Articles

The Missing Layer

FM Data Literacy as a Prerequisite for AI Success 

Part 1 of this series made the case that data quality is the structural barrier limiting AI adoption in facilities management. Clean, well-integrated, standards-compliant data is a prerequisite for AI systems to function reliably. But solving the data problem creates a second, equally important question: even when the data is ready, who is equipped to use what it produces?

The answer, according to emerging research from the International Facility Management Association (IFMA), is: not yet enough people. And the gap is growing wider faster than the profession is closing it.

The Rise of the FM Analyst

IFMA’s research report The Rise of the FM Analyst documents a fundamental shift in what facility management roles require. FM teams are building new data fluency capabilities in response to the analytical demands of modern building operations. Three competency areas are identified as most urgently needed: analytics proficiency, data governance literacy, and cross-stakeholder communication of data-driven insights.

From these demands, a new role archetype is emerging - the FM Analyst. This is a professional positioned at the intersection of building operations and data interpretation: someone who can extract meaning from the sensor streams, work order histories, occupancy patterns, and energy consumption data that modern buildings generate continuously, and translate that meaning into decisions that operators and executives can act on.

The emergence of the FM Analyst role reflects a structural recognition that AI tools and data platforms do not generate value autonomously. They require human judgment to function as intended - judgment that must be grounded in data literacy to be reliable.

What Data Literacy Means - and Does Not Mean

Data literacy is not the same as data science. The distinction matters practically and professionally. Data science involves building models, writing algorithms, and architecting analytical infrastructure. Data literacy involves reading outputs, asking the right questions of data systems, recognizing anomalous results, and translating findings into operational decisions.

Consider a predictive maintenance dashboard that flags an HVAC unit as high risk of failure within 30 days. A data-literate FM professional can interrogate that output: Is the sensor reading for this model calibrated? Has this unit generated similar alerts that proved to be false positives? Does the maintenance history suggest a pattern that explains this spike, or is something genuinely unusual? Without that layer of professional judgment, the alert is either ignored because it is distrusted or acted on blindly because it carries the authority of a data system.

Think of it as the difference between a pilot who can operate an aircraft and one who can also interpret instrument alerts, cross-reference competing readings, and make sound judgment calls when automated systems conflict. The autopilot handles cruise; the pilot handles the edge cases. AI in FM works the same way: the system handles pattern detection at scale; the FM professional handles the judgment calls that scale cannot replace.

IFMA’s Rethinking Data in FM research reinforces this framing, identifying data-driven decision-making and the capacity to translate analytical outputs into operational action as the defining competency shift for the profession in this decade.

The Widening Skills Gap

IFMA’s 2026 Global FM Trends research and FMJ synthesis both point to a compounding dynamic. As FM roles grow more complex and data-intensive, the profession faces a demographic challenge that amplifies the skills gap: significant retirement attrition is projected over the next decade, and the institutional knowledge leaving with retiring professionals is not being replaced at the rate the profession requires.

The result is a double gap. Technical data skills are insufficient - too few FM professionals have been trained to work with modern analytics platforms, interpret AI-generated outputs critically, or apply data governance principles in practice. Simultaneously, the accumulated operational experience that provides the contextual judgment needed to evaluate data outputs intelligently is attriting faster than it is being built into incoming talent.

This creates organizational fragility. An FM team with strong data infrastructure but weak data literacy will see AI tools produce outputs that nobody fully trusts, leading to either the underuse of expensive systems or an overreliance on their recommendations in contexts where human judgment should override them. Neither outcome serves the building owner, the occupants, or the profession.

The AI-Literacy Loop

There is a circularity problem embedded in the current moment. AI tools are frequently positioned as solutions to FM workforce capacity constraints - they are supposed to reduce the cognitive burden on FM professionals by automating routine analysis and surfacing actionable insights. But AI tools only deliver on that promise when the professionals using them have sufficient data literacy to evaluate whether those insights are sound.

Without data literacy, AI in FM risks reproducing a pattern the industry already knows well: expensive technology that generates outputs nobody can confidently validate, quietly accumulating in dashboards that are checked less and less frequently. The tools do not fail technologically. They fail organizationally because the human layer required to make them useful is absent.

