AI, capability and pace

Across Ireland and globally, we see organisations embedding AI at pace. It is starting to shape how work gets done, how decisions are made, and how services are delivered. Yet when leadership teams are asked a fundamental question - what value are you actually getting from AI? The answer is often unclear.

As KPMG’s Rory Timlin observes, this lack of clarity is becoming one of the defining leadership challenges of our time.

Article highlights:

Pace of change

Many organisations are adopting AI at pace – but still struggle to articulate the return
Pace constrained by the organisation

Fragmented AI use across tools, teams, and platforms is making value harder to see – and govern
Supporting roles

How leading organisations manage AI value over time

The AI value challenge leaders are facing right now

In our work with clients, we consistently see tangible benefits emerging from embedding AI. The challenge is that value is rarely defined, measured or governed in a way that reflects how AI is actually being used across the organisation. As a result, investment decisions are made, tools are deployed, and activity accelerates - but confidence in return does not.

That gap matters. Without a clarity on value, AI risks becoming a cost driven by hype and competitive pressure, rather than a deliberate capability anchored in business outcomes. For leaders, the question is no longer whether to adopt AI, but whether its value is being actively shaped - or passively assumed.

In light of that, many organisations are now examining their strategies through AI consulting, as a way to bring greater clarity to where value should be created, how it should be measured, and how ambition aligns with organisational maturity.

Why is AI value being missed?

For most organisations, the problem is not that AI isn’t working. It’s that value is being discussed at the wrong level.

Enterprise AI adoption is happening in multiple ways at once. It is embedded in purchased software platforms. It underpins personal productivity tools. It enables bespoke initiatives built by teams trying to solve specific local problems. Each of these can deliver benefits – but they rarely roll up neatly into a single value story.

As a result, leaders are often left with fragmented signals. Productivity gains are anecdotal. Efficiency improvements are assumed rather than measured. Revenue growth is aspirational. Investments are approved, but success criteria are vague.

When AI urgency outpaces value

This is compounded by the pressure many are feeling from the boardroom. The push to “do something with AI” is often driven by the fear of being left behind, rather than a clear articulation of where value should come from. That top‑down pressure can accelerate adoption initiatives, but it can also blur accountability for outcomes.

If all this feels familiar, you are not alone. It is a pattern we see repeatedly across organisations at different stages of AI maturity.

In that environment, AI activity increases – while confidence in return does not.

The risk of unclear AI value measurement

The risk of unclear value measurement is not just a financial metric. It becomes an organisational challenge.

Rory Timlin

Partner, Management Consulting

KPMG in Ireland

Consider the points below:

Scaling AI initiatives

When value is poorly defined, AI initiatives struggle to scale. Teams can demonstrate local wins but cannot justify broader investment.
Uncertainty causing delays

Foundational issues required for scaling AI value, like data quality, skills, process transformation, are deferred because the business case feels uncertain.
Fatigue and inertia

Over time, this creates fatigue. You might begin to question whether AI is delivering on its promise, and whether further investment is justified. What began as ambition turns into inertia.
Measuring the wrong thing

Without clear value-framing, organisations default to measuring what is easiest – often short‑term efficiency – rather than what actually matters.
Quantifying capabilities

Customer experience, decision quality, resilience and workforce capability are harder to quantify and drive longer term value. Ignoring them distorts the picture of impact.

How can leaders think differently about AI value?

Leaders need to get comfortable with three realities.

Value is contextual, not generic

AI does not create value in the abstract. It creates value in specific business processes, decisions and outcomes. Organisations that struggle to measure return often start with the technology, rather than the outcome they are trying to achieve.
Productivity is only one dimension of value.

Time savings matter. Cost reduction matters. But many of the most meaningful benefits of AI sit elsewhere, like in better prioritisation, improved insight, accurate first-time response and reduced error. If leaders only look at headcount or immediate savings, they will miss much of the return.
Value cannot be separated from maturity

Organisations at different stages of AI adoption should not expect the same returns. Early efforts are often about learning, capability building and data readiness. Expecting transformational impact too soon is unrealistic and leads to disappointment.

At its core, this is not a measurement problem alone. It is a leadership framing problem. Defaulting to cost reduction only can force a mindset of using AI as an efficiency tool, inserted into how you are working today.

Leading AI adopters are thinking bigger, and asking “what are we trying to achieve and how might we work differently with AI at the centre of how we operate?”

How can you define the value in AI?

The actions required are practical, not theoretical. They are about clarity, focus and intent. They start by asking sharper questions about what success actually looks like in your organisation.

Focus on outcomes, not tools

Before approving or expanding AI initiatives, be clear on what success looks like. That means anchoring AI application to specific business outcomes, such as improved cycle times, better customer insight, reduced risk, and enhanced decision quality - and agreeing upfront how progress will be assessed and measured.
Break value into measurable initiatives

Large AI ambitions need to be decomposed into smaller, testable initiatives. Each should have a clear hypothesis, defined metrics, and a review point. This reduces risk and makes value visible incrementally, rather than relying on a single end‑state business case. Create a portfolio of targeted initiatives to focus AI investment and avoid fragmenting and diluting effort.
Invest in foundations that enable value

Data quality, process clarity and workforce capability are not optional extras. They are prerequisites for sustainable return and the rails for AI scale. Organisations that under‑invest here often see impressive pilots that simply fail to scale. Foundational investment may not deliver immediate ROI but is a prerequisite for unlocking future value.

How best can you show AI value?

Organisations that are making progress tend to share a few characteristics.

They are explicit about where AI should create value – and where it should not.
They align initiatives to business priorities, rather than allowing experimentation to proliferate unchecked.
They accept that some early activity is about learning, not immediate payback.
They invest in the foundational capabilities for scale across data, technology, workforce transformation and governance.

Crucially, they review AI value regularly. Not to prove success at all costs, but to refine focus, stop initiatives that are not delivering, and double down where impact is emerging.

Value becomes something that is proactively managed, not hoped for.

AI expectations to watch in the coming months

Pressure on AI value will intensify. Boards and executives are increasingly asking harder questions about return. Regulatory and stakeholder scrutiny is rising. And as AI becomes more deeply embedded, the cost of unclear outcomes is likely to grow.

Waiting for perfect measurement frameworks is not the answer. Progress comes from disciplined experimentation, honest evaluation and leadership willingness to adjust course.

Making AI value intentional

AI is already part of the organisation. The question is whether its value is being deliberately shaped or passively assumed. You don’t need certainty to act. But you do need clarity on what you are trying to achieve – and how you will know if it is working. If you get this right, you will realise the true value of AI.

For organisations wanting to unlock that value, KPMG’s AI consulting team can help you harness the full power and potential of AI with speed, agility and confidence.