Ask around and you’ll hear the same answer: a good PM compliance rate is 90%. The figure commonly appears on CMMS dashboards and in vendor guidance.
The number is real, and it comes from a real place. SMRP lists above 90% as the best-in-class target for its PM and PdM Compliance metric. The trouble starts with what gets left out. That target is a goal SMRP prescribes, not a median anyone measured across industry, and the same program can post very different compliance numbers depending on which on-time rule it uses.
Get the definition right and above 90% earns its place on the dashboard. Quote it without the fine print and an experienced reliability engineer will pick it apart.
What PM compliance measures, and where the numbers come from
PM compliance is SMRP metric 5.4.14, “Preventive Maintenance (PM) and Predictive Maintenance (PdM) Compliance.” The formula is plain: PM and PdM work orders completed by their due date, divided by PM and PdM work orders due. It’s a work-order count, so a two-minute lubrication route and an eight-hour overhaul each count as one work order.
The part that moves the number is the definition of “on time.” The 6th-edition metric offers three completion criteria: completed by the required date, completed by the required date plus one day, or completed by the required date plus 20% of the PM interval, capped at a maximum of 28 days. Whichever criterion you pick has to be applied consistently. A monthly PM under the 20% rule gets about six extra days. The 28-day cap stops an annual PM from inheriting a 73-day grace window.
There’s a second, related SMRP metric that gets tangled up with the first. PM and PdM Work Order Compliance (metric 5.4.10) looks at each completed work order’s overdue variance and groups the results by variance range and asset criticality, rather than reporting one aggregate on-time rate. Its target section names the “10% rule” as the recommended goal: a time-based PM counts as compliant only if it’s completed within 10% of its interval. Allied Reliability illustrates the rule with a monthly PM landing within about 1.5 days on either side of its due date. Other practitioners read the 10% allowance differently (Jeff Shiver, for one, treats it as roughly plus or minus three days on a 30-day PM), which is one more reason to state your implementation.
That 10% goal sits in apparent tension with the 20% completion option in 5.4.14, and SMRP practitioners have publicly asked for the two to be reconciled. The practical fix is to state which metric and which timing rule you’re using rather than treating them as interchangeable.
One more metric to keep separate: PM and PdM Yield measures corrective-work hours found through PM and PdM, divided by PM and PdM hours. SMRP gives it no universal target number; it flags both very low and very high values as worth investigating, and the right range varies by plant.
The Reliable Confidence Score
PM Compliance Rate Benchmarks: the Reliable Confidence Score
We rate the confidence in each claim, not just the number attached to it.
The Big Takeaway
The honest answer to “what’s a good PM compliance rate” is above 90% against a consistently applied on-time rule, read as a trend, and paired with a reliability outcome so completions can’t be gamed.
Above 90% is SMRP’s prescribed target, measured against a fixed on-time rule. Read it as a goal, hold the on-time criterion steady, and it earns its place on the dashboard. The number that drifts is the one nobody defined.
What you can defend without flinching: the definition, the completion criteria, and the direction of travel. A program climbing from 62% to 84% over two quarters may be healthier than one parked at a flat 90% nobody trusts.
Why the numbers vary
Most of the disagreement comes from four choices that rarely get stated out loud.
The completion criterion. Required date, required date plus one day, the 20% rule, or the stricter 10% goal each produce a different percentage from the same work. SMRP’s answer is to pick one and apply it consistently. A program that quietly loosens its window can post a “rising” compliance rate that reflects a definition change, not better execution.
Count versus hours. SMRP 5.4.14 counts work orders. Some sites build an hours-based version so a big overhaul carries more weight than a filter change. Both can be useful. They produce different numbers, and only the count version is the SMRP metric.
What sits in the denominator. SMRP uses “work orders due.” Rolling overdue PMs forward into next period’s count, or writing them off, are local reporting choices. Neither is the SMRP calculation, and two plants doing identical work can post very different rates if they handle overdue PMs differently.
Schedule compliance bleeding in. Teams quote a weekly schedule-compliance figure and call it PM compliance. Doc Palmer’s argument lives here: a weekly schedule that routinely clears 90% usually wasn’t fully loaded, because a normal plant absorbs around 20% urgent break-in work, so he treats roughly 80% as the realistic mark for a fully loaded schedule. SMRP sets a high target for schedule compliance, Palmer questions the wisdom of chasing it, and that debate is about schedule compliance, a separate metric from PM compliance.
How to use the benchmark safely
Pick one completion criterion and lock it in the CMMS, so every site and every report means the same thing by “on time.” This is the highest-leverage move, and it’s what a CMMS is built for: the system applies the same grace-period rule to every work order automatically, rather than leaving it to whoever closes the ticket.
