Finance leaders must shift their teams from a scorekeeper role to a proactive business partner by using AI to free up human capacity, prioritize clean data, and continuously adapt talent for strategic insights.
Bruno J. Navarro
Senior Editorial Strategist, Finance
Workday
AI’s true value isn't just cutting costs, but fundamentally reshaping the finance function to drive top-line business transformation.
It’s a big-picture perspective finance leaders should understand to optimally position their teams to prepare for the future.
In a recent panel discussion, industry leaders—Karen Schreiber, principal, Advisory, Corporate Services at KPMG US; Josh Schwartz, CFO at Medidata; and moderator Zane Rowe, CFO at Workday—unpacked this critical theme.
Guide
Schreiber expects AI technology to bring about large-scale change. “Our hypothesis is that in the next three years, 80% of finance will be automated,” she said, quickly adding, “I didn't say eliminated. I said automated.”
The radical restructuring means that remaining finance roles will be higher value and more strategic, Schreiber said.
The future finance function will feature specialized teams: an AI execution team combining human and agent expertise, an AI business partnering role using models to generate first-pass insights, and an AI innovation team focused on continuous improvement and identifying new market opportunities. Notably, human expertise in governance, risk, and controls will remain essential and irreplaceable.
Schwartz reinforced this strategic view, stating that internal transformation is necessary to credibly sell external transformation. The CFO's role, therefore, shifts from scorekeeper to holding the organization accountable for transformative, long-term business decisions.
“As I look across to my team as CFO, my job is to hold them accountable to budget, to hold them accountable to make the right business decisions,” he said. “And if we're not able to do that in a way that is really transformative, we lose the ability to tell that story and how our technology can change how our customers think and work.”
“Our hypothesis is that in the next three years, 80% of finance will be automated.”
The discussion quickly highlighted that successful AI adoption rests on solid, often “unsexy,” infrastructure.
“No one wants to tell my CEO that we got to be smart and spend the money on, we should bring in, spend all this effort on cleansing data, building structured data. It doesn’t resonate,” Schwartz said. “Now when you have a flashy story about, ‘We're going to predictive forecast for the next three years in our six-year model,’ that moves the needle.”
Rowe noted that clean data represents the “oxygen of AI,” adding, “If you don't have good data, you'll struggle from the very beginning.” This highlights two critical areas for successful AI implementation:
The data imperative: AI reliability, particularly in granular tasks like forecasting, depends entirely on structured, high-quality data. Schreiber shared an example in which a KPMG client in the Fortune 100 achieved “wildly successful” cash-flow forecasting only after fixing broken data hierarchies across all its core systems.
AI is rapidly changing the skill requirements for finance professionals, effectively “flipping the pyramid” of experience, Schwartz said. Agents are now taking over the repetitive, entry-level work traditionally used to train new staff.
This creates a new leadership challenge: teaching young staff, who are comfortable using models, how to question the output and truly understand the business. The key mandate for the modern finance team is to stop being merely a data output function and start with a hypothesis about the business problem they are trying to solve.
“From a finance org, we got to empower ourselves and challenge our teams because it’s about getting the right margin profile that allows people to hit financial results,” Schwartz added.
Practical use cases are already emerging, Schreiber said, such as KPMG’s internal predictive close using agents for AP invoice/PO matching (focusing human effort only on exceptions) and generating first-pass variance explanations.
The panelists emphasized that the first crucial step in AI adoption is getting people familiar with the technology—and understanding what it means for their work.
Schwartz noted that initially, the concept of AI felt like it meant “everything and nothing” because it's used so much, leading to confusion about where to even start. To address this, Schwartz said Medidata's approach focuses on investing in people’s training and using methods like hackathons to help employees become familiar with AI and see successful applications.
“That’s something I feel really strongly about, sharing ‘win stories’ about how people are using it, because no one knows where to start,” Schwartz said.
Alluding to the concept of Rosie the Robot from the animated, futuristic, 1960s television series “The Jetsons,” Schreiber said, “I never thought that there would be a robot who would be really part of our everyday life.”
