PM and PdM compliance: are your scheduled tasks actually getting done on time?

At the end of the month a few hundred preventive and predictive tasks were due, and the team closed most of them on time. The headline number looks perfectly reasonable, until someone scrolls back through the year and notices that a few months ago compliance quietly dropped into the low sixties, and nobody asked why.

PM and PdM compliance is the simplest health-check a maintenance programme has. It tells you, without much ambiguity, whether the schedule of preventive care is actually being honoured.

What it actually measures

PM and PdM compliance is the percentage of scheduled preventive and predictive maintenance tasks that were completed by their due date in a given period. The phrase "by their due date" hides a crucial detail: a consistent grace period is normally allowed, so a task still counts as on time if it lands within a defined window past its nominal date. The usual rule is the due date plus twenty per cent of the task's frequency, capped at twenty-eight days. A monthly task therefore carries about six days of grace; a quarterly one around eighteen; anything on a six-month-or-longer cycle hits the twenty-eight-day ceiling. The exact window matters far less than applying it identically to every task and every period, because the moment "on time" means something different from one month to the next, the trend stops meaning anything at all.

How to work it out

PM and PdM compliance (%) = (Tasks completed on time × 100) / Tasks due

Of 520 tasks due in the month, 452 were completed within their grace window.

PM and PdM compliance = (452 × 100) / 520 = 86.9%

That sits just under target, but the single month is less revealing than the trend behind it. Plotting the whole year shows an earlier plunge into the low sixties, the kind of sharp drop that always has a cause worth finding.

Why it is a leading indicator

It is tempting to file compliance as mere administrative tidiness, but it is better understood as one of the earliest warning lights you have. Reliability itself is a lagging indicator: by the time failures rise, the damage is already done. Compliance leads it. When preventive tasks start slipping, the equipment has not failed yet, but the protection that was keeping it healthy is being quietly withdrawn, and the failures will follow in their own time. A falling compliance line is therefore a chance to act before the breakdowns arrive, which is exactly the kind of early signal a maintenance programme is supposed to give.

The way compliance slips is almost always the same quiet story. A breakdown swallows the week, so the preventive tasks due that week get bumped to next week, which is already full, so they get bumped again. Nobody decides to stop doing prevention; it simply gets borrowed from, a day at a time, to pay for today's fire. That is why a steady compliance figure is as much a cultural achievement as a technical one: it means the organisation is protecting its preventive work even under pressure, instead of raiding it whenever the plant gets loud.

What good looks like

The target is ninety per cent and above, and some operations hold themselves to ninety-five. The precise figure matters less than the discipline of reading the trend honestly and not letting a soft definition flatter it. One distinction is worth drawing clearly, because it trips people up: this metric measures whether tasks were done on time, a simple yes-or-no against the due date. A separate measure looks at how far past their interval the late ones ran, and whether that lateness is landing on critical assets. The two sound alike but answer different questions, and a mature programme watches both.

Where it can mislead you

• If a task is completed early, set the next occurrence from the actual completion date, not the original due date. Skip this and the next reading is artificially inflated, because you have quietly lengthened the interval.
• Apply the grace period consistently. Redefining "on time" between periods, or stretching the window to rescue a bad month, turns the whole trend into a fiction.
• It measures schedule adherence, not quality. A task ticked off on time but done carelessly, or barely done at all, still counts as compliant, which is why a high score has to be paired with an effectiveness measure before you trust it.
• A sharp month-on-month drop almost always has an identifiable cause: a shutdown, a resource shortage, a priority clash. Investigate it promptly rather than waiting for the line to recover on its own.

PM and PdM compliance asks the most basic question you can put to a maintenance programme: are the scheduled tasks actually being done on time? Everything more sophisticated, the yield of those inspections, their effectiveness, the reliability they ultimately buy, rests on the answer being yes, week after week, and not just on the months when someone happens to be watching.