Making the case for AI: What your board and CFO need to know

AI is moving at extraordinary speed. I see that reflected in almost every client conversation: there is real ambition, but also a very practical question about how to turn that ambition into measurable business value.

For many AI leaders and CIOs, that question quickly becomes a boardroom one. Before asking for further investment, they need to be able to explain what problem they are solving, what value the organization should expect, what it will take to scale, and why the investment makes sense financially.

The technology will continue to evolve, but for business leaders the challenge is already much more immediate: where do we start, how do we build a credible business case, and how do we avoid investing in something that never gets beyond a promising experiment?

These are the questions I hear from clients time and time again. They are also the questions boards and, in particular, CFOs are asking as AI investment grows: What will this actually deliver? How will we measure success? What are the risks? And what needs to be true before we scale?

I recently explored these questions with Tom Winstanley, Managing Director of Technology, New Ventures and Innovation at NTT DATA UK, and Henrik Blokhuis, Senior Director and CTO: Partner Organization, Europe, Middle East and Africa, at Cisco.

The message was clear: Start with a business problem that is worth solving, be clear about how success will be measured, and then make sure the foundations; infrastructure, security, data, governance and people, are ready to support it.

Build the foundations to move fast

Regardless of the industry you’re in, there is increasing urgency to move quickly on AI. The investment is already significant. GartnerÒ research indicates that “worldwide spending on AI is forecast to total $2.59 trillion in 2026, a 47% increase year-over-year.”*

But moving quickly does not mean skipping the fundamentals. In fact, sometimes getting the right governance, security, data and infrastructure in place can feel like you are slowing down at the start. The reality is that this is what allows you to move faster later, with fewer surprises, less rework and a much better chance of scaling successfully.

When those foundations aren’t in place, organisations often hit obstacles further down the line, particularly when they try to move from a promising pilot into the reality of the enterprise environment.

From a technology-readiness perspective, there are three areas that should be part of the conversation from the beginning:

1. Infrastructure has to be ready for the reality of the organization. An AI use case may work brilliantly in a sandbox, but the real test is whether it can operate securely and reliably in your existing environment. As adoption grows and more AI agents operate across the business, that demand will only increase.
2. Security needs to be designed in, not added later. AI agents need identities, access controls and clear management, just as people do. Organizations need to know what agents can access, how they are behaving and how they will contain the impact if something goes wrong.
3. Data and governance matter just as much. Fragmented data, organizational silos and weak governance can quickly become the difference between a promising use case and one that delivers measurable business impact.

And these cannot be separate conversations. The top-down business ambition and the bottom-up reality of infrastructure, data, processes and people have to come together from the start.

• ALSO READ → From AI ambition to real impact: Why infrastructure makes the difference

Start with the value, not the technology

One of the most important questions is also one of the simplest: is this a problem actually worth solving? AI should not become an answer looking for a problem. Before investing, be clear about the business outcome you are trying to change and why AI is the right way to change it.

In short, start with the business problem, not the technology.

Many organizations started with proofs of concept, smaller pilots and productivity use cases. Those experiments have been useful, but the conversation is now moving into the core of the business: where can AI genuinely improve how the organization grows, operates or serves its customers?

That means looking at the core value chain. In insurance, it could be underwriting and claims. In banking, loan origination or back-office operations. In telecommunications, autonomous networks. The opportunity will be different in every industry, but the principle is the same: Focus AI investment where it can make a measurable difference to the business.

That also creates a much stronger conversation with the CFO. Instead of trying to justify an isolated AI experiment, you can show how the investment supports revenue growth, reduces cost, improves time to value or changes the performance of a business process that already matters.

Measure AI against the business

A Cisco survey found that only 32% of organizations had defined clear key performance indicators for AI success. That matters because, if you have not defined success at the beginning, it becomes very difficult to make an objective decision about whether to scale, redesign or stop.

The board and CFO need something more concrete than enthusiasm. They need to understand what will change in the business, how you will measure it and what evidence will justify the next investment decision.

The good news is that measuring AI success does not mean inventing an entirely new set of business metrics. If the business already measures time to quote, time to resolution, revenue growth or cost to serve, those measures should remain relevant. The question is whether AI is materially improving them.

There will also be useful leading indicators around adoption, enablement and training. But ultimately, AI has to connect back to the measures that already define success for the business.

Let your employees play

AI transformation depends on people as much as it does on technology. Training matters, but so does giving people a safe opportunity to experiment. When employees can try the technology for themselves, test ideas and learn from what does not work, it becomes much easier to move from fear or uncertainty to confidence.

That experimentation needs clear guardrails. Cisco took this approach early, combining hands-on use with education on responsible AI and even gamifying parts of its training to encourage employees to develop new ideas and use cases.

At NTT DATA, we launched an AI Academy more than two years ago for 150,000 employees. Everyone started with the fundamentals as an AI “white belt”, with opportunities to progress through yellow, green and black belts. We have also built communities of practice so people can explore how AI applies within their own part of the business.

The penny often drops once people start using AI for themselves. That is when they begin to see both what the technology can do for them and the role they can play in shaping how it is used.

AI: Success by numbers

Our CEO, Abhijit Dubey, uses a simple 1-2-3-4 framing that puts the scale of AI transformation into perspective: for every $1 spent on technology, think $2 on data, $3 on process and $4 on people.

I think that is a useful reminder because the technology naturally attracts most of the attention. But technology on its own does not create the outcome. The value comes from bringing the data, processes and people around it with you.

For me, the opportunity around AI is enormously exciting, but clients do not need more hype. They need confidence: confidence that they are solving the right problem, measuring the right outcomes, building for scale and bringing their people with them. Get those fundamentals right and AI has a much better chance of creating lasting business value.

This article includes contributions by Henrik Blokhuis, Senior Director and CTO: Partner Organization, Europe, Middle East and Africa, at Cisco.

• ALSO READ → The real AI opportunity: Reinvent how your organization operates

* Gartner Press Release. Gartner Forecasts Worldwide AI Spending to Grow 47% in 2026. 19 May 2026.

GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally and is used herein with permission. All rights reserved.