The modern finance department is currently navigating a period of profound structural tension, characterized by a widening chasm between the heightened expectations of executive leadership and the practical limitations of existing data infrastructure. This assessment emerged as a central theme during a recent FP&A Circle webinar hosted by the FP&A Trends Group, an international professional organization dedicated to the advancement of Financial Planning and Analysis (FP&A). The session, titled "Analysis to Orchestration: How FP&A is Evolving in the AI Era," featured a panel of industry experts from SquareTrade, Neo4j, and Planful. The discussion illuminated a stark reality: while artificial intelligence (AI) is often marketed as a panacea for corporate efficiency, many finance teams lack the foundational data readiness, technical skill sets, and organizational support required to leverage these tools effectively.
The Stagnation of Finance Performance: A Five-Year Statistical Review
The webinar opened with a comprehensive survey presented by facilitator Hans Gobin, which painted a sobering picture of the current state of the profession. According to the data collected by the FP&A Trends Group, the performance of FP&A teams has remained largely static over the last half-decade, despite significant advancements in enterprise software. The survey revealed that FP&A teams currently spend a mere 14% of their time driving strategic action and only 18% generating insights. The remaining 68% of their capacity is consumed by "non-value-add" work, such as manual data entry, reconciliation, and correcting errors in legacy systems.
Furthermore, the survey indicated that only 22% of finance teams consider themselves to be performing at a high level, while 32% describe their operations as "struggling." This lack of progress is particularly notable in the context of skill acquisition. While 48% of organizations now prioritize "business partnering" as a primary hiring requirement, the technical roles necessary to support AI—such as data scientists and systems architects—remain at the bottom of the recruitment priority list. This creates a competency gap where teams are expected to provide high-level strategic advice without the underlying data architecture to support their conclusions.
However, the data also suggested a potential "AI dividend" for those who successfully integrate the technology. Teams that have effectively deployed AI reported a 7% increase in time spent on high-value activities and a 15% improvement in overall performance. The challenge, the panelists noted, lies in moving from experimental usage to scalable, functional implementation.
The Leadership Pincer Movement and the Resource Disconnect
Gaby Makstman, Director of FP&A at San Francisco-based SquareTrade, provided a detailed look at the pressures currently facing finance leadership. She described a "four-way pincer movement" that is squeezing finance departments from all sides. First, role expectations are evolving faster than internal capabilities. Second, planning remains siloed across Finance, Operations, and Sales. Third, there is a mounting demand for real-time forecasting that current tools are unable to satisfy. Finally, a pervasive lack of data readiness serves as a significant barrier to any meaningful progress.
Makstman highlighted a significant irony in corporate resource allocation: while executive leadership is demanding faster, more data-driven answers from Finance, the vast majority of AI investment is being directed elsewhere. "AI investment inside most organizations is going into Operations, call centers, and Marketing—not into Finance, Accounting, or FP&A," Makstman observed. This lack of investment has led to an "inferiority complex" within some finance functions, where lean teams fear that automation or data democratization might lead to further head-count reductions rather than providing much-needed bandwidth.
The disconnect is further exacerbated by the "hallucination" risks associated with Large Language Models (LLMs). Makstman warned that AI outputs are only as reliable as the data they ingest. Without a deep understanding of the data’s context, finance professionals risk presenting flawed projections to boards and investors, which could have catastrophic consequences for both the company’s stability and the professional’s career.
Technical Barriers: Deterministic Functions vs. Probabilistic Solutions
A critical point of analysis during the webinar was the inherent conflict between the nature of finance and the nature of AI. Rowan Tonkin, Chief Marketing Officer at Planful, noted that finance is a "deterministic function"—it requires exactness, auditability, and absolute accuracy. In contrast, AI is a "probabilistic solution," meaning it operates on likelihoods and patterns rather than fixed rules.
"You’re being asked to be exact, and the technology by its nature isn’t," Tonkin explained. This fundamental mismatch explains much of the hesitancy within finance departments to adopt AI for core functions like the general ledger or regulatory reporting. While a forecast that is "mostly correct" can still be useful, a financial statement that is only "mostly correct" is a liability.
