I’ve been thinking lately about how much time we spend looking at maintenance KPIs in manufacturing. We track preventive maintenance completion, equipment breakdowns, MTBF, MTTR, OEE, maintenance costs, schedule compliance, and just about anything else we can measure.
We put the numbers on dashboards, review them in meetings, compare them with last month’s results, and spend a lot of time discussing why something is red instead of green. And don’t get me wrong, I think maintenance KPIs are important. Without them, we’d have a hard time understanding how well we’re performing or whether the changes we’re making are actually working.
But there’s something I’ve noticed over the years that I think deserves more attention. Sometimes we become so focused on improving the numbers that we forget to ask whether we’re improving the way we work.
And those aren’t always the same thing.
When a Better Number Doesn’t Mean Better Maintenance
Let’s say we’re reviewing our monthly maintenance performance and notice that preventive maintenance (PM) completion is sitting at 70%. Our target is 90%, so naturally, we want to close that gap.
We discuss it in the meeting, remind everyone how important PM compliance is, and ask the team to make it a priority. The following month, completion reaches 90%. Great news, right? Well, maybe.
What if some of those PMs were rushed just to get the work orders closed? What if inspections identified problems that nobody followed up on? Or what if we’re still performing preventive maintenance tasks that haven’t been reviewed in years and no longer address the equipment’s actual failure modes?
On paper, we’ve made a significant improvement. But have we really made the equipment more reliable? That’s the question I think we need to ask more often.
I’m not suggesting that people deliberately manipulate numbers. Sometimes they’re simply doing what the organization has asked them to do. If we keep telling everyone that PM completion must reach 90%, we shouldn’t be surprised when most of the attention goes toward reaching 90%. The trouble is that completing a maintenance task and completing an effective maintenance task are two different things.
A PM program should help us prevent failures, detect developing problems, and keep equipment operating as intended. If it isn’t doing that, a high completion percentage doesn’t tell us the whole story. This is also why I believe in regularly reviewing and improving preventive maintenance programs. It’s not enough to complete the work. We need to make sure we’re doing the right work in the first place.
What Are the Numbers Actually Telling Us?
Now let’s look at equipment breakdowns. Imagine our monthly report shows that unplanned downtime has increased beyond the acceptable target. We could immediately tell the maintenance team that downtime needs to come down next month. We could make it a priority in every meeting and ask for daily updates until the number improves.
But what exactly have we changed? We haven’t necessarily improved equipment condition, eliminated a recurring failure, or addressed anything that caused the downtime. We’ve simply made everyone more aware of the number.
I’d rather start by understanding what’s behind it. Are we experiencing repeated failures on the same equipment? Are our inspections detecting problems early enough? Do technicians have the information and spare parts they need? Are we completing corrective work before minor defects turn into breakdowns?
Maybe the problem starts with how work is identified and prioritized. Maybe we’re postponing repairs because production can’t release the equipment. Or perhaps we’re repeatedly replacing the same component without understanding why it keeps failing.
These are the conversations that can actually lead to better reliability. And they’re very different from simply telling people to reduce downtime.
The KPI is still valuable because it tells us something needs attention. But it doesn’t automatically tell us what needs to change.
That’s where our understanding of the maintenance system becomes important.
The Problem With Chasing Targets
There’s another side to KPI management that I think we sometimes underestimate. People naturally pay attention to what they’re measured on. And when a particular number becomes the main measure of success, it’s easy for that number to influence decisions in ways we didn’t intend.
Take maintenance costs, for example. If the message from management is that costs must come down, postponing certain maintenance activities might make the monthly report look better.
But what happens six months later when deferred work turns into a major equipment failure? We’ve reduced today’s maintenance cost at the expense of tomorrow’s reliability.
The same thing can happen with other indicators. If we focus too heavily on mean time to repair (MTTR), technicians may feel pressured to get equipment running as quickly as possible, even when a little more time spent understanding the problem could prevent another breakdown. If schedule compliance becomes the priority above everything else, we might become reluctant to adjust the schedule when equipment conditions change. And if we’re only concerned about closing work orders, we may miss opportunities to capture information that could help us improve future maintenance activities.
None of this means these maintenance KPIs are bad. They’re useful measures, and we should absolutely pay attention to them. But we need to understand what behaviour we’re encouraging when we set targets.
I think a good question for any leader is whether the way we’re measuring performance is helping people make better decisions or simply encouraging them to produce better-looking reports.
Because a dashboard can be full of green indicators while the people working on the equipment know that things aren’t going nearly as well as the numbers suggest.
The Maintenance System Is Bigger Than the Dashboard
When I talk about improving the system, I’m not referring to one particular procedure or a piece of software. I’m talking about everything that goes into keeping a plant running reliably.
How do we identify maintenance work? How do we decide what’s most important? Are jobs properly planned before they’re scheduled? Do we have the necessary parts, tools, and technical information? Are we recording what we discover during repairs? Are we learning from failures? And how well do Maintenance and Operations work together when equipment needs attention?