Closing this loop requires treating data literacy not as a credential to add to a job posting but as a core professional competency to be developed systematically - through training programs, certification pathways, role design, and organizational investment in the FM Analyst capability that IFMA’s research identifies as the structural response.

What Leaders and Professional Bodies Should Do

The response to the FM data literacy gap operates at two levels: organizational and professional.

At the organizational level, FM leaders should:

•    Treat data literacy as a core operational competency - not a bonus skill - and reflect that in hiring criteria, role design, and professional development planning.

•    Create or designate FM Analyst roles that formally bridge building operations and data interpretation, rather than expecting existing staff to absorb these responsibilities without structural support.

•    Invest in practical data literacy training tied to the specific platforms and data systems in use - generic analytics training is significantly less effective than training anchored in operational context.

•    Design AI tool deployments to include explicit protocols for human review, flagging, and override - building in the judgment layer rather than assuming it.

At the professional development level, industry bodies have a parallel responsibility:

•    Integrate data fluency modules into existing FM certification pathways - the CFM and SFP credentials administered by IFMA are natural vehicles for embedding data literacy standards alongside traditional FM competencies.

•    Develop case-based learning resources that ground data literacy in building operations contexts, making the connection between data skills and FM practice concrete for practitioners at all career stages.

•    Create peer networks and communities of practice for FM Analysts, providing the professional recognition and knowledge-sharing infrastructure that accelerates competency development across the field.

The Data Quality Bottleneck

Why FM’s AI Ambitions Stall at the Data Layer

Building Lifecycle Management Initiative (BLMI) — April 2026

This article is Part 1 of a two-part series on FM AI adoption. This article, Part 1 - “The Data Quality Bottleneck” - examined the structural data infrastructure barriers identified in Johnson Controls’ 2026 AI & Digitalization in FM Report. Part 2 examines the workforce competency dimension: the data literacy gap that limits FM teams’ ability to use AI effectively even when the data is ready. The numbers tell a story of momentum. According to Johnson Controls’ 2026 AI & Digitalization in FM Report - a survey of 1,020 business and facility management leaders conducted in December 2025 - 65% of business leaders and 67% of FM professionals are already using artificial intelligence for facility operations. On the surface, this looks like a profession in full stride.

But the same survey reveals a critical friction point: when leaders were asked what most limits their ability to scale AI, data quality and integration ranked first - ahead of budget constraints, cybersecurity concerns, and expertise gaps. The tools are available. The will is present. What is missing is the connective tissue.

What 1,020 Leaders Are Saying

Johnson Controls surveyed 760 U.S. business leaders and 260 FM professionals for its 2026 report, producing one of the largest practitioner datasets on AI adoption in facilities to date. The findings are unambiguous about where the sector stands and what is holding it back.

47% of FM respondents plan to deploy predictive maintenance AI within the next 12 months. Yet one in three business leaders identified ease of integration as the single most desired improvement in their workplace management systems. That tension between deployment ambition and integration readiness defines the current moment.

Budget ranked lower as a barrier than data quality. When cost is not the primary obstacle, the problem is structural. The industry is not short on capital or intent. It is short on the data foundations that AI requires to function reliably.

The Real Bottleneck: Data Quality and Integration

Imagine training a skilled financial analyst to identify patterns in quarterly results - then handing them a filing cabinet of handwritten notes, half in different formats, several missing entirely, and none timestamped consistently. No analyst, however talented, can surface reliable insights from fundamentally inconsistent inputs. Artificial intelligence is no different.

Machine learning models require data that is consistent in structure, complete in coverage, and reliable in origin. Facility operations - spanning Computerized Maintenance Management Systems (CMMS) work orders, Building Automation System (BAS) sensor streams, space management platforms, Integrated Workplace Management Systems (IWMS), and energy metering systems - routinely fail on all three dimensions.

These systems were built independently, by different vendors, for different purposes, at different times. Data fields that should represent the same concept - a piece of equipment, a floor, a maintenance event - are labeled differently across systems. Values are entered inconsistently. Records are incomplete. The result is what practitioners sometimes call a “data swamp”: vast quantities of information that are individually plausible but collectively unusable for predictive analysis without significant remediation work.

