The current surge in artificial intelligence (AI) adoption mirrors the dot-com boom of the early 2000s, but with a distinct challenge: the last-mile problem. Back then, the bottleneck was inadequate telecommunications infrastructure, which slowed digital transactions. Today, organizations face a different hurdle—maximizing the value of AI investments despite readily available tools and capabilities.
The core issue lies not in the technology itself but in how it is implemented, supported, measured, and managed. Key obstacles include unclear return on investment (ROI), a focus on productivity over transformative improvements, and inadequate planning and change management. These challenges impact revenue, expenses, and ultimately, shareholder value, making it a critical area for CFOs to address.
The Complexities of AI Integration in Finance
A year ago, the expectation was that a $3 million AI investment would pay for itself within a year through efficiencies and headcount reductions. However, this outlook has shifted, with payback periods now extending to three to four years. As financial markets demand swift, substantial returns, organizations must confront several last-mile obstacles:
Unclear ROI Definitions
The costs of AI extend beyond subscription fees, including token consumption, internal time, process changes, and third-party services. On the return side, organizations often overlook benefits like time savings, accuracy improvements, and customer experience enhancements that are difficult to quantify. Some AI providers suggest adopting new metrics, such as cost per successful task and result dependability, to better capture the true value of AI outcomes.
Overemphasis on Productivity Gains
According to Gartner research, 84% of finance AI spending is directed toward productivity and process improvements, while only 16% focuses on high-value use cases that significantly alter business outcomes. This imbalance is prevalent across organizations, highlighting a missed opportunity for transformative change.
Consumption-Based Cost Models
Organizations are struggling to adapt to hyperscalers’ evolving pricing models, which introduce costs that were not anticipated just six months ago. This lack of preparedness underscores the need for better planning and financial forecasting.
Widespread Usage Without Formal Plans
While nearly three-quarters of CFOs report AI usage within their teams, fewer than half operate under a formal AI plan. This gap between adoption and intentionality captures the essence of AI’s last-mile problem.
Complex Workflows
The final 20% of AI enablement is the most challenging, involving complex workflows with exceptions and cross-functional handoffs. Taming this complexity is difficult but could yield significant value.
The CFO’s Role in AI Success
CFOs are uniquely positioned to lead AI investments due to their focus on ROI and value creation, commitment to data-driven decision-making, and growing role as innovation advocates. Their control over budgets and accountability for spending make them ideal for connecting AI investments to tangible value. Additionally, finance teams are equipped to understand and calculate new cost measures, such as cost per token and API call charges.
CFOs can encourage a holistic approach to AI deployment, replacing fragmented efforts with a cohesive strategy. By stepping back and asking critical questions about AI investments, they can validate current investments and pave the way for future ones. This disciplined approach ensures that AI’s potential is fully realized, driving innovation and value creation within organizations.