All these things are connected. For example, a technician might do an excellent job repairing a machine, but if the spare parts weren’t available when needed, the equipment could still be down for hours longer than necessary. A planner might prepare a detailed job package, but if production can’t release the equipment at the agreed time, the work may be postponed. Or a technician might discover the cause of a recurring failure, but if that information never makes it into the maintenance management system, the next person may have to troubleshoot the same problem all over again.
Improving reliability means looking at how these activities work together, not just measuring the final result.
And sometimes, improving one part of the process can have a much bigger impact than asking everyone to work harder.
Leadership Makes a Difference
This is where I believe leadership has an enormous influence on the results we get.
Imagine two maintenance managers looking at the same report. Both see that PM completion has dropped below target.
The first manager asks why the team isn’t completing the required work and tells everyone to improve the number.
The second manager asks what’s preventing the team from completing the work properly.
Maybe the PMs are being scheduled when equipment isn’t available. Maybe the team is spending too much time responding to emergencies. Maybe the workload is unrealistic, or perhaps the job plans aren’t clear enough.
The second manager isn’t ignoring the target or removing accountability. They’re trying to understand what’s happening before deciding what to do about it.
And I think that’s an important distinction.
Good leadership isn’t just about setting expectations and following up on results. It’s also about making sure people have the conditions they need to meet those expectations.
That’s one of the reasons I’ve always valued consistent leadership practices and daily management. When leaders stay connected to what’s happening in the operation, they’re in a much better position to recognize problems early and help their teams address them.
Of course, sometimes the issue really is poor execution or a lack of accountability. We shouldn’t ignore that possibility.
But we should make that determination based on what we discover, rather than assuming the people doing the work are always the problem.
A Simple Way to Think About Maintenance KPIs
I like to think of a KPI as being similar to a thermostat.
A thermostat tells us the temperature in a room. If it’s too cold, we know something needs attention. But staring at the thermostat, checking it every five minutes, or telling someone that the temperature needs to improve won’t make the room warmer.
We need to understand what’s happening. Maybe the heating system isn’t working properly. Maybe a window has been left open. Maybe the insulation isn’t good enough.
Once we address the actual problem, the temperature changes.
I think the same principle applies to maintenance performance.
If breakdowns are increasing, we need to understand why. If maintenance costs are rising, we need to investigate what’s driving them. If schedule compliance is poor, we should look at how work is planned, prioritized, and coordinated.
The KPI gives us a reason to investigate. It doesn’t replace the investigation.
And just like a thermostat, we need to make sure the indicator itself is giving us accurate information. Otherwise, we could spend a lot of time trying to solve a problem that isn’t what we think it is.
How Do We Know an Improvement Is Real?
Here’s something else worth considering. Let’s say we make a change to our maintenance process, and the following month our maintenance KPIs improve. How do we know the change actually worked?
Maybe we had fewer breakdowns because the equipment wasn’t running as much. Maybe maintenance costs dropped because several planned jobs were postponed. Or perhaps PM completion improved because the schedule contained fewer tasks that month.
The numbers may have changed for reasons that have little to do with the improvement we introduced.
That’s why I think it’s important to look beyond a single month’s results. We need to understand what changed in the process, whether people are following the new approach, and whether the improvement continues over time.
If we’ve improved work planning, are technicians spending less time waiting for parts and information? If we’ve revised our PM strategy, are we detecting more potential failures before they cause downtime? If we’ve introduced a new way to prioritize work, are critical defects being addressed sooner?
Those observations help us understand whether we’re making real progress.
And this is where effective daily management becomes valuable. Regular conversations, follow-up, and attention to what’s actually happening on the plant floor can tell us things that a monthly dashboard might miss.
To me, sustainable improvement means the process is working better, not just that the latest report looks better.
So, Should We Stop Focusing on Maintenance KPIs?
Not at all.
I wouldn’t want to manage a maintenance organization without good performance indicators. We need them to understand where we are, identify trends, set priorities, and see whether our efforts are producing results.
But I think we need to be careful about the role we give them. A KPI should help us ask better questions, not replace the need to ask questions.
When a number is below target, our first reaction shouldn’t always be to push people harder. We should try to understand what is happening and what needs to improve. And when a number looks good, we shouldn’t automatically assume everything is working perfectly.
We should be just as curious about why performance is good as we are about why it’s poor. Because if we understand what’s producing good results, we have a much better chance of repeating them.
For me, that’s the real value of performance measurement. It’s not about having the most impressive dashboard or making sure every indicator is green at the end of the month. It’s about using the information we have to make better decisions, improve how we work, and build a more reliable operation.
Maybe instead of asking, “How do we improve this KPI?” we should start asking, “What needs to improve in our system for this KPI to get better?”
The difference might sound small, but I think it can completely change the conversation. And over time, it can change the results too.
For more thoughts on maintenance, reliability, and leadership, feel free to connect with me on LinkedIn.
