Here is the short version. Both can feel true at once, but the cause is almost always one of two things. Either your compliance number is hollow, counting work orders that closed without the work actually being done, or your PMs are getting done on time and they are simply the wrong PMs doing the wrong things. Compliance only measures whether a work order closed inside your window. It tells you nothing about whether the work happened, and nothing about whether it was worth doing.
Compliance is a closure rate, not a work rate
When your dashboard says 95 percent, what it is really saying is that 95 percent of the PM work orders that came due got marked complete on time. That is it. A completed work order can represent a thorough, careful PM, or it can represent a record someone set to complete at the end of the month to keep the number green. The metric cannot tell the difference, and neither can you from the percentage alone.
The fastest way to see which one you have is to open the completed PM work orders and look at whether they are filled out. A real PM leaves evidence. There are labor hours and a labor cost on the work order, and if the job calls for parts, there is a material cost. A PM that closed with no labor hours and no materials, on a job that should have taken a tech an hour and a couple of parts, is telling you something. Maybe the work happened and nobody reported the time, which is its own problem. Or the work never happened and the order was closed anyway. Either way, that completion is not something you can trust.
Even a real completion fails if it was the wrong work
Say the work did happen. The tech showed up, recorded the time, closed the order honestly. You can still be at 95 percent and still be failing, because doing a PM on time is not the same as doing a PM that matters.
This is where the PM task detail earns its keep. If your job plan task says "inspect pump," you have left everything to the technician's memory, training, and mood on that particular day. One tech checks the seal, the coupling, the vibration, the oil level, and the mounting bolts. The next tech walks up, looks at it, and moves on. Both close the work order on time. Both count toward your 95 percent. Only one of them did maintenance.
Work depends on the tech. No evidence trail. Unverifiable completion.
Inspect coupling alignment
Record vibration reading (flag >4 mm/s)
Check oil level and color
Verify mounting bolt torque
Work is defined. Evidence is built in. Completion is verifiable.
My recommendation, and this one costs you nothing but planner time: get granular with your Job Plan tasks. A job plan task that lists the specific checks, the readings to take, and the limits that should trigger a follow-up makes the right work more likely to happen and leaves evidence on the record.
Read compliance next to reactive work
The single best test of whether your compliance number is real is sitting right next to it. Put PM compliance and reactive work percentage on the same screen and watch them over a few months. When the program is real, they move in opposite directions. Compliance climbs, reactive work falls, because the PMs you are completing are actually preventing the failures that used to drive emergency work.
If compliance is high and reactive work is flat or climbing, you have your answer. The PMs are either not really being done, or they are being done and they are not the right work on the right assets at the right frequency.
"But my crews would never fake a completion"
I believe you, and that is not usually what is happening. Deliberate fraud is rare. What is common is far more mundane. A status gets set in bulk at period end to clear the list. An automation closes PMs under certain conditions and quietly sweeps up more than it should. Labor never gets reported because reporting it is a hassle. Tasks are vague, so the work drifts. None of that requires a dishonest technician. It just requires a system that rewards closure and a number nobody audits.
So what do you do Monday
Pull your completed PM work orders from the last month. Look for the ones with no labor and no materials, and ask whether the work really happened. Open a handful of your PM job plan tasks and read them as if you were the tech, and ask whether they tell you what to actually do. Then put compliance and reactive work on the same chart. Those three checks will tell you, in about an hour, whether your 95 percent is real.
If you want a structured way to run that check, the PM Compliance Self-Assessment walks through the six questions that separate a real number from a hollow one. The PM Compliance Calculator will help you recalculate once you strip out the closures that were not work. Both are free.
This article covers why a high compliance number can coexist with ongoing failures. The Maximo KPI Guide to PM Compliance walks through the full implementation from compliance window definition through KPI Manager SQL, segmentation by PM vs. PdM, craft, site, and asset criticality, the drilldown from aggregate to root cause, and the business case to justify a compliance program overhaul.
Frequently Asked Questions
Why is my PM compliance high but my equipment still failing?
Either your compliance number is hollow, counting work orders that closed without the work actually being done, or your PMs are getting done on time and they are simply the wrong PMs doing the wrong things on the wrong assets. Put PM compliance and reactive work percentage on the same chart. If compliance is high and reactive work is not falling, one of those two problems is present.
What does PM compliance actually measure?
PM compliance measures whether PM work orders that came due were marked complete inside your compliance window. It is a closure rate. It does not measure whether the work was performed, whether labor and materials were used, or whether the PM was effective at preventing failures.
How can I tell if my PM compliance number is real?
Open your completed PM work orders and look for evidence. A real PM leaves labor hours, material costs, and task completions on the record. PMs that closed with zero labor and zero materials on jobs that should have taken time and parts are completions you cannot trust. Then put compliance next to reactive work. If reactive work is not falling as compliance rises, the number is not doing what it should.
Does a completed PM work order mean the work was done?
Not necessarily. A completed work order means the record reached a completed status inside your system. That can represent a thorough PM, or it can represent a record someone set to complete without performing the work, or an automation that closed it. The status alone does not confirm the work happened. Labor hours, materials, and task evidence confirm the work happened.
Quick Maximo questions are always free. Reach out on LinkedIn. I never charge for chatting.