In the evolving landscape of finance, the integration of artificial intelligence (AI) has become a focal point for many organizations. However, recent research from Payhawk reveals that even those finance teams that classify themselves as AI frontrunners still face significant challenges in adopting the technology within their core operations.
A global survey comprising 1,520 finance and business leaders indicates that while experimentation with AI is widespread, essential elements like governance and data readiness are lagging behind. Payhawk identifies “AI leaders” as individuals who rated their AI maturity from 7 to 10 on a scale of 10, representing a subset of 405 respondents.
Among these self-declared AI leaders, only 26% reported possessing all the foundational elements necessary for effective operational deployment of AI, as outlined in the study. In contrast, a staggering 74% indicated they were lacking at least one crucial component.
For finance teams, operational deployment encompasses the utilization of AI in high-accountability processes such as financial closing, internal controls, approvals, exception handling, audit trails, and governance of expenditure. These findings underscore a disconnect between initial AI adoption and the ability to integrate it into workflows that require rigorous oversight and transparency.
Five Essential Requirements
The study delineates five critical operational prerequisites for scaling AI within finance workflows: execution measures, minimum rules for AI utilization, requisite skills and tools, a dedicated budget, and data that is fit for AI-driven analytics.
Among the AI leaders, confidence levels varied significantly. A noteworthy 78% reported having ample AI skills and tools, while 69% indicated that they had allocated budget for AI initiatives. Execution measures were also relatively robust, with 64% affirming their existence.
However, governance and data preparedness presented as weaker areas. Nearly a third of AI leaders (32%) acknowledged possessing the skills but lacking minimum usage guidelines for safer operations. Meanwhile, 22% reported implementing AI measures without the necessary rules for consistent scaling. Data readiness was also identified as a limiting factor, with 39% unable to strongly affirm that their data was adequate for effective AI analytics.
The findings suggest that the limitations on AI adoption in finance stem less from the lack of skills and more from inadequate control frameworks and data integrity. The study highlights phenomena termed “rules debt” and “data debt,” where advancements in activity outpace the establishment of governance standards and the availability of reliable datasets.
Identifying the Operational Gap
Payhawk’s findings indicate a shift in the barriers that finance functions face as AI technology becomes more entrenched in operational activities. While many organizations showed an intent to invest and some act on governance, the scaling of AI efforts often falters when minimum usage rules are unclear or when systems cannot align AI-generated outputs with trustworthy financial information.
This situation is critical since finance processes necessitate clear accountability, dependable audit trails, and uniform controls. Though automation can alleviate manual tasks, it raises vital questions regarding decision-making, handling exceptions, and the alignment of outcomes with financial records. When AI is incorporated into approvals, reporting, and spend management, organizations must also contend with internal compliance requirements and, at times, external scrutiny.
The survey engaged senior professionals from various roles and sectors, including C-suite executives, vice-presidents, directors, and senior staff members. Participants were drawn from functions such as finance, accounting, sales, HR, and procurement across industries that included services, digital, manufacturing, healthcare, education, non-profit, and B2C companies. Company sizes varied from 50 to over 1,000 employees.
The geographical coverage included DACH, Spain, France, Benelux, the UK and Ireland, as well as the United States. Payhawk collaborated with IResearch, employing affirmative statements developed with finance and business leaders while conducting interviews across eight countries.
Headquartered in London, Payhawk offers spend management software that encompasses bills, cards, expenses, travel, and procurement, alongside a global money account that operates on top of an organization’s ERP system, catering to mid-market and large enterprises in over 32 countries.
The survey also indicates that finance teams may be rapidly testing AI tools without establishing consistent operational guidelines across their functions. Such an approach can be effective in pilot programs, where parameters are limited and oversight is direct. However, challenges arise when AI is expected to facilitate repeatable, controlled workflows across various teams, systems, and geographical locations.
“In finance, AI only matters when you can delegate real work inside controlled workflows like approvals, reporting, and audit trails,” stated Hristo Borisov, Payhawk CEO and Co-founder. “Our data shows the skills and experimentation are already there. What’s missing is the operating stack, minimum rules, and usable data that make AI accountable at scale.”
In conclusion, as organizations seek to integrate AI more deeply into their finance operations, addressing these foundational challenges will be essential. By establishing clear governance frameworks and ensuring data readiness, finance teams can unlock the transformative potential of AI and ensure its effective deployment within critical workflows.