Tonkin’s research indicates that 84% of current AI investment has been focused on "copilots" for individual productivity. While these tools help individual analysts write emails or summarize documents faster, they do not solve the organizational problem of data fragmentation. Tonkin argued that the industry must move from "single-player mode" to "multi-player mode," where AI is integrated into an operating model that connects foundation, intelligence, action, and feedback.
Orchestration and the Maturity Model: The Neo4j Perspective
Ankit Chopra, Director of FP&A and Cloud at Neo4j, introduced a framework for viewing the evolution of the finance function, shifting the focus from "analysis" to "orchestration." In this model, FP&A acts as a conductor, managing a complex arrangement of analysts (musicians), tools (instruments), and data.
Chopra identified three distinct stages of maturity for finance teams:
- Control: A reactive state focused primarily on accuracy and historical reporting.
- Agility: A proactive state characterized by real-time drivers and scenario elasticity.
- Orchestration: An advanced state involving continuous learning and coordinated action across the entire enterprise.
According to Chopra, most modern teams remain trapped at the "Agility" stage, unable to reach "Orchestration" due to three persistent pain points: the lack of a single source of truth, the inability to store context and causal relationships, and a lack of "explainability" in AI outputs. He noted that many teams are currently using tools like Claude or other AI agents in a siloed manner. While these tools may help an individual analyst, their outputs are not synchronized between teams, leading to conflicting forecasts and a lack of credibility in board-level conversations.
To combat this, Chopra proposed a four-layer implementation framework consisting of data, forecasting, LLM reasoning, and agentic action. His primary advice for teams looking to start this journey was to "start with one signal, one model, and one alert," rather than attempting a wholesale digital transformation overnight.
Case Studies: Real-World Applications and Results
Despite the challenges, the webinar highlighted successful case studies where organizations have overcome these hurdles. Rowan Tonkin cited Alltech, a global agricultural products and services company, as a prime example of successful digital transformation. Alltech manages global consolidation across 120 countries and 140 legal entities. By implementing a structured feedback and orchestration layer, the company was able to move from a "day-20" month-end close to a "day-7" close, significantly increasing the speed at which leadership can make informed decisions.
Another example provided was McLarens, a global insurance adjuster. McLarens utilized a feedback layer to provide its board and private equity sponsors with real-time visibility across multiple business units, regions, and currencies. This transparency allowed the organization to move away from static, retrospective reporting and toward a dynamic model that reflects the current state of the business.
The Road Ahead: Implementation and Guardrails
The consensus among the panelists was that the path forward for FP&A requires a disciplined, incremental approach. The experts recommended treating AI as one would treat a "new intern"—a tool that has great potential but requires extensive context, clear instructions, and constant supervision.
The following strategic steps were identified for finance leaders:
- Document Processes: Before introducing AI, teams must document their existing manual processes to identify where automation can provide the most value.
- Prioritize Automation: Audience polls during the webinar showed that 52% of attendees believe the "automation of repetitive tasks" is the area where AI can provide the most immediate benefit, far ahead of predictive forecasting or anomaly detection.
- Establish Guardrails: Validation is non-negotiable. Putting AI-generated data into executive decks without rigorous human verification creates unacceptable levels of corporate risk.
- Address the Skill Gap: Organizations must move beyond hiring for "soft skills" alone and begin investing in the technical data science capabilities required to manage AI systems.
Broader Impact and Implications
The implications of this shift are significant for the broader corporate landscape. As AI continues to permeate the enterprise, the role of the CFO is evolving from a "scorekeeper" to a "strategic orchestrator." However, this evolution is contingent upon the willingness of organizations to break down data silos and invest in the foundational infrastructure that AI requires.
The webinar concluded with the observation that data is rarely perfect and is almost always siloed. Fixing these silos is not just a technical challenge but a political one, as those who control the data often control the narrative within a company. For FP&A teams to truly move the needle on performance, they must overcome the "inferiority complex" and advocate for the resources necessary to turn their data into a strategic asset. If the profession fails to adapt, it risks remaining in a state of perpetual struggle, spending the majority of its time on manual tasks while the rest of the organization moves forward into the AI-driven future.
