A maintenance director opens the monthly operations review with 96% PM compliance on slide three. On slide four, the previous week’s incident log includes a low-loader whose gearbox seized on the way to a job. The failure code on the closed work order is blank, and the last three scheduled services on that asset all closed on time. On slide five, the reliability engineer’s own bad-actor list has been kept on a laptop for the past eighteen months, because the report the maintenance team looks at every month reports compliance and reports almost nothing else. The number climbs quarter on quarter, and the incidents climb with it.
This is the shape of most maintenance scorecards in asset-intensive operators. A single number, agreed with a regulator or a board sub-committee a decade ago, has become the headline; every other measure sits in an appendix nobody opens. PM compliance in IBM Maximo Manage measures whether the planned schedule was executed inside its window. That is a valid schedule-execution metric. Steering a maintenance programme on it as the primary KPI is where the operating model breaks. The argument here: PM compliance belongs on the audit page, and a compact steering set of four other measures belongs on the operating scorecard the director carries into the ops review.
What 96% PM compliance actually measures
SMRP defines Preventive Maintenance Compliance as the percentage of PM work orders completed inside a defined window against the total scheduled in that window, with a widely quoted benchmark of ninety per cent. The metric answers one question: were the scheduled tasks executed on time. It says nothing about whether the task detected anything, whether it corrected anything, whether the interval was right, whether the operator ran to failure between visits, or whether the technician had the parts to do the work properly.
An operation running 96% PM compliance can carry an out-of-date PM library, miss failure codes on every corrective closeout, generate no follow-on work from its PM visits, and accumulate a backlog older than most of the crew. None of that lowers the number. Compliance rises when the schedule is executed, and the schedule is easier to execute when the tasks are short, cursory, and generate no downstream work. As a headline, the metric selects for the wrong behaviour.
The four measures that steer the programme
Four measures, taken together, tell a maintenance director whether the programme is working. All four have SMRP definitions, are widely used in reliability practice, and are recoverable from Maximo Manage without a data lake.
- PM yield. The share of PM and PdM work orders that identify corrective follow-on work. SMRP treats PM yield as the ratio of corrective work orders raised from a PM parent to the PM work orders completed in the same window. A PM library that yields nothing is one of two things: a fleet in genuinely stable condition, which is rare, or a task list the technician cannot fail. Ten to thirty per cent is a healthy operating range for most rotating equipment; sustained single digits are a signal to review the tasks.
- Schedule compliance. The share of scheduled labour hours completed against the hours in the frozen weekly schedule. Schedule compliance is a separate SMRP metric from PM compliance, and it captures whether the plan and the reality agree at the resource level. Where PM compliance is calendar-driven, schedule compliance is capacity-driven, and it exposes the crew shortage and the emergent-work churn that PM compliance hides.
- Backlog age distribution. Ready-to-schedule backlog work orders bucketed by age: zero to thirty days, thirty-one to ninety, over ninety. The distribution matters more than the total. A backlog that stays flat overall while the “over ninety” bucket grows is a backlog that is quietly ageing into a safety and reliability risk.
- Failure reporting completeness. The share of corrective work orders on critical assets that close with a problem code, a cause code, and a remedy code populated. Without this, the failure history built on the estate is decoration, and the case for any failure-code-driven reliability programme is unwinnable.
Four numbers, refreshed monthly, tell a director whether the PM library is finding anything, whether the crew can actually execute the plan, whether the backlog is ageing, and whether the record being built is worth reading in three years.
Producing the four in Maximo Manage
None of the four requires new tooling. All four are produced from records the platform already writes.
- PM yield. Corrective work orders generated from a PM parent carry that parent on the work order (
PARENTandWORKTYPEonWORKORDER). A saved query on child work orders raised within, for example, seven days of the parent PM completion date, grouped by asset class, is the raw numerator. Where the workflow does not raise follow-on work from the PM screen, a small piece of adoption discipline (raise the follow-on from the PM screen, not as a standalone) fixes the denominator problem at source. - Schedule compliance. Maximo Manage carries the scheduled labour hours on
WPLABORand the actual hours onLABTRANS. Where Scheduler is in use, the frozen weekly plan is the reference; where it is not, the crew supervisor’s committed schedule for the week is. A start centre KPI or a KPI Manager entry reads both totals on Sunday night. The value the director sees is a single percentage, with a drill-down into the outliers. - Backlog age distribution. A saved query on work orders in
WAPPR,APPRandWSCHstatus, with an age function onSTATUSDATE, produces the three buckets. A start centre result set on the maintenance manager’s dashboard shows the trend; an escalation flags any critical-asset work order that crosses ninety days without a schedule date. - Failure reporting completeness. The
FAILUREREPORTrecords attached to a closed work order are queryable directly. A KPI reading the share of critical-asset corrective work orders (defined by a criticality field onASSETorLOCATIONS) with all three codes populated is a single figure on the same dashboard.
The heaviest lift here is agreement, not technology: the definition of “critical asset” and the compliance window for schedule compliance. Both are working sessions with the reliability lead and the planning lead. The Maximo Manage configuration around them is small.
Where PM compliance still earns its place, and what stays hard
PM compliance does not leave the estate. Statutory inspection regimes under HSE-adjacent frameworks, sector regulators such as Ofgem or the CAA, and asset management system audits under ISO 55001 all expect evidence that planned work was executed on time. That evidence is real, and PM compliance is the natural language for it. The measure stays live on the compliance and audit page, tracked by asset criticality band and by regulatory category, and reported to the safety and audit function on its own cadence.
Three things stay hard. The board takes six to nine months to trust a new headline; PM compliance has been the shorthand for “the maintenance function is under control” for a generation, and no scorecard change accelerates the rebuilding of that reflex. Failure reporting completeness moves only as fast as the technician adoption behind it, which is a supervisor discipline, not a KPI change. And schedule compliance exposes crew and backlog realities that some operations have been quietly avoiding, which is the operating model conversation the KPI change is meant to force.
The position
PM compliance measures whether the schedule was executed. That is what regulators and auditors need to see, and it belongs on the compliance evidence page. It does not belong at the top of the maintenance scorecard, because as a headline it selects for a maintenance programme that executes an unchallenged plan, and it hides the four things a director actually needs to see: whether the PMs are finding work, whether the crew can execute the plan, whether the backlog is ageing, and whether the record being built is credible. Move PM compliance to the audit page, and put those four on the operating scorecard. The first quarter after the change is uncomfortable; the second quarter is where the programme starts to produce data worth acting on.
For the operating-model implications of steering a Maximo estate on outcomes rather than on execution volume, the shift from project mode to product mode covers the governance and procurement changes that most naturally follow.