Too many business leaders are asking the wrong questions of AI
Businesses still struggling to demonstrate meaningful business value are approaching a “cliff edge”, according to Lindsay Ratcliffe, chief innovation and transformation officer at CreateFuture. This is an inflection point at which the laggards will have little chance of keeping up with those competitors that are already generating ROI through AI platforms; and she anticipates it will happen within two years.
Two very different perspectives on AI have emerged in recent months: Forbes’ latest CxO Growth Survey found that the C-suite views AI-driven technology as the single biggest growth opportunity, with 95% planning to increase AI investments this year. However, PwC’s AI Performance Study suggests that a mere 20% of organisations are hoovering up any significant AI-driven returns.
The precise scale of the gap is open to interpretation, but the underlying disconnect between AI investment and realised returns has become difficult to ignore. As the leaders pull ahead, organisations that have yet to generate meaningful ROI are likely to find it much harder to close the divide.
The AI transformation problem
The pressure is real, and it’s coming from every direction: competitors moving faster, startups with nothing to unlearn, boards asking harder questions, shareholders wanting returns, customers expecting more… the list goes on.
Most organisations aren’t short of AI initiatives; they lack an integrated strategy that ties business outcomes to people, process and technology together.
What actually matters isn’t a tighter AI strategy. In our experience, the key difference between those organisations delivering value and ROI, and those that have yet to see an impact, is a cohesive approach that makes systemic improvements at an organisational level.
That’s a big undertaking - and rightly so, given how much the future success of an organisation depends on it. Leaders who are pulling ahead refuse to let the scale of the challenge impede progress. They break it down into component parts and tackle these piece by piece.
Rather than chasing a standalone ‘AI strategy’, these leaders lean into a clear business strategy powered by AI.
Crucially, success isn’t defined by the tech you buy, but by the questions you ask, and these start with two fundamentals. First, which core organisational challenges do we need to solve? The second has to be in what order, because trying to tackle everything at once is a fast track to failure.
What to prioritise
Layering AI on top of a problem - be that broken processes or a data swamp - doesn’t fix the problem, it only accelerates it. The smartest move is to step back and use AI as a diagnostic tool to understand what underlying issues, whether technical or structural, need to be addressed first. It can be equally useful to ask the people who use the systems every day, employees and customers, to identify what their pain points are.
Once the core challenges have been shortlisted, split them into foundations and initiatives. Foundations are the unglamorous work, data infrastructure, platform engineering, domain modelling, that rarely shows immediate ROI but determines how effective everything else can be. These aren’t optional extras to cut when budgets tighten. They’re gated by one question: does this de-risk multiple other initiatives? If yes, it stays in, regardless of payback time.
Initiatives are where prioritisation earns its keep. Multi-year big-bang transformation is high risk and hands nimble competitors room to extend their lead. The pragmatic approach is quick wins that turn cynics into champions, building momentum that compounds.
The formula for picking them
ROI captures total value, revenue, cost to serve, speed. Confidence reflects how solid that estimate is, not how exciting it sounds. Effort is the resourcing and complexity required. Score every initiative, then gate for delivery: can it prove impact in a 30 to 90 day sprint? If yes, it’s a quick win. If not, it may still be worth doing, just judged on its own terms.
Skip this segmentation and the maths misleads. Foundational work always looks like a bad initiative on paper: high effort, indirect ROI, low near-term confidence. Score it against quick wins with the same yardstick and you’ll defund the thing your quick wins depend on.
How to bring people with you
Theoretically, all this is fine. But the biggest hurdle is often making AI relevant and tangible to the people expected to use it. Turning ambition into practice requires people to change how they work, a behaviour shift that rarely happens without resistance. That makes trust the defining issue, from resetting the vision at the top to reshaping everyday workflows.
That trust must be earned. Showing that AI is a net win for the workforce and not just the balance sheet demands a high level of transparency and clear communication. A critical leadership skill is to actively listen to anxieties and prove that AI is built to augment. Appointing a chief transformation officer who understands people, operations, technology and finance will help to balance the strategy, communicate the change at the right time in the right way, and measure impact.
Equipping people emotionally for change is one part of the equation, the other is giving them the skills they need to evolve alongside the business. Upskilling programmes start with tailored training to help people at all levels get to grips with the new AI strategies, their applications and how they can augment roles, not just replace them.
Cultural change takes time to embed and it’s an ongoing process, not a one-off town hall or a single training day. What sustains it is a formalised learning framework, one that’s resourced and mentored so that best practice evolves alongside the tools and use cases. This needs to be supported by a clear communications framework that covers what’s changing, why, and what it means for each team. And it needs to be repeated consistently at frequent intervals and across multiple channels.
Take away the noise, and the reality checks for both leaders and employees are this: AI isn’t hype, it’s the closest thing to genuine step-change we’ve seen in decades and its impact is already reshaping every dimension of the enterprise.
But that impact only lands if it’s learned and applied properly. AI is a tool, not a shortcut. Mastering it is what separates transformation from theatre. In a market accelerating fast, becoming AI native isn’t optional, it’s what determines who survives the coming era and who’s left trying to explain what happened to it.
Ultimately, this has never been about adopting technology; it’s about delivering business outcomes. The gap between companies compounding value from AI and those falling behind is widening – and the window to close it is narrowing. Within two years, the divide will be clear between businesses that have rebuilt their strategy around AI and pursued transformation programmatically, and those left with a collection of disconnected initiatives delivering little strategic return.
The question isn’t whether you’re using AI. It’s whether your efforts are compounding value or simply accumulating activity.
