Five years ago, most technology investments came with familiar financial models. A company upgraded its ERP system, improved its infrastructure, or invested in automation. The expected costs, timelines, and returns could be estimated with reasonable confidence.
AI has changed that equation.
Today, finance leaders are being asked to approve investments whose eventual impact may extend far beyond the department requesting the budget. For instance, a customer support team may ask for an AI tool to reduce response time. The immediate benefit appears operational. Six months later, the same investment could influence customer retention, cross-selling opportunities, and revenue growth of the organization.
That is what makes AI fundamentally different from many previous technology cycles. The business case rarely stays confined to one function.
At the same time, the pressure to invest has become difficult to ignore. According to IDC, global spending on AI infrastructure touched $318 billion in 2025 and is expected to reach $1 trillion by 2029. Boards are asking management teams about AI strategy. Investors increasingly expect companies to articulate how they are using AI to improve productivity and customer experience. In many sectors, competitors are already moving from pilots to scaled deployments.
For CFOs, the challenge is not whether AI deserves investment. The challenge is deciding where capital should go when there are dozens of possible use cases competing for attention.
The Real Question Is Not ROI. It Is Opportunity Cost.
Many discussions around AI begin with return on investment. That is a reasonable starting point, but it is no longer the only lens through which these decisions should be viewed.
A more important question is: what happens if we do nothing?
Consider customer experience. Consumers increasingly expect instant responses, personalised interactions, and round-the-clock support. If one company continues relying entirely on manual processes while its competitors use AI to resolve queries faster and at lower cost, the financial impact may not appear immediately on a quarterly report. It often surfaces later through lower customer retention, slower growth, and rising service costs.
This is why opportunity cost has become a central part of AI investment discussions.
A useful example comes from customer service operations. Industry studies show that AI-powered support systems can handle a significant share of routine customer queries without human intervention. The value is not limited to cost reduction. Human agents spend less time on repetitive tasks and more time resolving complex issues that require judgment and empathy. Customers receive faster responses. Businesses gain capacity without proportionately increasing headcount.
The strongest AI investments usually have one thing in common: they solve a clearly defined business problem.
When businesses begin with technology and then search for applications, results are often disappointing. However, when they start with a business problem and look for solutions within the AI ecosystem, the probability of success increases at a rapid scale.
CFOs Need New Ways to Measure Value
Traditional capital allocation frameworks were designed for investments with relatively predictable outputs. AI usually does not follow this pattern.
A company may invest in an AI assistant to support customer-facing teams. Some benefits can be measured quickly through productivity improvements and lower operational costs. Others emerge gradually through higher customer satisfaction scores, reduced churn, and stronger revenue retention.
The challenge is that these gains do not always fit neatly into a 12-month ROI model.
A study from McKinsey indicates that organisations are beginning to see measurable financial benefits from generative AI, particularly in customer operations, marketing, software development, and service delivery. At the same time, there remains a significant gap between experimentation and enterprise-wide value creation. Many companies have deployed AI tools. Far fewer have established systems to measure their long-term contribution to business performance.
This is where finance teams have an important responsibility.
The objective should not be to demand certainty before approving investment. However, if that standard were applied rigidly, many transformational technologies would never have received funding. Instead, CFOs need clear milestones, measurable outcomes, and regular reviews that assess whether an initiative is delivering what it promised.
In practical terms, that means moving away from vanity metrics. The number of pilots launched or tools deployed says very little about business value. Metrics such as customer retention, revenue per employee, service costs, productivity improvements, and margin expansion provide a far more meaningful picture.
The companies creating the strongest results from AI are not necessarily spending the most money. They are making disciplined choices about where AI can create measurable business impact and where it cannot.
