For years, boards have been approving multi-million-dollar AI and digital transformation projects. Yet many are failing, not because of the technology, but because of a foundational flaw management isn't talking about—a flaw so profound that, according to the MIT Sloan Management Review, the cost of insufficient data is already consuming between 15% and 25% of total revenue for most companies. This isn't a minor operational drag; it's a direct tax on the top line. For directors and investors, it’s time to start asking more complicated questions, because the most significant risk to your company’s future isn't the failure to invest in AI, but the quality of the data that renders those investments worthless.

In boardrooms across the globe, a familiar ritual plays out. A presentation, rich in the vocabulary of Industry 4.0—Artificial Intelligence, digital twins, and predictive analytics —culminates in a significant capital expenditure request. The promised returns are astronomical: radically improved efficiency, the elimination of unplanned downtime, and a decisive competitive edge. The board, fulfilling its duty to drive innovation and shareholder value, approves the investment.

Yet, for a troubling number of these companies, the promised revolution never arrives. The projects stall in "pilot purgatory," the ROI remains a rounding error, and the organisation is left with a costly hangover of disillusionment. When directors ask for an explanation, the answers are often vague, citing the complexity of AI or the challenges of integration. The real culprit, however, is a far more fundamental and damning failure —one that rarely makes it into a PowerPoint slide: the company’s data is junk.

This is the "data delusion," and for any board director or investor, it should now be considered a primary strategic risk. The failure to establish a high-fidelity data foundation is not a mere technical oversight; it is a profound business malpractice that wastes capital, destroys shareholder value, and leaves a company dangerously vulnerable. The most significant liability in your organisation today may not be debt or litigation; it may be the vast, unacknowledged swamp of low-quality data upon which your entire digital strategy is being built.

The Fiduciary Duty to Scrutinise "Digital-Washing"

As stewards of the company, a board's fiduciary duty extends beyond financial oversight to strategic risk management. In the digital age, this must include a rigorous scrutiny of a company's data infrastructure. The market is currently rife with "digital-washing"—the practice of using the hype around AI to mask a lack of genuine capability. Companies boast about the petabytes of data they collect, but they are silent on its quality.

This is where directors must become more discerning. The volume of data is a vanity metric; the fidelity of data is what drives value. A company collecting averaged vibration data every 15 minutes from a critical asset is not in the same league as a competitor capturing the full, high-frequency vibration waveform from that same asset at regular intervals and on exception if any reading materially changes between scheduled readings. The former collects digital noise; the latter collects actionable intelligence. The former can create a pretty dashboard; the latter can prevent a multi-million-dollar failure.

Approving an AI project without first auditing the underlying data fidelity is the equivalent of approving the construction of a skyscraper without conducting a geological survey of the foundation. It is a dereliction of strategic oversight.

Five Questions Every Director Must Ask a CEO

To cut through the digital washing and assess the true health of your company's digital strategy, directors need to move beyond high-level promises and ask probing, specific questions. The next time a digital transformation budget is on the agenda, consider this your essential questionnaire:

1. "Show me the data. Not the dashboard, the raw data. What is the actual granularity of the information we are feeding our AI models?" This question immediately shifts the focus from the superficial output (the graph) to the fundamental input—demand to see the difference between the low-resolution averages and the high-fidelity waveforms.
2. "What percentage of our analytics team's time is spent cleaning data versus analysing it?" This is a powerful diagnostic for the health of your data ecosystem. If, as many studies suggest, that number is approaching 50%, your company doesn't have an analytics program; it has a costly data janitorial service.
3. "How have you engineered our systems to be reliable at scale? Specifically, how do we prevent data collisions in dense sensor environments?" This technical question gets to the heart of scalability. Any vendor can make a pilot with 10 sensors work. A credible strategy must account for the chaotic reality of a factory with 10,000 sensors. If management cannot answer this, their plan is built for the lab, not the real world.
4. "What is our strategy for unifying data from different operational silos? Are we buying point solutions or building a foundational platform?" This reveals the long-term vision. Buying a collection of disparate, single-purpose products is a recipe for a costly integration nightmare. A true strategy focuses on building a single, modular data backbone that can serve multiple applications over time.
5. "Can you quantify the cost of a single, critical asset failure that our current system failed to predict?" This final question brings the discussion back to tangible value. Frame the investment in a high-fidelity data foundation not as a cost, but as an insurance policy against a precisely defined, multi-million-dollar risk.

Data as a Core Asset

The companies that will dominate the next decade will be those whose boards recognise that data infrastructure is not an IT expense. It is a core strategic asset, as critical as the company’s factories, intellectual property, or brand.

Therefore, the investment in a high-fidelity data platform must be evaluated differently. It is foundational. It enables not just one application, but all future AI and analytics initiatives. It is the bedrock upon which future efficiency, innovation, and resilience will be built.

As a director or investor, your role is to look beyond the immediate horizon. The pressure to "do something with AI" is immense, but the risk of doing it wrong is catastrophic. By demanding a culture of data discipline and prioritising the investment in a high-fidelity foundation, you are not just mitigating a hidden liability. You are steering the organisation toward a future of genuine, sustainable, and defensible value creation.

By Gert Botha

Keywords: AI, IoT, Digital Twins

Follow Us On