Published on :
June 19, 2026
by
Anisha Bhattacharjee
Facilities management has always been measured by what gets done. SLA attainment, PPM completion, CMMS work order closure, CAFM compliance scores, so these became the language of FM performance because they reflected what the operating model was built around: executing maintenance work correctly and on time. That framework served FM well within its context. The challenge is not that it was wrong. The challenge is that something is now achievable that simply was not before: continuous, portfolio-scale verification of whether maintenance work produced the result it was intended to produce. That changes what FM can measure, report, and demonstrate to the people who fund it.
The FM operating model was built around work execution. A fault occurred. A technician attended. Work was completed. The work order closed.
The KPIs that emerged reflected that model honestly:
These were the right measures for an execution-led model. They provided consistent visibility into activity across portfolios, contractors, and sites.
What they could not do, and were never designed to do, was confirm whether the activity produced the intended result. Not because anyone overlooked this, but because verifying outcomes at portfolio scale was not practically achievable. The data existed in fragments across CMMS records, BMS logs, contractor reports, and asset histories, systems that were never designed to communicate with each other. Connecting them meaningfully across hundreds of assets and multiple sites was not realistic.
So FM measured what it could measure. And for a long time, within its context, that was sufficient.
Consider two portfolios, both managed diligently, both hitting every number.
Traditional reporting reads both as performing equally well. They are not.
In Portfolio A, interventions address underlying conditions. Repeat calls on the same assets reduce over time. The PPM programme adapts based on what failure data is showing. Reactive demand falls quarter on quarter.
In Portfolio B, faults are attended and closed correctly. SLAs are met. But the same assets keep generating tickets. The PPM schedule runs as written regardless of what asset history suggests. Reactive demand quietly builds.
The KPI framework reads both identically because it records whether work happened, not what the work produced.
This difference was always present. What has changed is the context around it. Budget scrutiny has tightened. Asset owners and finance functions are asking harder questions about what maintenance spend is delivering, and activity metrics now need to be accompanied by evidence of what the work actually produced. Until recently, there was no practical way to generate that evidence at scale.
The reason outcome verification was not achievable was structural. There was no layer that could sit above CMMS, CAFM, BMS, and maintenance execution data simultaneously, connect those signals continuously, and translate them into evidence of whether a decision worked.
That layer now exists.
This is what a System of Decisions does. Where a CMMS is a system of record that captures maintenance activity, a System of Decisions is a governance layer that asks what that activity produced and whether the decision behind it was correct. It does not replace existing FM infrastructure. It sits above it.
Xempla is built as a System of Decisions for FM. Its governance layer uses the DIIV Cycle - Discover, Investigate, Implement, Verify, to make every intervention traceable from decision to outcome.
Discover — anomalies and maintenance signals are identified before they become reactive calls, drawing on asset history, BMS data, and failure pattern recognition across the portfolio.
Investigate — the issue is analysed in context: asset criticality, maintenance history, risk profile, and comparable failure patterns across the estate.
Implement — a guided intervention is executed, with the reasoning behind the recommendation logged at the point of decision.
Verify — did the intervention produce the intended result? Did the asset return to expected performance? Did the fault recur? If the outcome is not confirmed, the cycle reopens.
On a 280 kWp commercial rooftop solar installation, Xempla's agent detected an abrupt drop in PV output. Multiple strings across two inverters were underperforming by 30 to 40% against peer strings. Investigation identified heavy soiling on central panel sections. A targeted cleaning intervention was recommended and executed. Verification confirmed a 14% improvement in energy generation, approximately 250 kWh of additional output per day. Fault found, cause identified, intervention made, outcome confirmed.
When the DIIV Cycle runs continuously across a portfolio, every asset, every intervention, every outcome logged and verified, it generates something FM has not previously had: a continuous, evidence-based measure of operational health.
This is the Assurance Score.
Xempla's Central Assurance Layer evaluates the health of facilities and locations continuously, distilling complex FM signals into a single score at two levels, asset and facility, and portfolio-wide.
At the asset and facility level, four signals combine into a confidence measure for each asset:
At the portfolio level, these signals aggregate across all assets, sites, and contractors into a single governance metric, giving leadership a continuous view of whether the programme is producing operational value, not just completing tasks.
This is not a completion rate. It is a confidence measure. It answers the question a work order closure rate cannot: how well is this portfolio actually being managed right now?
In practice: India's fastest-growing PE-backed FM business deployed Xempla as its System of Decisions across a scaling portfolio. Asset Performance Assurance moved from 30% to 85%. Hard FM effort reduced by approximately 40%. The central support model stayed lean with no new senior operations roles added as the portfolio grew.
Client conversations shifted from activity reporting to evidence of asset performance improvement and outcome-aligned spend.
Explain reactive demand before it escalates. The Portfolio B pattern, rising reactive work despite strong KPI numbers, becomes visible at asset level in the Assurance Score before it reaches a contract review.
Address repeat faults at root cause. Rather than closing the same work order repeatedly, the governance layer identifies the underlying pattern and the intervention that resolves it.
Demonstrate maintenance value in outcome terms. When interventions are verified rather than assumed closed, FM teams can show reliability improvement and risk reduction alongside SLA compliance, not instead of it.
Scale without proportional overhead. The governance layer carries the analytical load across every site consistently, regardless of local team capability, making growth manageable without adding central support headcount.
Activity KPIs such as SLA compliance, PPM completion, and work order closure confirm that maintenance processes were followed. Outcome KPIs confirm whether those processes achieved their intended result, such as improved asset reliability, reduced reactive demand, or eliminated repeat faults. The shift is not about replacing one with the other but about adding verification to what already exists.
The Assurance Score is Xempla's measure of verified maintenance outcomes, operating at both asset and portfolio level. It combines four signals, Stability Confidence, Triage Confidence, Decision Assurance, and Preventive Multiplier, into a continuous confidence measure reflecting how well assets and facilities are actually being managed, not simply whether tasks were completed.
The DIIV Cycle, Discover, Investigate, Implement, Verify, is Xempla's operational framework for AI-governed maintenance resolution. It is how Xempla works through every maintenance intervention, ensuring each one is traceable from the initial signal through to verified outcome. The Verify stage is what generates the evidence that feeds the Assurance Score.
A CMMS records maintenance activity. A System of Decisions is a governance layer that connects maintenance decisions to outcomes and verifies whether those decisions achieved their intended result. Xempla is built as a System of Decisions, sitting above existing FM infrastructure to make outcome verification possible at scale.
Traditional KPIs remain a necessary operational foundation. SLA compliance, PPM completion, and work order closure rates provide essential visibility into process adherence. Outcome verification does not replace them. It adds a layer above them that confirms whether the processes being followed are producing the results they are designed to produce.
Paragraph
Block quote
Ordered list
Unordered list
Bold text
Emphasis
Superscript
Subscript
© Xempla. All rights reserved.
Terms & Conditions