The CMMS is doing exactly what it was built to do. That is part of the problem.
Computerized maintenance management systems track work orders, schedule preventive maintenance, manage technician dispatch, and produce the reports that keep facilities organizations running. They are functional, mature, and embedded in nearly every institutional portfolio. They are also incomplete in a way most operators have stopped noticing.
A CMMS records what happens to systems. It does not give those systems a permanent identity.
When a chiller is replaced, the new unit typically inherits the old asset record. When a building changes hands, the operational history may transfer in a static export, or it may not transfer at all. When a CMMS vendor is replaced, service context gets compressed into whatever fields the new system can ingest. The systems themselves continue operating across all of these transitions. Their records do not.
This is not a software problem. It is a structural one. The infrastructure layer underneath every building's mechanical and energy systems is missing something that every other significant asset class already has.
What Other Industries Figured Out a Long Time Ago
Every vehicle on the road has a permanent 17-character identifier assigned at manufacture. The VIN is not owned by the dealer, the driver, the insurer, or the DMV. It is owned by the vehicle. When the car is sold, the VIN goes with it. When the insurance policy changes, the VIN is unaffected. When the dealership goes out of business, the VIN persists. CARFAX exists because that persistent identifier exists. The history report is a downstream consequence of the underlying identity layer.
Healthcare made the same move. The National Provider Identifier is a permanent number assigned to every healthcare provider in the United States. It survives practice changes, employer changes, and licensure transitions. Insurance, billing, and credentialing all reference the same persistent identifier across the provider's career.
Real estate works the same way. Land title registries assign permanent identifiers to property and track ownership through every transaction. The record outlasts every owner.
Each of these industries reached a point where the cost of fragmented records exceeded the cost of building a persistent identity layer underneath. Each industry now treats that layer as basic infrastructure.
Buildings have nothing equivalent.
Why the Gap Matters More Now Than It Used To
Three forces are converging that make the absence of system-level identity increasingly expensive.
Electrification is multiplying the asset count inside every building. Heat pumps, battery storage, solar arrays, EV charging infrastructure, microgrids. Each new system adds a lifecycle that needs to be tracked across decades. The volume of equipment requiring lifecycle continuity is rising faster than the systems underneath have adapted.
Capital markets are starting to underwrite against infrastructure records. Lenders increasingly request system-level documentation in due diligence. Insurers price risk based on assumptions about equipment age and maintenance history. Buyers discount portfolios with weak mechanical records. When records are fragmented or partial, that uncertainty gets priced into the transaction. The cost of missing identity is becoming quantifiable.
Asset management standards are formalizing globally. ISO 55000 frameworks, regional asset management planning regulations, and governance-grade reporting requirements are converging on the same need: defensible, traceable lifecycle records. The standard is shifting from best effort to auditable.
What was once an inconvenience has become a structural exposure.
What Persistent Identity Actually Looks Like
A persistent identity layer is not another platform. It is a reference layer that sits underneath whatever software is in use.
At a high level, a persistent identity layer would establish a consistent reference for each system that remains stable over time, regardless of how the building is operated, who services it, or what software is in use. The reference would remain associated with the system itself, rather than with any single owner, contractor, or platform. In practice, this would allow service activity, ownership transitions, and operational changes to be understood against a consistent system-level reference.
The CMMS still does what it does well. Work orders, dispatch, scheduling, reporting. The persistent identity layer does what the CMMS was never designed to do. It carries continuity across the transitions that break ordinary records.
The two layers are complementary. One manages workflow. The other preserves identity.
What Facility Managers Can Ask For Today
The persistent identity layer is still emerging in the built environment, but operators do not have to wait for the industry to standardize before they start protecting continuity in their own portfolios.
Three changes are worth considering immediately.
At handover, demand identity continuity. When new equipment is commissioned, require that the manufacturer, model, serial number, install date, and contractor information be captured in a structured format that does not depend on the current CMMS. Treat that record as a property of the equipment, not the project.
At ownership transition, treat infrastructure records as part of the asset. Insist that buyers and sellers transfer not just deeds and warranties but the operational record of every system in the building. If that record cannot be produced, treat the gap as a diligence finding, not a footnote.
At software migration, audit what gets lost. When a CMMS is replaced, do not assume the export carries everything. Map what existed in the previous system against what survives in the new one. The gaps reveal where identity was tied to the software rather than to the equipment.
None of these require new technology. They require treating the equipment record as something that belongs to the building, not to whoever currently happens to be operating it.
The Broader Point
Facilities managers have been working around the absence of persistent identity for decades. The workarounds are familiar: manual exports at handover, walking the building with a clipboard after a CMMS migration, reconstructing service history from contractor invoices when the previous operator left.
These workarounds were defensible when buildings were simpler and ownership cycles were longer. They are becoming less defensible as electrification expands the asset count, as institutional capital prices infrastructure records more carefully, and as governance standards formalize what defensible documentation actually looks like.
The work the CMMS does is real and necessary. It is not, by itself, a system of record for the equipment. The industries that have addressed this gap did so by building an identity layer underneath their workflow software. Buildings will eventually do the same.
The question worth asking is how operators choose to respond as expectations around infrastructure records continue to evolve.
