Why the overwhelming majority of treasuries producing cash flow forecasts still use Excel and won’t likely stop anytime soon.
Despite decades of technological advances, Excel remains the predominant tool that treasury teams use to generate cash flow forecasts, according to the results of NeuGroup’s May 2022 Cash Forecast Survey (see chart below). That includes teams that have a treasury management system but use Excel along with it because the TMS doesn’t offer the same degree of modeling flexibility.
- Yes, a TMS is a more effective solution to automating the collection of data than Excel, which is a lot more cumbersome and prone to error. But “the TMS does not have enough flexibility in analysis, scenario planning and reporting,” a member of NeuGroup for Growth Tech Treasurers said.
- Given that cash forecasting efficiency and effectiveness has been a perennial challenge, many in treasury are disappointed by the lack of progress by TMS vendors. However, many are also skeptical about vendors’ ability to provide more robust functionalities.
- The survey revealed that less than half of treasury teams expect to invest in new forecasting technologies in the next year, either because they already implemented a TMS or they don’t intend to.
The opacity of data. One reason it has been so hard to automate the curation of data to support the cash forecasting process is that cash data resides in disparate source systems.
- While some data can be pulled from banks and ERPs, this approach provides only a limited picture because additional data resides outside of those sources. There is also the issue of “black box” areas, e.g., international accounts that are outside the cash pooling or IHB structures. Another is M&A.
- “Our main challenge is visibility. Access to actuals is a very big impediment because it involves the automation of collection of data from multiple sources’ systems,” one survey participant said.
- That’s why treasuries continue to use a combination of direct and indirect forecasting methods. While treasury would prefer a direct approach, it needs to fill in the gaps with information from the business and other systems.
In real time? For TMS vendors and users of other solutions, there is good news. While system connectivity and data “fetching” had traditionally required expensive integrations, with the advent of APIs, some data compilation can be more easily automated.
- Current applications of APIs are primarily related to gathering actual information, e.g., bank balances. They have not yet developed into supporting the forecasting process, e.g., by building a direct link to analytics solutions.
The next generation. Technologies are catching up with treasury’s complex forecasting requirements. A handful of respondents to the survey have already deployed AI and machine-learning tools to facilitate data collection, develop more reliable models that continuously learn from experience, and produce customized reports.
- While the use of AI/ML may be rare, this is a good time to present treasury use cases. “You have to take advantage of the entire company’s digital transformation process and resources,” said one member.