The Standards Architecture Response: OSCRE’s Smart Data Highway

In February 2026, OSCRE International announced a significant evolution of its approach to building data standardization. Rather than continuing to develop data definitions in isolation, OSCRE unveiled a “Smart Data Highway” vision - a semantic framework designed to add intelligent layers to its Industry Data Model (IDM), enabling systems to not merely define data fields but comprehend their inherent relationships and broader context.

The initiative is designed to make AI, digital twins, and advanced analytics genuinely functional across the built environment’s fragmented data landscape. OSCRE describes the goal as enabling “intelligent, automated understanding and the seamless flow of information” - language that speaks directly to the integration barrier the Johnson Controls research identifies as the primary AI constraint.

This matters because the data quality problem in FM is not simply a matter of cleaning up existing records. It is a matter of structural alignment across the entire building data lifecycle - from design and construction handoff through decades of operational change. Without a common data model that systems can reference, cleaning data in one system does not prevent it from becoming inconsistent again as it flows into another. The ConnexFM-OSCRE partnership, formalized in January 2025, reflects the FM practitioner community's recognition that data management requires coordinated standards development and systematic implementation across platforms.

The CAFM/IWMS Integration Gap

Despite significant investment in Computer-Aided Facility Management (CAFM) and IWMS platforms over the past two decades, integration between these systems and emerging AI layers remains one of FM’s most persistent operational challenges. These platforms remain siloed in most operational environments, with significant manual coordination required to move data between them.

The consequence is that even organizations with modern FM platforms often cannot present AI systems with the unified, consistent data feeds those systems require. Predictive maintenance models need clean historical work order records, real-time equipment sensor data, and asset specifications - typically resident in three separate systems with three separate data models. Research finds that more than two-thirds of FM workers spend significant time on manual coordination across tools rather than higher-value analytical work - a ratio AI is supposed to invert, but cannot when data integration is absent.

As AI capabilities in the CAFM and IWMS space mature, industry observers anticipate convergence toward unified platform environments. But that convergence is years away for most portfolios. FM organizations face a choice: invest in data integration infrastructure now, or continue deploying AI tools that underperform because the data they depend on is not ready.

What FM Leaders Should Do Now

The path forward is not to wait for platforms to converge or standards to achieve universal adoption. It is to begin building data governance disciplines now, while those standards and platforms mature.

• Audit the quality of audit data before selecting or expanding AI tools. Understanding where data is inconsistent, incomplete, or siloed is a prerequisite for meaningful AI deployment planning.
• Align data architecture with the OSCRE Industry Data Model. The IDM provides a governance-grade framework for mapping FM data across systems - organizations that align with it now will benefit as AI tools increasingly depend on it.
• Treat integration as an infrastructure investment, not a project cost. The tendency to treat data integration as a one-time implementation expense rather than an ongoing operational capability is a primary reason integration debt accumulates.
• Build data governance into FM operations. Assign data stewardship responsibilities, establish data quality standards for system inputs, and create review cycles. Data quality is not an IT problem - it is an operational discipline.

 Key Takeaways

• AI adoption in FM is real and accelerating - 65–67% of leaders are already using it for facility operations.
• Data quality and integration is the #1 barrier to scaling AI in FM, outranking budget constraints, cybersecurity concerns, and expertise gaps.
• The problem is structural: fragmented systems, inconsistent data models, and absent governance - not a lack of technology or intent.
• OSCRE’s Smart Data Highway initiative (February 2026) provides the standards architecture the built environment needs for AI to function reliably at scale.
• FM leaders should audit data foundations and align with the OSCRE Industry Data Model before expanding AI tool deployments.

Sources

• Johnson Controls. (2026). 2026 AI & Digitalization in FM Report. Johnson Controls International.
• OSCRE International. (2026, February). OSCRE unveils vision for a Smart Data Highway. PRWeb.
• ConnexFM & OSCRE International. (2025, January). ConnexFM and OSCRE International partner on managed services agreement. PR Newswire.
• REMI Network. (2026). Five AI insights for FMs in 2026. reminetwork.com.
• eFACiLiTY / Sierra ODC. (2025). AI in Facility Management enters the GenAI era. efacility.in.
• Facilities Dive. (2026). Coverage of AI adoption and cybersecurity in facilities management. facilitiesdive.com.
• Building Lifecycle Management Initiative (BLMI). (2026). FM AI Adoption Series. blmi.org.