March 11, 2026 | Blog
Data-led maintenance in UK FM: Risk of inaction
Written by: Jevon Hughes, CEO & Co-Founder
The UK facilities management (FM) sector in 2026 stands at a pivotal juncture.
Scroll to discover more
Executive summary
Every year, the UK’s commercial buildings waste an estimated 54% of the energy they consume. Planned preventative maintenance schedules send engineers to service assets that are running perfectly while genuinely failing equipment goes undetected until it breaks. Over half of FM organisations report being understaffed, and 73% of teams resort to reactive firefighting on a weekly basis. Meanwhile, the regulatory environment has never been more demanding. The Building Safety Act, golden thread documentation requirements and tightening ESG obligations are raising the compliance bar faster than most FM operations can adapt.
This is not a cyclical downturn. It is a structural reckoning.
Facilities management is one of the last major industries to undergo meaningful digital transformation. While sectors such as logistics, manufacturing, financial services and healthcare have spent years embedding data, automation and artificial intelligence into their core operations, much of the FM industry still operates on spreadsheets, fixed schedules and institutional memory. That gap was survivable when the alternatives were immature or expensive. In 2026, it is no longer either.
AI is now commercially proven, readily deployable and improving at a rate that compounds quarterly. Technology companies, unburdened by legacy processes, labour constraints or the cultural resistance to change that characterises much of the FM sector, are already building platforms that can ingest building data, diagnose faults, optimise energy consumption and generate maintenance decisions faster, cheaper and more accurately than traditional service delivery models.
This whitepaper argues that the UK FM industry has arrived at a defining crossroads. Those who pivot toward data-led maintenance now will strengthen their operations, reduce costs and future-proof their businesses. Those who remain indecisive, waiting for perfect information, deferring investment, clinging to the status quo, will find themselves progressively displaced by organisations that treat data and AI not as optional enhancements, but as core infrastructure.
The risk of inaction is no longer theoretical. It is measurable, it is accelerating, and it is existential.
The last major industry to transform
Why FM’s technology lag matters now more than ever
Consider the trajectory of other industries. Manufacturing adopted predictive maintenance and digital twins over a decade ago, and production lines now self-diagnose and auto-schedule interventions in real time. Logistics companies route millions of deliveries daily using machine learning, adjusting dynamically to weather, traffic and demand. Financial services firms process billions of transactions through automated compliance and fraud detection systems that improve with every cycle. Healthcare is deploying AI for diagnostic imaging, drug discovery and patient triage at a pace that would have seemed implausible five years ago.
Facilities management, by contrast, remains largely tethered to the same operating model it has used for decades. Calendar-driven schedules, paper-based compliance, reactive interventions, and a dependency on tacit engineering knowledge that walks out of the door with every retirement. The SFG20 State of Facilities Management Report 2025 revealed that 75% of facility managers identified budget constraints as their biggest challenge, and 40% reported a budget decrease year on year. Skills gaps in compliance, digital capability and sustainability are reported by over 40% of organisations. Over half are understaffed.
The industry has been aware of these pressures for years. What has changed in 2026 is the convergence of three forces that make continued inaction genuinely dangerous.
The first is the regulatory ratchet
The Building Safety Act 2022, now firmly in its enforcement phase, demands that accountable persons maintain a golden thread of digital building information throughout the asset lifecycle. The Fire Safety (England) Regulations 2022 have sharpened accountability and raised the baseline for compliance. Information governance is moving from an aspirational goal to a measurable, auditable requirement. FM providers who cannot demonstrate real-time, traceable compliance data are carrying material risk.
The second is the AI acceleration curve
Artificial intelligence is no longer a proof of concept technology. It is commercially deployed, improving at compounding speed, and increasingly accessible to organisations of any size. The gap between what AI can do and what traditional FM operations actually do is widening with every quarter that passes.
The third is the arrival of technology-native competitors
The most significant threat facing the FM industry is not from within. It is from technology companies and PropTech platforms that view building operations as a data problem, not a labour problem. These organisations can deploy AI-driven maintenance, energy optimisation and compliance automation without the legacy overhead, cultural inertia or labour constraints that slow down traditional FM providers. They are faster, leaner, and they scale differently.
This convergence is what makes 2026 a crossroads, not merely another year of incremental change.
The cost of inaction
What standing still actually costs
Inaction is often framed as a neutral position, a decision to wait, to gather more information, to avoid risk. In practice, it is the most expensive decision an FM organisation can make. The costs compound silently, and by the time they surface, the damage is structural.
Financial Exposure
Traditional time-based maintenance generates enormous waste. Engineers are dispatched on fixed schedules regardless of whether assets need attention, consuming labour hours and materials on interventions that deliver no value. Simultaneously, assets that are genuinely degrading go undetected because the schedule has not yet triggered an inspection. The result is a paradox. Organisations are simultaneously overspending on maintenance and underprotecting their assets.
