Unlocking the Power of Internal Data:
A Call for Predictive and Proactive Facilities Management
Figure 1: Representation of internal FM data analytics. Image created using an AI research assistance tool.
Every facilities management firm is sitting on a goldmine it rarely mines. Work orders, maintenance logs, energy records, asset histories, these accumulate daily, yet from experience studying built environment organisations across New Zealand and beyond, most of this internal data is rarely fully utilised and sometimes locked in siloed systems. My research on internal data analytics keeps returning to the same finding: firms are more comfortable importing external benchmarks than trusting the intelligence of their own systems. That instinct is understandable, but it is also a strategic vulnerability. As digital transformation reshapes the built environment, the firms that lead FM’s next phase will be the ones that treat their own data as a genuine asset, not an administrative by-product.
From Reactive to Predictive: The Data Imperative
Most FM operations still run on the reactive or preventive logic of ‘fix it when it breaks, or service it on a fixed calendar regardless of actual condition’. Internal data analytics offers something better: predictive maintenance, where sensor readings and operational patterns are read together to anticipate failure before it happens. The gains are real and well within reach since most firms already run CMMS platforms capturing exactly the data that predictive work needs. In my own analysis, scheduling and cost data are usually the first things firms analyse well, but the ceiling rises sharply once data governance mature. Governance is what makes data consistent enough to inform decisions, which will translate to longer asset life, lower costs, more reliable service.
Digital Twins: Making the Invisible Visible
Digital twin technology gives FM teams a way to see what was previously invisible. A digital twin is a live virtual replica of a building or asset, continuously refreshed through sensor feeds, turning raw data into something you can watch, test, and interrogate. It facilitates monitoring conditions in real time, prioritising maintenance dynamically, trialling energy scenarios before touching the building or disturbing occupants. However, what research keeps underlining is that a digital twin is only as good as the data feeding it. Where CMMS data is integrated cleanly, firms get an adaptable intelligence platform, but where internal data management is poor, the initiative tends to stall before it delivers value.
Generative AI: Turning Data into Decision Support
Generative AI adds a further layer of capability, which is more accessible to the everyday FM practitioner than any tool before it. It lets teams interact with their own data conversationally, querying performance data and maintenance histories, drafting specifications, and generating compliance documents. FMers can operationalise generative AI at a pace that used to require a data science background. If trained on FM-specific datasets, generative AI can go further by producing failure mode libraries, energy benchmarks, and asset replacement recommendations grounded in a firm’s own history rather than external generic averages. The catch again is data quality. Without standardised metadata, these tools cannot scale, which is why consistent data protocols must come first.
Agentic AI: Autonomous Action at Scale
Beyond generative AI sits agentic AI. These are systems capable of autonomous, multi-step decision-making without someone standing over them at every turn. In FM, this could mean a system that monitors performance, detects an anomaly, raises the work order, schedules the technician, and closes the loop within one workflow. For large, complex estates, this is a genuine productivity lever. But agentic AI is unforgiving of poor data. My work with colleagues has repeatedly found that fragmented systems such as BIM, IoT, and ERP platforms that do not talk to each other, remain a major brake on analytics maturity in the built environment. Firms that invest now in integrated, well-governed data will be positioned to use agentic AI when it matures.
CMMS Data and the Robotics Frontier
CMMS data is also starting to power a new generation of robotic applications. Autonomous inspection robots rely on asset location and maintenance history to assess infrastructure that is difficult or dangerous for people to reach. Cleaning robots use occupancy data to optimise their own schedules, often anchored by the same digital twins discussed earlier. None of this happens automatically, thus CMMS records need to be structured, geocoded, and granular enough to give a robot something meaningful to act upon. Again, it brings us back to the same truth: internal data quality is the strategic prerequisite for almost everything FM firms want to do with technology.
A Call to Action
Here is the encouraging part, the infrastructure for this transformation already exists inside most FM firms. What is missing is not technology but leadership commitment to data governance, analytical capability, and a strategic vision that finally treats internal data as the FM sector’s most underleveraged asset. Our research had proposed a four-domain framework to guide this shift: governance frameworks, capability and culture development, alignment and maturity models, and emerging technologies with ethical integration. This structured pathway from intuition-based practice toward evidence-based, analytics-enabled management will see to internal data maturity levels that matter. Digital twins, generative AI, agentic systems, and robotics form a compelling technological ecosystem for FM, but that potential becomes real only when the data feeding it is reliable, standardised, and deliberately governed.
Organisations that build data analytics into their core capability, treating internal data as a strategic asset rather than a by-product of operations, see measurably better outcomes in forecasting, risk management, and governance accountability. The firms that act on this now, while the data foundations are still being laid, will define the next generation of FM service excellence.
By Professor James O.B. Rotimi, School of Built Environment, Massey University.
Related articles by the author:
