PM compliance tells you whether preventive maintenance was completed when it was due. It does not tell you whether the task was technically capable of preventing, detecting, or controlling the failures that matter. High compliance can coexist with poor reliability when the maintenance strategy itself is weak.
Compliance is necessary because even a technically excellent maintenance strategy cannot help if the work is never executed. But once a site can reliably complete scheduled PM work, the next question must be whether the work itself deserves to exist in its current form.
PM compliance answers:
Did we complete the preventive maintenance that was due, inside the completion window we defined?
A common calculation is:
The site must define “on time” consistently. That may mean by the due date or within an approved tolerance based on the task frequency and risk.
PM effectiveness asks:
Did the task actually prevent a failure, detect deterioration early enough to act, restore a required condition, or control the failure mode it was designed for?
It is not one universal percentage. It is a body of evidence connecting PM content to asset failure behavior, findings, corrective follow-up, reliability, and maintenance burden.
That matters. A technically excellent PM program that is never completed cannot protect the equipment. But once execution is under control, the next question has to be whether the work itself deserves to remain in the program.
Scheduled PM work is being completed inside the organization’s defined timing rule.
The task addresses the correct failure mode, uses the correct method, or has a defensible frequency.
A completed inspection produced useful findings, corrective action, or a measurable reduction in recurrence.
An effective PM system connects execution to condition information, corrective action, and changed failure behavior. If the chain breaks after completion, compliance can look excellent while reliability does not improve.
When a plant tells me PM compliance is above 90% but breakdowns remain high, I do not start by asking how to push compliance higher. I start by asking what is actually failing, which PM task is supposed to control it, what that task tells the technician to do, and what happens when the technician finds something wrong.
The PM is executed exactly as written, but the recurring failures are caused by conditions the task never inspects, measures, restores, or detects.
“Inspect motor,” “check conveyor,” and “service pump” can all close on time while giving technicians no consistent inspection point, method, or acceptance criterion.
A legacy interval can be executed perfectly while being too short, too long, or simply unrelated to the deterioration pattern it is supposed to manage.
An inspection finds looseness, leakage, wear, contamination, or abnormal condition, but no corrective work is generated and the defect survives until failure.
Duplicated, overly intrusive, or low-value tasks can consume labor and equipment access without creating enough risk reduction to justify the maintenance touch.
Plot execution discipline and technical effectiveness separately. The same compliance percentage can describe very different maintenance systems.
The plant is neither executing the program consistently nor confident that the technical strategy is sound. Stabilize work management while prioritizing the highest-risk PM gaps.
The task may be technically sound, but reactive work, access, labor, scheduling, parts, or production pressure is preventing consistent execution.
This is the false-positive state. The KPI is green, but repeat failures, emergency work, weak findings, or poor failure-mode coverage show the strategy needs redesign.
The work is executed consistently, the tasks are technically defensible, defects are detected early, corrective work follows, and failure behavior supports the strategy.
A stronger PM review combines execution data with failure behavior, finding quality, corrective follow-up, and task-level evidence.
Are assets continuing to experience the same failure modes even though the relevant PMs were completed on time?
Which inspection or condition tasks actually identify actionable deterioration, and which repeatedly close with no useful condition information?
When a PM finds a defect, does it create, prioritize, and complete corrective work before functional failure?
Do credible asset failure modes have an appropriate maintenance strategy, or are important mechanisms uncovered while low-value tasks remain?
Where the data is trustworthy, are failure frequency, downtime, reactive labor, emergency work, or similar asset-level signals improving after strategy changes?
How much labor, access, production interruption, parts consumption, and planning effort does the PM program require for the risk it controls?
There is no single universal PM-effectiveness percentage that answers every maintenance question. Use several measures that connect the PM program to condition, corrective work, failure behavior, and resource burden.
| Measure | What it tells you | What it does not prove |
|---|---|---|
| PM compliance | Whether scheduled PM work was completed inside the defined timing rule. | That the task was necessary, technically correct, or effective. |
| Finding yield | How often inspection or condition tasks identify actionable deterioration. | That every low-yield task should automatically be removed. |
| Corrective conversion | Whether PM findings become tracked corrective work rather than dying in notes. | That corrective work was completed before failure. |
| Repeat failure after PM | Whether the same failure mode persists despite completion of the relevant PM. | That PM frequency is automatically the root cause. |
| Failure-mode coverage | Whether credible failure mechanisms have an appropriate management strategy. | That every failure mode requires a scheduled PM task. |
| Reactive burden | Whether emergency labor, downtime, and reactive work are trending as strategy changes. | That PM is the only factor influencing the trend. |
| PM labor burden | How much capacity the PM program consumes. | Whether high or low labor by itself is good or bad. |
A common reaction to failure is to increase PM frequency. Sometimes that is appropriate. Sometimes it simply creates more maintenance touches without controlling the mechanism.
If the task does not inspect, measure, restore, or test the condition associated with the failure mode, performing it twice as often does not make it technically relevant.
Random failures, installation defects, contamination, overload, design weakness, operating conditions, or poor repair quality may not respond to a shorter calendar interval.
Condition monitoring, failure-finding, redesign, operating control, improved installation standards, spares strategy, or run-to-failure may be more appropriate than more time-based PM.
NASA's reliability-centered maintenance guidance emphasizes matching maintenance actions to specific failure modes and evaluating whether the task is applicable and effective. That is the discipline missing when frequency becomes the default answer to every recurrence.
A conveyor line completes 95 of 100 scheduled PMs inside its defined completion window. Compliance is 95%.
But the same drive-end bearing continues to fail. The monthly PM says only “inspect conveyor.” Technician notes contain no bearing condition, temperature, vibration, lubrication condition, contamination finding, or acceptance criterion. No corrective work is generated before failure.
The execution metric is strong. The maintenance strategy is not.
The corrective response is not “push compliance to 100%.” The response is to validate the failure mode, determine the appropriate maintenance strategy, rewrite the task, establish the technical basis and acceptance criteria, and verify whether the new strategy changes the failure behavior.
You do not need to redesign the entire PM program to find the first problems. Pick one recurring failure and trace it through the maintenance system.
A completed wrong task is still the wrong task. The goal is not perfect compliance with inherited maintenance. The goal is controlled execution of technically defensible maintenance.
XRVE Reliability reviews PM master data against corrective work history, repeat failures, task quality, frequency basis, failure-mode coverage, and available technical evidence. The goal is not simply fewer PMs. It is a more defensible maintenance strategy.
See the full preventive maintenance optimization workflow, including Keep, Modify, Consolidate, Investigate Frequency, Convert Strategy, Add, and Remove dispositions.
Review a fictional PM effectiveness analysis showing task-level findings, evidence controls, confidence, and human-review requirements.
XRVE Reliability uses external standards and technical guidance as reference points while keeping site engineering, OEM requirements, approved procedures, operating context, and customer authority controlling.
The stronger question is whether the right maintenance was performed, at a defensible interval, with a method capable of controlling the failure modes the asset actually experiences.
Start free Maintenance Health Check Explore PM optimization