr. DeVita make in the podcast: cutting-edge tools deliver the most value when they’re built on a consistent, well-defined data infrastructure.

“You can’t just take data for data’s sake and throw it at AI,” Mr. Barrie says. “You have to give it appropriate definitions.”

Mr. DeVita adds that without proper organization and definitions, data “just lays on the table like a bunch of puzzle pieces.” When those pieces are put in the proper context, corporates can connect transactions to entities and accounts, spot exceptions and use AI to identify key trends.

From onboarding to forecasting. The pair outline Treasury4’s process, which starts by pulling in years of banking history, then layers on definitions tied to the statement of cash flows and finally automates reconciliation and reporting. That workflow, they say, creates the structure needed for AI and machine learning to generate forecasts that complement FP&A projections and give treasury teams new insight.

Both of them stress that adopting AI in treasury is a journey, not a flip of a switch. It doesn’t require a massive overhaul, Mr. Barrie notes—companies can start with one dataset and expand over time. He predicts a gradual shift in treasury teams, with some adding specialists he calls “ prompting experts ” who know how to ask the right questions of AI tools to generate insights for decision-making.