Reactive maintenance, the inevitable consequence of missed early warning signs, costs significantly more than planned intervention. Emergency callouts carry premium labour rates, expedited parts procurement, and the knock-on costs of operational disruption, tenant complaints and SLA penalties. Research indicates that 73% of FM teams resort to reactive problem solving on a weekly basis, with missed SLAs eroding both margins and client confidence.
Energy waste compounds the financial picture. With an estimated 54% of energy wasted across commercial real estate, buildings without intelligent monitoring are haemorrhaging operational expenditure every hour of every day. In a market where energy costs remain elevated and carbon pricing is tightening, this is not a marginal inefficiency, it is a balance sheet liability.
Regulatory and compliance risk
The regulatory environment in 2026 leaves no room for ambiguity. The Building Safety Act requires accountable persons to maintain golden thread documentation, a continuous, digital record of building information that must be accurate, accessible and audit-ready throughout the building’s lifecycle. Fragmented systems, incomplete records and unclear data ownership are now classified as material risks, particularly for higher-risk and complex buildings.
FM providers operating on legacy processes cannot produce this level of documentation reliably. Manual compliance tracking is inherently prone to gaps, delays and human error. When regulatory enforcement actions are resulting in fines of tens of millions of pounds, the cost of a compliance failure dwarfs the investment required to digitise maintenance and asset data.
The SFG20 State of Facilities Management Report 2025 report notes that while 72% of FM professionals identify compliance and safety as their top priority, progress toward meeting these standards is not happening at the speed or scale required. This gap between intention and execution is precisely where risk accumulates.
What data-led maintenance actually is
From fixed schedules to living intelligence
Traditional planned preventative maintenance relies on SFG20 instruction sets with frequency-based tasks executed regardless of actual asset condition. An air handling unit is serviced every twelve weeks whether it is operating flawlessly or showing early signs of bearing failure. A boiler receives its annual inspection on the calendar date, not when its performance data suggests it needs attention. This approach made sense when buildings produced no usable operational data. In 2026, it is indefensible.
Modern commercial buildings generate vast volumes of telemetry every minute. Building management systems, IoT sensors, energy metering platforms and CAFM systems continuously produce data on temperature, air quality, equipment status, occupancy patterns and energy consumption. The problem is not a lack of data, it is that this data is typically fragmented across multiple systems and vendors, creating silos that prevent meaningful insight.
Data-led maintenance solves this by providing a centralised intelligence layer that aggregates, normalises and analyses information from across the entire building environment. Once unified, layered intelligence can be applied. Deterministic rule-based diagnostic engines identify known fault patterns in real time. Machine learning anomaly detection surfaces subtle deviations that would be invisible to human inspection. And hybrid physics-informed neural networks predict remaining useful life and performance degradation trajectories.
How it works in practice
The platform generates engineering-grade asset health scores on a normalised 10 to 100% scale, starting at an 80% baseline. Green scores trigger PPM suppression, the system confirms the asset is healthy and the scheduled intervention is unnecessary. Amber or red scores accelerate interventions, routing the right engineer to the right asset with the right diagnostic context before a failure occurs.
Strategy 1
Recommended for rapid rollout, front-loads the suppression of PPM instruction sets, replacing them with 24/7 monitoring and rules engines. At month end, the service provider reviews triggered rules and authorises interventions that are automatically raised in the CAFM system. This approach delivers measurable savings from month one.
Strategy 2
Dynamic optimisation, adds live asset health scoring to enable triage and selective suppression. This strategy provides deeper operational intelligence and continuous model refinement while preserving full audit trails.
AI as an operational tool
Predictive maintenance
Predictive maintenance is one of the most valuable and immediately accessible applications of AI in the built environment. By analysing historical performance patterns alongside real-time telemetry, AI models identify indicators of asset degradation long before a failure occurs. Changes in vibration patterns, energy consumption differentials, temperature drift or abnormal runtime behaviour signal emerging issues that would be invisible to calendar-based inspection.
When these insights are integrated into facilities management workflows, maintenance teams intervene earlier, prevent failures, reduce downtime and extend asset lifespan. Critically, maintenance resources are deployed where they are genuinely needed, not where a schedule says they should be.
Anomaly detection
Buildings are complex systems, and inefficiencies develop gradually. Traditional alarm systems detect only extreme conditions, equipment failure or temperature thresholds being breached. They are binary. Everything is fine, or something has already gone wrong.
AI-driven anomaly detection operates in the space between these extremes. By analysing time-series data across multiple sensors and systems, it identifies patterns that fall outside normal operational baselines. These may reveal equipment operating inefficiently, systems running outside scheduled hours, unexpected energy consumption patterns, faulty sensors or control loops, and early indicators of asset degradation that would not trigger a conventional alarm for weeks or months.
