Turning AI adoption into strategic decision advantage

Artificial intelligence is quickly widening what organisations can analyse, but that does not mean they will automatically make wiser choices. McKinsey’s 2025 global AI survey found that 88% of respondents said their company now uses AI in at least one business function, up from 78% a year earlier, yet only 39% reported any EBIT impact. The figures suggest a widening gap between adoption and value creation, and they underline a broader point: more analytical capacity is not the same...

Generative AI is making analysis cheaper, faster and more widely available across the enterprise. More employees can interrogate data, test assumptions and produce insights at scale. But an organisation is not improved simply because it can produce more charts, models and summaries. Without the right decision structures, AI can just as easily accelerate old habits as replace them.

That is the central risk for leadership teams. Companies may invest heavily in AI, analytics and specialist talent while leaving untouched the authority structures, incentives and routines through which decisions are actually made. In that case, new tools can create more competing interpretations, more confidence in weak assumptions and more confusion over who is responsible for acting on evidence.

The strategic question, then, is not only how to adopt AI, but how to turn analytical capacity into decision capability.

A useful starting point is not technology, but the decisions an organisation most needs to improve. Which choices most affect customer value, capital allocation, risk, growth and operating performance? Where is judgment being exercised without enough evidence? Where does analysis already exist but fail to shape action? And where are analytical resources disconnected from the business problems they are meant to support?

Once leaders can answer those questions, talent planning becomes more than a hiring exercise. Technical skills still matter: modelling, forecasting, statistical analysis, visualisation and fluency in tools such as R, Python and other specialist platforms remain important. But technical ability alone is not enough. Organisations also need people who can interpret quantitative evidence, persuade colleagues, translate findings into commercial action and work across functions where evidence must meet judgement.

That is because analytics is not just a technical discipline; it is an enterprise capability. If analytical leadership sits only with the CIO, CTO or data team, evidence may remain peripheral to the real decisions that shape performance. Senior leaders across the business determine whether analysis becomes part of normal management or stays trapped in reporting cycles. They decide what gets measured, which issues receive attention, when evidence is deemed sufficient and whether data is used to challenge assumptions or merely to confirm what has already been decided.

Culture determines how much authority evidence really has. At an enterprise level, culture is not just a matter of morale or values statements. It is reflected in what leaders monitor, how they respond when things go wrong, where they direct resources, what behaviour they reward and who progresses through the organisation.

The things executives repeatedly ask about send a powerful signal. If a metric is routinely reviewed and challenged, it matters. If it is ignored, it may be strategically invisible, no matter how important it is. Equally revealing is what happens under pressure. When markets shift, performance slips or a major initiative disappoints, leaders can either engage with evidence that challenges their assumptions or retreat to hierarchy, intuition and familiar narratives. Organisations tend to learn more from those moments than from strategy presentations.

Budgets tell their own story too. Analytical ambition without enough time, access, technology and decision authority remains aspiration rather than capability. Likewise, leaders set the tone by the way they use evidence themselves. An executive team that demands discipline from others but relies mainly on assertion or selective information creates a contradiction that employees quickly notice. Judgment still matters, but it should be informed, and at times corrected, by evidence.

Rewards are equally important. An organisation may say it values analytical rigour, yet still promote speed over scrutiny, certainty over inquiry or agreement over constructive challenge. Over time, people learn whether raising inconvenient evidence improves decisions or damages their prospects. That lesson often matters more than any formal analytics initiative.

Hiring, promotion and succession decisions also shape the culture. The capabilities that are advanced through the organisation reveal what it truly values. If analytical judgement is strategically important, it should be visible in the criteria used to appoint people to greater responsibility.

Even so, there is no single correct way to organise analytical capability. Some resources need to sit close to the business to understand operating realities and focus on the decisions that matter most. But too much decentralisation can fragment standards, duplicate effort and prevent learning across teams. Too much centralisation can create the opposite problem: technically strong analysts who are increasingly detached from the decisions they are meant to improve.

The best structure depends on strategy, maturity, scale and the types of decisions being supported. The real governance challenge is to keep analytical expertise close enough to influence action, while connected enough to preserve consistency, learning and an enterprise-wide view.

That distinction matters because data, analytics and AI expand what organisations can know, but they do not determine what those organisations will do. Decision rights define who has authority. Accountability defines who carries consequences. Incentives shape what gets noticed. Culture decides whether evidence can challenge power. Resource allocation determines where capability grows. Organisational design decides whether insight is linked to action.

That is why the return on AI will not depend on technology alone. It will depend on whether organisations are designed to convert intelligence into judgement, coordinated action and learning. For many executives and boards, the more important question is no longer whether they have enough data, but whether their leadership system is built to use it well.

As McKinsey’s latest research suggests, AI adoption is spreading quickly, but value capture is still uneven. For some organisations, the gap between analytical capacity and decision capability will remain a drag on performance. For others, closing that gap may become a lasting advantage.

Source: Noah Wire Services