Today’s treasury teams operate with urgency in complex, global crises.
With companies under pressure to keep up with all the dynamic changes in their cash and liquidity operations, corporate finance divisions are more critical than ever.
But, of course, treasury organisations across the world are far from homogeneous.
Echoing the spectrum of companies and industries that they keep afloat, these teams vary in size, scope and structure. They also use different tools and technologies to manage cash, liquidity and risk.
A significant and perhaps surprising proportion of treasury teams still use spreadsheets. These manual systems are slow and error-prone, preventing treasurers from keeping up with today’s fast pace of change.
It’s not surprising, then, that many are looking for more professional tools.
Treasury management systems (TMSs) help them automate and integrate their financial operations in a single platform, with the most innovative solutions increasingly integrating machine learning (ML) tools.
The 2008 financial crisis showed the criticality of treasury functions for companies to survive and thrive. It also exposed the harms of cumbersome and error-prone procedures and tools to manage financial operations.
While the demands on treasury teams increase, there is a grand digital transformation unfolding, spearheaded by ML.
Over the past few years, we’ve seen technology innovators invest in ML to deliver greater efficiency and insights to treasury organisations of all sizes and complexities.
ML models with the capability to bring together vast swathes of data within seconds are underpinning cash and liquidity management practices. In doing so, these models provide larger and leaner treasury teams with more time and scope to convert their data into valuable insights.
Today’s ML models can save CFOs and their divisions time, elevating the accuracy of their cash forecasts and invigorating retrospective and predictive analysis.
While ML technology is impressive and sophisticated, it can be deployed with relative ease by all types of organisations. However, getting the most from ML requires focus on three key things:
Many companies believe they need to employ data scientists and ML experts to implement ML technology, but this is not the case.
Technology providers design ML algorithms in a way that they can be used by all kinds of companies. During the implementation project, they are tailored to an organisation’s requirements.
Once that’s done, the models have to be trained rigorously with structured, clean data. During the process, specialists from the technology provider work with treasury teams to make sure the models and data best fit the company’s needs.
Today, more innovative TMSs arrive with ML capabilities built in.
As the treasury system acts as a central source for financial data, the ML algorithms help unlock the potential of this data. They deliver new insights and help treasury teams better plan, manage and invest their cash flows.
ML helps to reduce manual tasks, transforms operational processes and establishes data-driven decision-making. Overall, it enables the shift to a more strategic, value-added treasury function.
No one can doubt the transformative potential of ML for treasuries. When supported by forward-thinking, flexible technology providers, organisations will witness the remarkable benefits of ML technology.
Viola Hechl-Schmied is ION’s product owner for ML and an expert in the training of ML and AI models for treasury and cash management functions
This article was taken from Issue 2, 2022 of The Treasurer magazine. For more great insights, log in to view the full issue or sign up for eAffiliate membership