To begin incorporating AI technology into the daily routine, Schreiber said she challenged her team to develop AI that would say to her each morning, “‘Hey, Karen, here are the five deals that didn’t close last night that we expected. Here’s the three people you need to call. And here’s some suggestions of what you need to do to get that deal closed.’
“And at first they looked at me like, ‘I don't know that we can do this.’ I said, ‘It doesn't need to be perfect, but let's build it.’”
The exercise becomes part of an iterative process, Schreiber said, in which the finance team can push back on the answer or ask the agent further questions.
The panel distinguished between the immediate benefits of AI and its ultimate value.
At the moment, AI is delivering on efficiency, productivity, and capacity. Yet the most profound benefit is capacity—freeing up finance talent to do different work, transitioning them out of “building the data and analysis” and into developing insights.
However, this transition involves a profound risk, known as the pricing conundrum. If a business like Medidata uses AI to solve a customer’s problem, such as accelerating clinical trials, it runs the risk of commoditization and shrinking revenue if it doesn’t fundamentally change its pricing model and value metric to reflect the outcome achieved by AI, rather than the effort expended.
Strategic outcomes become an important component of measuring ROI of AI investments.
Schwartz noted that the goal at Medidata is to enable people to do “different work, not more work,” by using AI to allow teams to focus on higher-value activities. “Right now the reality is most of the AI that we see in our internal uses is capacity, efficiency, and productivity,” he added.
Schwartz's vision of empowering teams to focus on higher-value activities through AI manifests in several key areas, including:
Shifting the role of finance: The finance team’s job isn’t just to report the news, Schwartz said, but to “change the outcome before it happens.” In practice, Schreiber added, that means asking strategic questions like, “What should the margin profile for product X be?” rather than just pulling data.
Driving top-line growth: The ultimate ROI is tied to improving the business. This includes getting new products to market faster and, in Medidata’s case, allowing customers to run fewer, smarter clinical trials because AI helps them predict the outcome.
Schreiber also highlighted the current inadequacy of traditional metrics as a key challenge to measuring returns.
“Traditional KPIs or traditional metrics are very focused on cycle time, error rate, number of transactions, or invoices processed per FTE,” she said. “The problem right now is: The benchmarks don’t have enough use of AI in them to actually have a comparison.”
However, Schreiber noted, the focus is shifting toward measuring improvements in quality, accuracy, and the ability to grow revenue without a proportional increase in headcount.
“We’ll learn more as the technology continues to change, and those KPIs will change,” Schreiber said.
“Right now the reality is most of AI that we see in our internal uses is capacity, efficiency, and productivity.”
The journey to harness AI’s transformative value in finance requires a continuous commitment to evolution. Industry leaders agree: Beyond mere efficiency, the ultimate benefit of AI lies in freeing up capacity—redeploying finance talent to focus on higher-value activities and strategic insights that drive top-line business growth.
Success in this new era hinges on a few core mandates:
Build the ‘oxygen of AI’: Ensure a solid foundation of structured, high-quality, clean data.
Flip the talent pyramid: Invest in training and upskilling talent to transition from repetitive tasks to questioning AI output, generating hypotheses, and focusing on business problems. Hackathons and sharing “win stories” are crucial to get teams comfortable with the technology and demonstrate successful applications.
Imagine new metrics: Acknowledge that traditional metrics (like cycle time and transactions per FTE) are no longer adequate. The focus must shift to measuring improvements in quality, accuracy, and the ability to grow revenue without a proportional increase in headcount.
The future finance function will feature specialized teams combining human expertise with AI agents, focusing on execution, business partnering, and innovation. Understanding this approach will empower the CFO to move beyond budget reconciliation and fully embrace the mandate to drive transformative value.
Finance leaders are facing increased expectations from both internal and external stakeholders. Download this report to uncover the top five reasons CFOs are moving to Workday to optimize their finance operations.
Guide
Pressure for finance leaders to invest in AI is at an all-time high, but where are CFOs spending their technology budget? Read on to learn more about the potential impact of widespread AI implementation from 2,355 senior business executives.