By now it should be obvious that artificial Intelligence (AI) and agents in all of their forms are on the brink of changing how finance and accounting departments operate. The basic outlines are already in place, but it’s not clear how or how rapidly day-to-day operations will evolve, as well as (by definition) what surprises are in store.
One of the most profound impacts of AI will be on the role of the financial planning and analysis (FP&A) group as much of today’s mechanical, low value-add workloads are taken over by technology, especially data preparation and narrative situational reporting. This will enable FP&A to assume a more strategic and more consultative role in supporting forecasting and planning in business units. Increasingly capable business planning software platforms will be able to support rapid, high-participation, action-oriented planning cycles that combine operational and financial planning to give senior leadership teams, executives and managers more immediate situational awareness and performance insights, as well as useful guidance on go-forward options and their consequences. The main obstacles to achieving this bright vision are data and change management.
Timely, accurate and usable data is critical for supporting AI in all of its forms. Because of this, software providers have been designing business applications to include what I call a
I use the term data pantry to conceptually simplify this type of data store, but the technology that supports it is quite sophisticated and the quality of results from its design depends on the quality of the software engineering that it is built on.
An important advantage of a data pantry is that it facilitates the use of external data in forecasting and planning. Until recently, it was adequate for organizations to regard external data as a “nice to have” item, but that’s no longer the case. External data enhances the predictive capabilities of models by dealing with the problem of endogeneity— specifically, the issue of omitted variables—in forecasting and analysis. To illustrate, consider a vendor that sells ice cream on the beach. On most days, the price does not change, but on cooler than normal days, when there are fewer people on the beach, they often drop the price so that they are able to sell as much inventory as possible. Conversely, on very warm days when the beach is crowded, they raise the price and they are able to sell even more than average because people are willing to pay more. At the end of the season, the vendor brings in a consultant to advise on pricing for the coming year. If that numbers-driven consultant only considers the daily sales totals and the price received each day, the “obvious” conclusion is that higher prices lead to more unit sales, so just raise prices. But that illogical conclusion is only possible because the analysis is missing a consequential variable about the external environment: the daily temperature.
External data is essential for creating robust and performative models. AI and agents are already embedded in planning applications with much more to come. AI that applies continuous machine learning (ML) to ensure the predictive quality of forecasting and planning models will produce more useful projections. AI and agents can enhance the breadth of analytics available to improve situational awareness and decision-making. But the results will be sub-par in many cases if enterprises only use internally generated data. For these purposes, external data is almost always necessary to create performant models because such data can have high explanatory value in determining demand, supply, prices and costs and, in so doing, enables systems to avoid undermining their credibility. Robust data stores that hold a large and diverse set of data from which inferences are gleaned will create more useful and accurate forecasts.
ISG Research asserts that by 2029, one-third of organizations will incorporate comprehensive external measures in performance reviews and benchmarking to improve
Becoming a predictive finance department will also require a heavy dose of change management to redefine the work that’s performed and the resulting shifts in roles and responsibilities. I don’t expect the number of analysts in an enterprise to change, but only if the FP&A group’s mandate is broadened and deepened to become a planning and analysis center of excellence.
A good deal of the alarmist commentary on the future of jobs in an AI-dominated environment is based on the simplistic assumption that increasing efficiency alone is the determinant in deploying technology. That certainly will be the case in some instances, but not all. History is instructive. In mid-century America, tens of thousands of people were needed to create monthly bills for utilities and other basic services. Early computers eliminated that work, while decades later voicemail and personal computers did away with typing pools and “secretaries.” In both cases, the scope of work performed by individuals could largely or entirely be automated.
That doesn’t apply to FP&A under the basic rubric, “the questions you ask are as important as the answers you receive.” A good analyst uses an inquisitive open mind to frame business issues and uncover the sometimes-hidden drivers of outcomes in order to present insights and logical solutions. Rather than having to spend their time on data preparation, slinging spreadsheets and purely formulaic reports about what just happened, these analysts will spend the bulk of their time in more of a forward-looking advisory role. That won’t happen all at once or automatically.
A good deal of FP&A’s traditional role in executing the budget process and providing financial and management data and analyses to the company can be automated with AI and agents, so the group’s mission has to focus on making forecasting and planning more valuable to executives and managers. Planning and budgeting software has helped the CFO and FP&A organization create planning models, produce a detailed budget and analyze “the numbers” with less effort than was required in the past. That is important, but it is not enough.
To make budgeting more valuable to running an organization, FP&A must also have the skills and business knowledge to provide guidance and leadership in modeling and measuring the “things,” or resources, that budget owners use to achieve their business objectives, not just calculate their monetary value. When business unit managers plan (and budget), they usually think in terms of the things they need to run their operations, such as headcount, raw materials, parts, advertising campaigns, facilities, laptops and other items. They need to track timelines and time-on-task. The business planning system must be able to assist in identifying drivers and key performance indicators to give managers a means of assessing performance. It also must simultaneously translate this list of resources into accounting line items so the system can aggregate the financial data into an organization-wide budget and financial forecast. The evolving AI and agent-enabled business planning software will facilitate that process, especially in assisting in statistical analysis, model building and model maintenance. But to realize the full advantage of these capabilities, the role of FP&A in that environment must be expanded. And that will require a considerable change management effort.
I strongly recommend that CFOs and FP&A leaders have a roadmap for evolving into a center of excellence to serve as a resource for executives and business unit management to provide them with the means to forecast and plan in short, streamlined cycles, provide situational awareness and the ability to intelligently set courses of action, as well as assess operational and financial performance. I recommend that business planning software providers offer the means to support this expanded role, with software that is engineered to optimize performance versus cost. Providers must also offer libraries of external data that can easily be incorporated into their data store and seamlessly used by customers.
Regards,
Robert Kugel