PM Compliance: The Metric and the Target

PM compliance is the share of scheduled preventive maintenance tasks that get completed, in full, within their due window, on-time PMs divided by scheduled PMs, times 100. It is the heartbeat of a preventive maintenance schedule: the single number that says whether the PMs you committed to are actually happening on time, or quietly slipping.

It is also one of the most gameable metrics in the plant, and one of the most misunderstood. Two things have to be true before the number means anything: the compliance window has to be defined, and a PM has to be counted only when every task on it was actually done, not just when the work order was closed. Get those right and PM compliance is the best early-warning leading indicator you have. Get them wrong and it is a comforting number that hides a failing program. This guide covers the formula, the window, the 90% target, the two ways it gets gamed, and how it differs from planned maintenance percentage.

What is PM compliance?

PM compliance measures execution discipline: of all the preventive tasks that came due in a period, how many were completed on time and in full. It is a leading indicator, you can move it this week by changing how work gets done, and it sits upstream of the lagging outcomes it drives. When PM compliance holds high for a couple of quarters, reliability rises and emergency work falls. When it sags, breakdowns follow a few months later, because the inspections and services that would have caught them never happened.

What is the compliance window?

The compliance window is the tolerance you allow around a PM's due date before it counts as missed. It exists because "on time" is meaningless without it: a monthly PM will rarely be done on the exact calendar day, so you define an acceptable band, a common convention is plus or minus 10% of the interval. For a 30-day PM, that is about a three-day window; for an annual PM, about a five-week one. Complete it inside the band and it is compliant; complete it outside, and it is late, a miss.

One subtlety: doing a PM far too early can be as wrong as doing it late, because it wastes labor and, on time-based intervals, resets the clock so the next one drifts. A good window is tight enough to protect the interval and loose enough to be achievable. Whatever you choose, define it explicitly and apply it the same way to every asset, the discipline behind any honest maintenance KPI.

How do you calculate PM compliance?

Count the PMs that came due in the period, count how many were completed on time and in full, and divide. If 200 PMs were scheduled in a month and 176 were finished inside their windows with all tasks done, PM compliance is 176 ÷ 200 × 100 = 88%. The two counting rules do the heavy lifting: a PM completed but three days past its window does not count, and a PM whose work order was closed with half its checklist skipped does not count either. Closing a work order is not the same as completing the work.

How do you run PM compliance well? Six steps

1. Define the compliance window first. Set the tolerance, commonly plus or minus 10% of the interval, before you measure anything, and write it down. Everything else depends on this.
2. Count only PMs completed in full. A PM counts as compliant only when every task on it was done and recorded. Closing the work order without doing the steps is the most corrosive form of gaming, because it destroys the metric and the reliability at the same time.
3. Measure by criticality, not just plant-wide. A 90% plant average can hide 70% compliance on your most critical assets. Break the number out by asset criticality and hold the tightest expectations where failure hurts most.
4. Do not chase 100%. Perfect compliance usually means the schedule is too loose or people are closing work orders they did not do. A realistic world-class number is around 90%, with genuine misses visible and explained.
5. Review weekly and act on the misses. PM compliance is a weekly metric. Each miss is a small signal, a parts stockout, a staffing gap, an inaccessible machine, and the value is in fixing the cause, not just recording the shortfall.
6. Audit the schedule the compliance is measuring. High compliance on the wrong PMs is wasted effort. Periodically ask whether each PM still earns its place, so you are not perfectly executing tasks that do nothing.

What is a good PM compliance percentage?

The commonly cited world-class benchmark is 90% or higher and below roughly 80% the PM program is generally considered not to be functioning effectively. Those figures are consistent with the metrics the Society for Maintenance and Reliability Professionals (SMRP) publishes in its Best Practices library, the field's closest thing to a standard and the right place to anchor a benchmark rather than a vendor number (SMRP Best Practices, Metrics & Guidelines). The reason the target sits so high is that PMs are cheap insurance: the U.S. Department of Energy's FEMP O&M guidance, maintained by Pacific Northwest National Laboratory, documents that keeping maintenance proactive rather than reactive offers savings that can exceed 30-40% (PNNL, O&M Best Practices), savings you forfeit every time a PM slips and a preventable failure gets through. With skilled crews getting scarcer, the U.S. Bureau of Labor Statistics projects 13% growth for industrial machinery mechanics and maintenance workers from 2024 to 2034 (BLS), the hours a missed PM eventually costs in emergency repair are the most expensive hours you can spend.

How does PM compliance get gamed?

Two failure modes make a good-looking PM compliance number lie, and both are common.

The first is false completion: closing work orders for PMs that were not actually done in full. Compliance reads high, the reliability it was supposed to protect keeps eroding, and the gap surfaces only when a machine that was "maintained" fails anyway. This is why the "in full" rule matters as much as the "on time" one, and why spot-auditing completed PMs against their checklists is worth the effort.

The second is subtler: high compliance on the wrong schedule. You can execute a bloated or poorly designed PM program with perfect discipline and still fail, because you are flawlessly doing tasks that do not address the failure modes that actually break your equipment. Compliance measures whether you did the work, not whether the work was worth doing.

How is PM compliance different from PMP and schedule compliance?

PM compliance is often confused with two neighbors. Planned maintenance percentage asks whether work was prepared before it started (proactivity); PM compliance asks whether the scheduled PMs were completed on time (execution). Schedule compliance asks whether the whole week's scheduled work, PM and corrective, got done in its window. A department can hit high PM compliance while PMP stays low, because its corrective work is all reactive. Read the three together, not one alone. And remember the failure-detection side: catching problems on the P-F interval is what lets you plan repairs instead of reacting, which lifts every one of these metrics at once.

Where PM compliance fits your reliability program

PM compliance is the first metric most plants should get right, because it is where proactivity becomes visible and because it is cheap to move. It belongs on the weekly review next to backlog and PMP, broken out by criticality, with real misses shown rather than hidden. Held high on a schedule that has been audited for value, it drives the equipment reliability gains that show up months later in downtime and cost, and it is the metric that tells you whether the discipline of total productive maintenance is actually taking hold on the floor.

The measurement depends entirely on trustworthy completion data: PMs with defined windows, checklists that must be filled to close, and misses that get logged instead of quietly reopened. When that lives on paper or in a system nobody trusts, PM compliance becomes theater. Capturing completions honestly, enforcing the window and the checklist, and putting compliance by asset on a live dashboard is the pattern on our platform overview and in a CMMS done well; the CLS case study shows automated reporting that makes a metric like this believable, no rip-and-replace of the systems already running.