Cloud and AI investments do not automatically create business value. That was the verdict from a recent executive roundtable hosted by TechCentral in partnership with Cloud On Demand and Microsoft on 3 September 2026, where C-suite executives and senior technology leaders examined what they called “The Margin Illusion”: the assumption that adopting cloud and AI translates automatically into returns.

The room dispelled that notion quickly. Technology is not the answer. It is the enabler.

A recurring theme was that organisations rush to deploy cloud and AI before defining the business problem they are trying to solve. The pressure to innovate, stay competitive or show progress turns technology into the objective rather than the means of reaching it.

Participants agreed that successful transformation starts with understanding the outcome being sought: revenue growth, operational efficiency, customer experience or risk reduction. Organisations that set clear business objectives before deployment realise more value. Technology without a defined purpose rarely delivers the returns executives expect.

This led to a broader discussion about the foundations of transformation. While cloud and AI dominate boardroom conversations, executives repeatedly returned to three interconnected pillars: people, processes and data.

People were the pillar most frequently overlooked. Organisations invest heavily in platforms and infrastructure but underestimate skills development, change management and organisational readiness. Technology can be implemented in weeks. Changing behaviour and embedding new ways of working takes considerably longer. Ignoring that reality is an expensive mistake, however advanced the technology.

Measuring value

The discussion also challenged conventional thinking about value itself. While financial returns remain important, executives questioned whether value should be measured exclusively through cost savings or revenue generation.

Data sovereignty emerged as the clearest example. Investments in governance, compliance and data localisation rarely produce immediate financial returns, but they reduce risk, strengthen customer trust and build a more resilient platform for growth. Several participants argued that governance is not merely a compliance obligation but a value driver that protects the organisation from future harm.

This expanded view of value carried into the conversation on cloud dependency and vendor lock-in. Cloud platforms have enabled unprecedented speed and scale, but organisations at the table described growing dependence on particular technology ecosystems.

Few organisations intentionally create strategic dependency. It emerges over time as platforms become embedded in critical business operations. Executives weighed innovation and convenience against long-term flexibility and control. Portability, exit strategies and keeping options open are now part of the investment discussion, not an afterthought.

If there was one topic that generated near-unanimous agreement, it was data. Participants consistently highlighted data quality as both a governance imperative and a prerequisite for meaningful AI outcomes. An AI system can only be as effective as the information that underpins it. Poor-quality data produces poor-quality outcomes, however sophisticated the model.

Several executives went further, describing institutional data as one of the most valuable forms of intellectual property an organisation owns. Customer insights, operational knowledge and proprietary datasets are among a company’s most sustainable competitive advantages. The quality, governance and ownership of that data matter more than the AI tools deployed on top of it.

The conversation also covered cloud wastage and extracting more value from existing investments. Rather than focusing only on reducing unused capacity, participants explored how underutilised resources can be repurposed for new business opportunities and future growth initiatives.

That surfaced a bigger insight: pure cost reduction is an incomplete measure of success. Cost optimisation still matters, but value also shows up as agility, resilience, innovation and risk reduction. Traditional financial metrics alone do not capture much of what cloud and AI create.

As the session drew to a close, one theme continued to surface: uncertainty.

Despite real progress in cloud and AI adoption, there is no universally accepted framework for measuring value, because value depends on perspective. CEOs focus on growth and competitive advantage. Boards prioritise governance and long-term sustainability. CISOs view value through resilience and risk reduction. Business leaders focus on productivity, operational efficiency and customer outcomes.

Value is contextual

The most important conclusion of the session, then, was not that cloud and AI create value but that value itself is contextual.

There is no one-size-fits-all model. Organisations that succeed will be those that clearly understand what outcomes matter most to their business and align their technology investments accordingly.

That is the real lesson behind the margin illusion. The challenge is not proving that cloud and AI work. It is understanding what value means for your organisation, and making sure technology, people, processes, governance and data all work towards the same objective.