In many cases, anomaly detection alone uncovers significant opportunities for energy optimisation and sustainability improvement, value that is captured immediately, not after a capital investment programme.
Intelligent rules and continuous learning
AI works most effectively when combined with existing operational frameworks. Many smart building platforms rely on rule-based monitoring to track asset performance and trigger alerts. These rules typically monitor thresholds, schedules or specific operating conditions. However, static rules require ongoing manual tuning and generate false alerts when building conditions change.
A dynamic logic engine continuously evaluates building telemetry and operational data, using AI techniques to analyse historical system behaviour and optimise monitoring rules. This creates a continuously improving intelligence layer where the system learns from operational data and refines its detection capabilities over time. The longer it runs, the smarter it gets.
The crossroads
Pivot or fade
The FM industry has talked about digital transformation for years. Conferences have been held, reports have been published, pilot projects have been launched and quietly shelved. The rhetoric has been consistent but the action has not.
What makes 2026 different is that the window for gradual adoption is closing. AI capabilities are not improving linearly, they are compounding. The platforms available today are materially more capable than those available twelve months ago, and the platforms available in twelve months will be materially more capable again. Organisations that delay adoption are not standing still. They are falling further behind with every quarter.
Meanwhile, technology-native companies do not carry the legacy overhead that slows traditional FM providers. They do not need to retrain a workforce accustomed to clipboard-based processes. They do not need to integrate decades of fragmented data. They do not need to overcome cultural resistance from stakeholders who view AI as a threat rather than a tool. They start clean, move fast and scale efficiently.
This does not mean traditional FM providers cannot compete. It means they must move now. The advantages that incumbent providers hold, engineering expertise, client relationships, regulatory knowledge, site-level operational understanding, are genuine and significant. But they are depreciating assets if they are not augmented by data and technology. Engineering expertise combined with AI is formidable. Engineering expertise competing against AI is a losing position.
The organisations that will thrive are those that recognise data-led maintenance not as a technology project, but as a fundamental shift in how buildings are operated and maintained. They will embed AI into their core service delivery, use data to demonstrate compliance and value to clients, and build operational models that improve continuously rather than degrade with every staff departure.
The organisations that will fade are those that treat this moment as another cycle of industry hype, defer investment pending clearer signals, and continue to rely on operating models that were designed for a world without data. By the time the signals are clear enough for the most cautious, the market will have already moved.
Measurable outcomes
Why data-led maintenance delivers
The case for DLM is not aspirational. The outcomes are measurable, documented and commercially proven across live building portfolios.
Reduced reactive maintenance costs through early fault detection and predictive intervention, eliminating the premium costs of emergency callouts and unplanned downtime.
Increased asset reliability and lifespan by servicing assets based on condition rather than arbitrary schedules, reducing both over-maintenance and under-maintenance.
Improved operational efficiency through automated work order generation, intelligent triage and the elimination of unnecessary PPM visits.
Lower energy consumption via continuous optimisation of building systems, identification of energy waste at a granular level and automated adjustment of operating profiles.
Stronger compliance and audit readiness through real-time digital documentation, automated golden thread maintenance and traceable records that satisfy regulatory requirements without manual intervention.
Workforce resilience by capturing operational intelligence in data systems rather than relying on institutional knowledge that is vulnerable to staff turnover.
For organisations managing large property portfolios, these benefits compound across sites. The intelligence gathered from one building improves the models applied to the next. Scale becomes an advantage rather than a source of complexity.
Conclusion
The UK facilities management industry has arrived at a moment that will determine its shape for the next decade. The convergence of regulatory pressure, workforce shortages, cost constraints and AI maturity has created conditions where the status quo is no longer a viable strategy.
Every month of inaction increases financial waste through unnecessary maintenance and missed fault detection. It deepens compliance exposure as golden thread requirements move from guidance to enforcement. It widens the capability gap between organisations that are learning from their building data and those that are still collecting it in silos. And it gives technology-native competitors more time to build the platforms, relationships and track records that will eventually displace traditional service models.
Data-led maintenance is not a future technology. It is available today, commercially proven and deployable over existing building systems without capital-intensive infrastructure upgrades. The barriers to adoption are not technical, they are organisational. Inertia, risk aversion and the false comfort of familiar processes.
The FM industry has been the last major sector to embrace digital transformation. That distinction now carries a cost. The question every FM leader, managing agent and asset owner must answer in 2026 is not whether to adopt data-led maintenance. It is whether they can afford not to.
Those that pivot will survive and strengthen. Those that remain indecisive will fade.
The crossroads is here.
The time to choose is now.
About the author
