Can AI deliver the perfect predictions Corporate Treasurers need to ...

Home / Artificial Intelligence / Can AI deliver the perfect predictions Corporate Treasurers need to ...

Can AI deliver the perfect predictions Corporate Treasurers need to survive and thrive ?

Reading Time: 4 minutes


It is said that when the great god Zeus overcame the Titans, he spared two, Prometheus and his brother, Epimetheus and gave them the task of going to Earth to make its creatures. Just before he sent them down from the heavens, he gave Epimetheus gifts to offer their creations. Using the abundant river clay of earth, the brothers began to mold their creations. Epimetheus worked quickly, shaping and molding the animals and each time he finished a creation, he handed out of one Zeus’ gifts. However, by the time he was finished, he realized he had given away all of Zeus’ gifts, and he had nothing left for his brother to give the human beings. Because he had lacked foresight, humans spent their initial years in the caves shivering in the cold, dark night, terrified of the many powerful beasts Epimetheus had created.


While it is an ancient tale that underlines the importance of foresight, and anticipating the future, the adage is just as true today. A recent survey by a Denmark business school Aarhus BSS found that companies that prepare for the future and change their course of action accordingly, are 33 per cent more profitable than those that don’t. They were also found to achieve a 200 per cent higher growth rate than the average company.


Preparation for the future usually requires investment, and hence companies need to have comprehensive visibility of their cash flow. A study carried out by Terry Ward and Benjamin Foster in 1997 found that companies that had depreciating cash flows for one or two years end up filing for bankruptcy. Cash flow, not current profits, determines the viability of an organization as a business can end a fiscal year showing a solid profit but still find itself out of money. As the saying goes “turnover is vanity, profit is sanity, cash is reality”.


Cash flow forecasting thus becomes imperative to spot signals of financial weakness and gauge the risk of failure in the future. Accurate forecasting provides the management team the information needed to guide the company’s strategies for the future including exploring new markets/opportunities or producing different products. However in too many companies forecasts are generated by a series of Excel spreadsheets, using inputs from disparate groups across the company that needed to be aggregated and rationalised in a very manual and tedious process. In a 2017 survey by PWC, nearly half of the surveyed companies cited a lack of automated tools as a key challenge to forecasting. Automated processes that gather and consolidate different inputs can significantly improve forecast accuracy and reliability while also increasing the efficiency of corporate treasurers- who can then focus on the more qualitative aspects of forecasting.


Clearly, forecasting is about much more than automating the collection of information, and here, too technology can help. Accuracy of cash forecasts can be increased by analysing historical data including payments and collection flows which can provide valuable insights and visibility into working capital trends, seasonality and anomalies. Since most of these transactions flow through the companies’ bank partners, more and more banks have started providing digital cash flow forecasting services using sophisticated predictive analytics solutions.  In September 2018, BNP Paribas announced its collaboration with a Fintech – Cashforce – to offer digital cash flow forecasting services for its business clients. Their partnership will allow a cross-integration of corporate treasurers’ existing accounts and functions for a more holistic, streamlined view of cash positions.


Artificial Intelligence could be a game-changer:

Until recently, most cash flow forecasting solutions used statistical regression models. These solutions often struggle to process the complex patterns and inter-relationships between variables such as collection patterns, cost of sales and industry trends. Also, forecasts relying only on past demand patterns can be inaccurate sometimes since they don’t take into consideration both the consumers’ and industry’s changing needs. As a result, the forecasting process is never completely automated and remains dependent on the skill and experience of corporate treasurers. However intuitive cash forecasting methods involving Artificial Intelligence hold great promise. Unlike the usual statistical methods, neural networks can identify complex patterns in cash usage and automatically adapt and adjust the forecasting model.


In July 2018, Mastercard and EU FinTech Strands announced a collaboration to develop a cash management and corporate payments platform for banks, designed to leverage AI and machine learning technology to help their small and medium business customers predict cash flow, expenses and balances. Last year, Microsoft Treasury and Microsoft’s Cortana intelligence team partnered together to build a machine learning forecasting solution for their own accounts receivable operations.



As with any emerging technologies there are still some wrinkles to be worked out with AI. While it does remove a degree of subjectivity from decision making, studies have shown that it isn’t immune to cognitive biases that can sway decisions. There is also the ‘black box’ problem – where it isn’t obvious why an AI has made a particular decision. However, progress is being made on these areas. In September 2018, IBM released and open-sourced an AI bias detection and mitigation toolkit, and it also released a new cloud-based service, designed to make the “black boxes” transparent. Google has also unveiled a bias-detection feature for its TensorFlow machine learning web application, dubbed the What-If Tool.


There is a definite potential of Artificial intelligence in cash forecasting and according to industry experts, early adopters stand to benefit the most. In fact, Gartner predicts that by 2020, embedded AI will become a key differentiating factor in finance systems evaluations, and vendors with this capability will be able to highlight greater functional advantages. Not only would AI-driven forecasts enable organizations to take more informed decisions now but it will also assure them that their future is in safe hands – their own.





  1. Foster, Benjamin, P., Ward, & Terry, J. (1997). Using Cash Flow Trends To Identify Risks Of Bankruptcy CPA journal, 67(9), 60.
  2. Gartner (2018): How AI Will Transform Financial Management Applications
Share this Post

About Nucleus

At Nucleus Software we are committed to providing efficient, modern yet proven software solutions for the global Banking and Financial Service industry. We have been pioneers in developing Retail Banking Software, Corporate Banking Solutions, Transaction Banking, Cash Management and Internet Banking Software since 1986. Our success spreads across more than 50 countries, and we serve our customers globally through our direct and partner operations across US, Europe, Asia-Pacific, Africa and the Middle East. We are known for our world-class expertise and innovation in lending and transaction banking technology. Our two flagship products, built on the latest technology are: FinnOne™ and FinnAxia™.