Then read the rate as a trend, and segment by asset criticality where you can, because a fleetwide average hides whether your important assets are the ones slipping. Pair compliance with PM and PdM Yield (the measure of what your PMs are finding) and at least one reliability outcome such as mean time between failures, so a high completion rate has to be backed by results.
Where teams go wrong
Closing PMs on paper. Auto-closing overdue work orders can distort compliance if they are excluded from the calculation or recorded with inaccurate completion dates; back-dating completions directly inflates it. A 95% rate sitting next to a falling MTBF deserves a look, though the cause can be paper completions, the wrong PMs, or something outside the PM program entirely such as asset aging, an operational change, or a new failure mode.
Conflating the two metrics. Reporting schedule compliance and calling it PM compliance, or holding one to the other’s target, produces goals that fight each other.
Chasing 100%. A perfect score can signal an under-sized PM program or a window so loose that “on time” means little. Legitimate 100% is possible; it just warrants a second look.
Tracking compliance without yield. PM compliance tells you the work got done. PM and PdM Yield tells you whether the work is finding anything. Track them together or you’re measuring activity in isolation.
Methodology
We rated the confidence in each claim, not just the figure. The definition, the three completion criteria, and the PM and PdM Yield formula were checked against the text of the SMRP metric and rated High. The above-90% target is rated High as a prescribed SMRP best-in-class value; it’s a target rather than an observed average, the exact value sits in the paid 6th-edition guide, and sources citing that edition report it consistently. Round-number thresholds with no basis in the metric (95% for critical assets, sub-80% “problem” lines) are rated Low even where they’re directionally sensible, and we name them as CMMS vendor conventions rather than standards.
We treat PM and PdM Compliance (the aggregate on-time rate, metric 5.4.14) and PM and PdM Work Order Compliance (the variance-based metric, 5.4.10, whose recommended goal is the 10% rule) as two distinct SMRP metrics. The negative finding is scoped on purpose: in the standards and in Plant Engineering’s 2020 maintenance study, which tracks PM adoption rather than achieved compliance, we found no published distribution of the rates plants actually hit, so it’s rated Medium rather than stated as a universal absence.
The Short Version
PM compliance is the share of due PM and PdM work orders completed by their due date, the way SMRP defines it in metric 5.4.14. SMRP prescribes above 90% as best-in-class, but that’s a target, not a measured average, and it only means something once you fix the on-time rule (required date, plus one day, or plus 20% capped at 28 days) and apply it consistently. Lock one rule in your CMMS, watch the trend, segment by criticality, and pair compliance with PM and PdM Yield so completions don’t get gamed.
Sources
- Society for Maintenance and Reliability Professionals (SMRP), Best Practices, Metrics and Guidelines, 6th Edition: PM and PdM Compliance (metric 5.4.14), PM and PdM Work Order Compliance (metric 5.4.10), and PM and PdM Yield. https://smrp.org/Learning-Resources/SMRP-Library/Best-Practices-Metrics-Guidelines
- Doc Palmer, “What does schedule compliance measure?” Plant Services. https://www.plantservices.com/planned-maintenance/planning-and-scheduling/article/11290088/what-does-schedule-compliance-measure
- Doc Palmer, “Is it okay to break the weekly maintenance schedule?” Plant Services. https://www.plantservices.com/planned-maintenance/planning-and-scheduling/article/33004872/is-it-okay-to-break-the-weekly-maintenance-schedule
- Richard (Doc) Palmer, Maintenance Planning and Scheduling Handbook, 4th edition, McGraw-Hill.
- Jeff Shiver, “30 days’ grace? Sorry, not so with a PM,” Plant Services (on applying the 10% rule). https://www.plantservices.com/home/blog/11300684/30-days-grace-sorry-not-so-with-a-pm
- Allied Reliability, “5 Tips for Optimizing a Preventive Maintenance (PM) Program” (states the 10% rule’s ±1.5-day reading for a monthly PM). https://www.alliedreliability.com/blog/5-tips-for-optimizing-a-preventive-maintenance-program
- SMRP All Member Open Forum, “Conflict between metric 5.4.10 and 5.4.14?” (members quote 5.4.10’s 10% rule and 5.4.14’s required-date-plus-20% option from the Best Practices guide). https://exchange.smrp.org/discussion/conflict-between-metric-5410-and-5414?hlmlt=VT
- eWorkOrders, “Preventive Maintenance KPIs” (source of the CMMS-vendor 95% critical-asset and sub-80% conventions noted in the table). https://eworkorders.com/preventive-maintenance/preventive-maintenance-kpis/
- Plant Engineering, 2020 industrial maintenance study (documents PM adoption, e.g. 88% of facilities follow a PM strategy, not achieved compliance rates). https://www.plantengineering.com/the-maintenance-function-like-manufacturing-itself-is-a-rapidly-changing-environment/
