Digital twins have become a symbol of modern, data-driven operations. They promise optimisation through simulation, precision through models, and control through visibility. They look convincing, behave predictably, and produce outputs that feel objective.
That sense of objectivity is exactly what makes them dangerous when they are built on incomplete or poor-quality data.
Most digital twins do not fail because the models are badly designed. They fail because they create confidence where certainty does not exist. A simulation can be mathematically sound and visually compelling, yet still be detached from what actually happens on the shop floor. When that happens, optimisation does not merely underperform. It quietly pushes decisions in the wrong direction.
When data creates false confidence
There is a fundamental difference between having no data and having bad data. When data is missing, teams know they are making assumptions. Decisions are framed as estimates, uncertainty is explicit, and judgement remains visible.
Bad data removes that awareness. It presents assumptions as facts.
Low-accuracy, high-latency, or inconsistent location data rarely looks obviously wrong in a dashboard or digital twin. It looks precise enough to trust. That trust is what makes it dangerous. Once false facts enter a system, they cascade outward. Layout changes, automation investments, safety policies, and workforce decisions all begin to optimise a reality that does not exist. The more convincing the visualisation, the harder it becomes to question the conclusions behind it.
In optimisation, being confidently wrong is worse than being consciously uncertain.
Why optimisation initiatives stall
Many optimisation initiatives stall at the same point. This is often explained as resistance to change, lack of discipline, or cultural inertia. In reality, a more uncomfortable explanation is frequently closer to the truth: the system cannot see deviation.
On the shop floor, this often surfaces as a simple reaction to new process improvements: “That doesn’t make any sense.” This response is commonly dismissed as emotional or defensive. In practice, it is often reality pushing back against theory.
People resist change when the system asks them to work in ways that feel less rational than before. When improvement initiatives are based on assumptions rather than observation, operators recognise the disconnect immediately. What looks optimal in a model can feel illogical, inefficient, or even counterproductive in practice. This is not a cultural failure. It is a visibility failure.
Clean models, messy operations
Digital twins tend to optimise for a simplified world: shortest paths, ideal flows, perfect timing, and uninterrupted execution. Real operations do not work that way. They are full of detours, delays, micro-decisions, interruptions, and workarounds.
These deviations are not edge cases. They are the system.
When a digital twin cannot see them, it does not eliminate them. It hides them. KPIs remain stable while inefficiencies accumulate underneath. Over time, the gap between what systems believe is happening and what people experience on the floor grows wider, even as dashboards continue to look reassuringly precise.
Why RTLS quality matters more than RTLS presence
At this point, many organisations conclude that they need more data and turn to RTLS. That decision alone is not enough.
Having RTLS is meaningless if the data it produces is not trustworthy. Accuracy, latency, and reliability are not technical details; they determine whether digital twins improve reality or degrade it. Inaccurate or delayed location data does not merely reduce insight. It introduces structural error into every system that depends on it.
In this context, bad RTLS is worse than no RTLS at all. It replaces conscious assumptions with unconscious falsehoods, and those falsehoods quickly become embedded in decisions that are difficult and expensive to reverse.
A different way to interpret resistance
When optimisation efforts fail, the instinct is to blame people, change management, or process discipline. A more honest interpretation is often simpler. If the system cannot see how work actually happens, it cannot improve it. If the digital view drifts away from physical reality while still appearing precise, optimisation becomes theatre.
Digital twins are not the problem. RTLS is not the solution by default.
The real question is whether the digital view is continuously anchored in what actually happens, or whether it is slowly drifting away while retaining the appearance of certainty. That distinction determines whether optimisation leads to meaningful improvement or to increasingly confident mistakes.
Is your data ready for optimisation?
If your digital twin looks convincing but you are not sure it reflects reality, it’s time to examine the data that underlies it.
Let’s take a closer look at the quality of your real-time visibility.
