KYLE WINTERBOTTOM - CEO, Orbition Group

Beyond the Hype: 

What It Really Takes for CIOs to Deliver on Data and AI

For all the excitement surrounding artificial intelligence, many organisations are still wrestling with a far more fundamental problem. They have invested heavily in data, analytics platforms, governance programmes and specialist teams. Yet years into that journey, many boards are still struggling to answer a simple question: what value has all of that investment actually delivered?

According to Kyle Winterbottom, CEO of Orbition Group, that question sits at the centre of many of his conversations with CIOs, Chief Data Officers and executive boards. We have personally known each other for several years now, and he has built a reputation for saying the unfashionable thing in a market addicted to hype. So this conversation was always going to tread a fine line between cold, hard fact and hard-earned industry lessons. "If you were to underpin the problem statement," he says, "most organisations have invested far more money than they've seen in return from all of the work they've done with data analytics and AI." It is a statement that immediately cuts through much of the noise surrounding AI.

Winterbottom is not speaking from the perspective of a technology vendor. Orbition's business sits at the intersection of leadership, talent and transformation. The company works predominantly with FTSE 100 and FTSE 250 organisations, helping boards appoint Chief Data Officers, Chief Information Officers and the senior leaders responsible for delivering data, analytics and AI capabilities.

What has changed over the past few years is Orbition's own focus. "We've probably become the most known for executive search in the data analytics and AI leadership space," says Winterbottom. "Over the last 12 months, we've doubled down on that space." That decision came from recognising a recurring pattern. "We exist to try and solve one key problem," he explains. "Helping organisations to effectively fix what is fundamentally a leadership problem." It is a theme we return to repeatedly throughout our conversation.

Despite the pace of innovation and the growing excitement surrounding AI, many organisations continue to face longstanding challenges. The view from the Orbition camp is that the issue is often not a lack of technology or investment, but a failure to connect those investments to clearly defined business outcomes.

The Accountability Challenge

Over the past decade, the role of the Chief Data Officer has become established within many large organisations. Businesses recognised the growing importance of data and invested heavily in specialist leadership, dedicated teams and enterprise-wide transformation programmes.

In theory, the model made perfect sense.In practice, the results have often been mixed."Nine times out of ten, they bring somebody in with a flawed mandate around what they expect the role to be," says Winterbottom. The leader arrives with responsibility for building capability. Teams are assembled. Governance frameworks are introduced. Platforms are modernised. New reporting and analytics capabilities emerge.

On paper, much of the programme may appear successful. New teams are in place, governance frameworks have been established, and reporting capabilities have improved. The difficulty often emerges when organisations attempt to measure the commercial impact of those investments. "The people who are in the business using the data, the business unit leaders and the board, they're equally not accountable for driving better outcomes by using the data, but neither is the CDO because that wasn't the mandate they were given." Winterbottom describes this as an accountability gap and, in his view, it lies at the heart of many failed data programmes. A Chief Data Officer may successfully deliver everything they were asked to build, yet still struggle to demonstrate commercial impact because responsibility for achieving business outcomes was never clearly defined.

The result is a familiar cycle. Leaders are appointed. Significant investment follows. Capabilities are created. Expectations rise. Questions begin to emerge about value. Eventually, organisations reconsider the structure and start searching for alternative approaches. At the same time, Winterbottom is keen to emphasise that successful examples do exist. "There are quite a number of organisations out there that have delivered a ton of commercial value and ROI and massively improved business performance by using data analytics and AI more effectively." Examples of success are not difficult to find. The more useful question is what separates the organisations generating measurable value from those that continue to struggle despite significant investment.

Why the CIO Is Back at theCentre of the Conversation

One of the most visible developments in recent years has been the shifting relationship between the CIO and the CDO.Many organisations originally moved data leadership away from IT and established independent data functions. More recently, Winterbottom has observed responsibility returning to the CIO in a growing number of businesses. "We started out with the CIO in terms of where the CDO reported to, then it moved away for exactly the reason I just mentioned, and now it's moved back again for the exact same reason." While many organisations continue to maintain independent data leadership functions, Winterbottom has seen a growing number move responsibility for data and AI back under the CIO. In his view, that shift is often influenced by how boards perceive the role data should play within the organisation.

