Deutsche Bank: AI Can Be a Game Changer for Treasurers

Innovation should be applied wherever there is a pain point.

This is particularly true when it comes to the enterprise treasury function, where technology stacks are becoming more advanced than ever.

Treasurers today must be more agile in decision-making, Claudia Villasis-Wallraff, head of data driven treasury at Deutsche Bank, told PYMNTS.

“Companies need to adopt new technology,” she said. “And with this, I not only mean adopting API connectivity, but also cloud functions and artificial intelligence.”

While treasury management systems (TMS) and enterprise resource planning (ERP) systems have traditionally been focused on operational tasks like accounting and financial instrument valuation, the dynamic macro backdrop and ongoing rate environment have transformed the calculus around effective treasury management and opened a new horizon of opportunity and investment for businesses.

Existing treasury management programs frequently fall short in aiding the kind of treasury decision-making that is necessary to capture the growth opportunities today’s environment entails.

As Villasis-Wallraff pointed out, onboarding an entity or a bank account in a TMS requires effort from multiple departments, often leading to decisions based on partial information. This scenario underscores the need for modern treasurers to adopt new technologies that can provide comprehensive, real-time data and insights.

The Impact of AI on Treasury Functions

Still, for treasurers, more data does not necessarily translate into better cash flow forecasting or decision-making — and that’s where the role of AI comes in.

One of the most significant applications of AI in treasury is in cash flow forecasting, specifically direct forecasting, Villasis-Wallraff explained, noting that AI-driven models can predict when clients are likely to make payments by analyzing past behavior and market variables.

This capability extends beyond forecasting; AI can also empower treasurers in deploying recommendation models to make more efficient funding, hedging and investment decisions, she added. These models can align with a company’s risk appetite and policy parameters, offering tailored recommendations that enhance decision-making processes.

But that’s not all AI can do. Villasis-Wallraff highlighted another critical, yet less discussed, use case: the categorization of bank transactions. AI can automate the identification of transactions as operational or non-operational, or salaries, taxes or payments to vendors. This automation can streamline treasury operations, allowing treasurers to focus on more strategic tasks.

The Path Forward for Treasury Teams

While AI offers substantial benefits, integrating it into treasury workflows is not without challenges. One of the primary hurdles is the quality of data, as well as the engineering and technical resources to activate that data. As Villasis-Wallraff put it regarding data quality, firms must be aware that “garbage in equals garbage out.”

Treasurers need to start structuring and collecting high-quality data to ensure accurate and reliable AI model outputs in the future. This data-driven approach may require a cultural shift within organizations, where both finance and technology teams need to work collaboratively.

For AI to truly transform treasury functions, C-level executives must recognize the potential return on investment that these technologies can bring. Villasis-Wallraff explained that as interest rates shift and instant payments become more prevalent, the demands on treasury teams will increase — a situation that companies need to get ahead of now by investing in education and fostering a closer alignment between technology and finance teams.

“Shareholders and the C-level are going to start asking more and requesting more from their treasury teams,” she said, adding that looking ahead, the ability to create operational cash flow forecasting without manual intervention will be a game changer for treasury teams.

It’s a future that Deutsche Bank is at the forefront of creating, Villasis-Wallraff said, by working closely with clients to develop and implement AI-driven solutions for cash flow forecasting and transaction categorization.