Many boards still view data primarily as a technology challenge rather than a business capability. When organisations struggle to demonstrate commercial returns from data investments, responsibility often gravitates back towards the executive leader already responsible for technology, infrastructure and enterprise systems.

In many cases, that responsibility has fallen back to the CIO. However, the rapid rise of AI has added further complexity to the discussion, creating new debates around ownership, accountability, and where responsibility for data-driven transformation ultimately lies.

AI Has Exposed Old Problems Rather Than Solved Them

The arrival of generative AI has transformed boardroom conversations over the past two years. Yet Winterbottom argues that one of AI's most significant contributions has been exposing weaknesses that already existed. "The whole LLM world has become a little bit more commoditised. Most people now are using ChatGPT or Claude rather than Google." As organisations explore how to deploy AI internally, they quickly encounter a familiar reality. "If we're going to use this internally, it's only as good as the data that we're feeding it." That observation has brought renewed attention to disciplines that previously struggled to attract executive interest. "Things like data governance probably didn't get the amount of time and investment that they should have historically. Now, all of a sudden, every organisation is scrambling to put loads of money into data governance so that they can get better outcomes with AI." While many data leaders see AI as an opportunity to demonstrate the strategic value of data, leadership teams are increasingly having to decide where responsibility for the agenda should sit and how that responsibility aligns with broader business objectives.

The Land Grab Around AI

Winterbottom describes the current environment as "a land grab". The phrase is deliberately provocative, but it captures a genuine reality. AI has become one of the most important strategic priorities in many organisations. Unsurprisingly, multiple leadership functions see it as falling within their remit. "The CIO obviously wants it. The CDO obviously wants it." Where ownership ultimately sits often depends on how the board views AI itself. "If it's just a tool that allows people to be more productive and more efficient, it tends to err on the side of the CIO because it's a piece of technology they can roll out." However, organisations that view AI as a driver of commercial performance often reach a different conclusion. "If it is something that can accelerate our speed and agility towards specific commercial goals and value levers, then it tends to lean more towards the CDO." These distinctions are becoming increasingly important as organisations develop their long-term operating models. The emergence of Chief AI Officers and Chief Data & AI Officers reflects an industry still working through fundamental questions around ownership, accountability and value creation. Winterbottom believes the answer is rarely universal. "I personally think that is massively contextual to the business and what the leadership of that business thinks about data analytics and AI."

Why There Is No Blueprint

Leadership teams frequently look for proven models. Successful transformation programmes are analysed. Operating structures are compared. Best practice frameworks are adopted. While Winterbottom believes organisations can learn from one another, he is sceptical about the idea of a universal blueprint. "At a very high level, there are probably some principles that can be followed." Successful organisations tend to align leadership mandates with business objectives. They hire differently. They structure teams differently. They approach accountability differently. Beyond those broad principles, however, similarities begin to disappear. "If you take it down a level to the execution of the strategy, I think there are often too many nuances and variables."

Industry, geography, organisational maturity, leadership philosophy and culture all influence how data and AI strategies should be designed. "There is no right or wrong answer to that question." That helps explain why strategies that work well in one organisation cannot always be replicated successfully in another. The businesses generating the strongest returns from data and AI are typically those that have aligned their approach to their own objectives, culture, operating model and leadership priorities rather than attempting to follow a prescribed formula.

Starting With Outcomes Rather Than Capability

One theme emerged repeatedly throughout our conversation. Many organisations begin their data and AI journey from the wrong starting point. "Most businesses start the wrong way around," Winterbottom suggests. The traditional approach is understandable - Leaders recognise that data should create value. They invest in platforms, governance frameworks, architecture and specialist talent. The expectation is that business value will emerge once the capability is in place. According to Winterbottom, that logic rarely delivers the desired outcome. Instead, organisations should begin by identifying the business goals they are trying to achieve. "What are the goals of the business? How do we measure our performance against those goals?" Only once those questions have been answered should leaders begin to define roles, capabilities, and investment priorities. Starting with business outcomes changes the conversation. The leadership mandate becomes easier to define, investment decisions become more focused, and organisations gain a much clearer understanding of the capabilities required to achieve specific goals. Most importantly, accountability can be tied directly to measurable business performance rather than activity alone. "When businesses do that and start at that end, it's very clear and obvious to see by who they hire, the type of mandate they give them and the scope of role that they give them."

The Reality of Legacy Data

Discussions around data quality are hardly new. Long before generative AI arrived, organisations were investing heavily in programmes designed to improve governance, standardisation and trust in their data. Winterbottom takes a more pragmatic view. "They are never going to get to the point where your data estate is clean." For large enterprises managing decades of systems, processes and acquisitions, perfection is not a realistic target. "We've all heard the phrase trying to boil the ocean. That's kind of what it is inside these big businesses." The organisations making progress are not attempting to solve every data quality challenge simultaneously. Instead, they focus on the data that matters most. If revenue growth is the objective, attention is directed towards the decisions, metrics and datasets that influence revenue.

The same thinking applies to operational efficiency, cost reduction or customer experience initiatives. Rather than attempting to solve every data challenge across the organisation, the focus should be on the information that directly supports the outcomes the business is trying to achieve. "These are the data sets we need. So it's imperative that these are clean." Focusing on datasets that directly support business objectives helps create a much closer connection between the board's priorities and the data function's priorities. Without that connection, it becomes difficult to demonstrate how improvements in data quality, governance or reporting contribute to the outcomes the organisation actually cares about. While executive teams are typically measured on revenue growth, profitability, customer retention, or operational performance, data leaders often report progress against entirely different metrics. "A CEO and a CFO are looking at each other going, 'Well, so what?'" It is a blunt observation, but one that many data leaders will recognise.

Reducing Time to Impact

This focus on outcomes shapes Orbition's own approach. While executive search remains a core part of the business, Winterbottom sees recruitment as only one component of the wider challenge. The ultimate objective is to reduce the time between hiring a leader and generating measurable business value. "A lot of HR teams measure things like the cost of hiring someone and the time it takes to hire someone." Orbition measures something different. "Time to impact." Achieving that often requires organisations to rethink the role before recruitment begins. "A lot of the work that we do at the outset is trying to bridge as much of the gap as possible at the strategic level."

That means understanding business objectives, identifying value levers and defining the outcomes a future leader will be expected to deliver. Technical expertise remains essential Winterbottom describes it as "table stakes". The differentiator is whether a leader can connect capability to commercial performance. That distinction increasingly determines whether organisations generate value from their investments or simply add another layer of complexity.

Looking Ahead

Predicting the future of AI is becoming increasingly difficult. The pace of change is accelerating. New models emerge almost weekly. Capabilities that seemed advanced six months ago quickly become standard. Even so, Winterbottom believes several trends are already becoming clear. The first is that the gap between organisations is likely to widen. "Anybody that isn't acting is probably going to get left behind." The second is that investment alone will not determine success. Organisations moving quickly without a clear understanding of outcomes may struggle just as much as those moving too slowly.

Finally, Winterbottom expects further convergence at the executive level. Traditional boundaries between CIOs, CDOs, Chief Digital Officers and Chief AI Officers are beginning to blur. "We're starting to see businesses look at whether they can amalgamate a lot of these roles together."

In some organisations, this may lead to the emergence of broader transformation or innovation leadership positions that combine responsibilities previously spread across multiple executives. "It feels like the CIO, the CDO, the Chief Digital Officer and the Chief AI Officer all rolled into one." Whether that model becomes the norm remains to be seen.

What seems increasingly clear, however, is that the next phase of data and AI leadership will be defined less by technology choices and more by organisational choices. While the pace of innovation shows little sign of slowing, Winterbottom believes the organisations that succeed will be those that remain focused on outcomes rather than technology alone.

The real test is not whether businesses can deploy AI, but whether they can create the leadership structures, accountability and commercial focus needed to turn that capability into measurable results. In the end, the future of AI may depend less on the models organisations deploy and more on the leaders they choose to deploy them.

Sound familiar to your business? Get in touch with Kyle and the team - They will more than likely have the answers.