How to Uncover the Causes of High Costs and Variability in your Maintenance Organization

Most sites have no shortage of KPIs.  You can see cost, downtime, PM compliance, reactive work, backlog size, MTBF, and plenty of other measures.

But what happens when a KPI moves in the wrong direction?  When performance varies from month to month, department to department, or one site to another, it’s difficult to know the underlying causes and how to fix them.  The KPI tells you what’s happening – it doesn’t necessarily tell you why.  When you attempt to dig into the details it’s frustrating to get confusing and conflicting points of view.  You come away not knowing if you can trust the story behind the numbers or what’s causing them.

You shouldn’t have to dig through every work order to understand what’s happening.  Leaders shouldn’t have to become maintenance analysts to make good maintenance decisions.

The answer isn’t more data.  What you need is insight – the ability to know what actions will drive the results you want.

There’s nothing in the data that will tell you what you need to do.  The objective is to focus on the cause of the business problem and move from KPIs to actions.

KPIs → Insights → Decisions → Actions

The confidence for decisive action only comes through insight.  Here’s something that maintenance professionals and site leaders often miss: insights are made, not found.  

A structured approach is necessary to make insights that can lead to action.  It starts by defining what matters to the business and identifying the questions that need to be answered.  It then uses the right combination of data and targeted investigation to gather evidence.  Finally, the causes are translated into a plan of action that feeds fact-based decision making.

Discover: Start with What Matters

Define the Problem

The first step is to define the business problem before getting into the details of maintenance performance. 

If maintenance costs are increasing it’s tempting to focus 100% of your attention on that one KPI.  Since maintenance costs are made up primarily of labor and materials most business leaders jump straight to solutions and assume tighter control is needed.  How often have you seen a ban on overtime or Maintenance Managers having to start approving POs above some arbitrary dollar amount? 

A better approach would be to understand the overall situation and see maintenance costs in context.  Often, the business is much more concerned with the overall cost of goods sold which is impacted by much more than overtime and maintenance material purchases.  Consider how framing the problem as maintenance costs vs cost of goods sold will lead to different possible solutions.

If the focus becomes maintenance costs the possible solutions are limited and may even be counterproductive.  Focusing on overtime and parts often leads to increases in contract labor and deferred repairs, both often more costly than the original problem.

The goal is to understand what’s contributing to the business problem and what actions need to be taken.  Starting with the business problem helps keep the analysis from assuming the answer before the investigation begins.

Frame the Investigation

Once the problem is clear, you can determine how to investigate.  This is where tools such as decision trees and hypotheses become useful.

The tools aren’t the objective. They help turn a broad business problem into questions that can be answered with evidence.  The goal isn’t to collect every available piece of data, it’s to identify the right evidence to answer the questions that matter.

This creates a focused path for the analysis and a repeatable way to make insights:

Business Problem → Questions → Hypotheses → Tests →  Evidence

Instead of treating a KPI as the answer, you use it as a starting point to help determine where to look and what you need to understand.

When facing a situation where costs are climbing or variability is increasing between sites, first consider what problem you want to solve and ask yourself a few questions:  

• Is there a real business problem?  
• How can I look beyond the immediate to focus on what matters most?
• How will I keep an open mind before jumping to conclusions?
• What process will I use to test my reasoning and make informed decisions?
• How does Maintenance fit into the overall business strategy and what other organizational elements are at work?
• What is the ideal outcome I want to see in the next 12 to 24 months (not just the next month)?

By focusing on the right problem and approaching it with a structured process the investigation and solutions will be of much better quality. 

Analyze: Turn Data into Insight

Once you know what business problem you’re trying to solve and what you need to understand, the next step is to gather and evaluate the evidence.

If our initial investigation narrows in on costs as a major component of cost of goods sold we will need to determine what parts make it up and the contribution of each. 

You might start with production and maintenance data:

• Production output vs costs
• Equipment downtime and production losses
• MRO inventory value over time and stockout rates
• Labor broken down by internal and contractor costs

The objective isn’t to find a number that proves maintenance is the problem.  It’s to understand the relationship between maintenance activity and costs.

Bring Different Perspectives Together

Numbers are only one source of evidence.  People working in the operation often see things that aren’t obvious in the data.

In our example, the production data might show that a particular line is responsible for a disproportionate share of high costs. An operator might explain that the line routinely runs under conditions that aren’t reflected in the equipment history. A technician might recognize that many of the failures appear to have the same symptoms. A planner might point out that repairs are frequently delayed because parts aren’t available.

Each perspective creates a potential avenue for investigation.

Follow the Evidence Deeper

Once you identify a potential driver, keep asking questions.

Perhaps the data shows that the troubled production line has repetitive failures involving the same conveyor drive.  Work-order history reveals constant corrective repairs.  Interviews indicate that operators frequently notice abnormal vibration before the failures occur, but there is no formal condition-monitoring practice in place.

The analysis might reveal an opportunity to introduce condition monitoring, change the maintenance strategy, address an operating practice, improve a repair process, or conclude that the equipment isn’t actually the right place to focus.

Pareto analysis, trend analysis, failure history, work-order analysis, equipment history, interviews, and targeted investigation are all useful when they help narrow the path from what is happening to why it’s happening.  The specific tool matters less than the question it helps answer.  

The objective is to keep narrowing the problem until you understand enough to make a decision.

Deliver: Move from Insight to Action

Once the causes of the business problem are understood, the final step is to translate understanding into action. This is where insight becomes valuable.

Returning to our example, now that the analysis has answered the question of what is happening and why, leadership needs to answer a different question: “What are we going to do about it?”

That decision might involve changing the maintenance strategy, addressing an operating practice, improving planning and scheduling, investing in a reliability improvement, or some combination of actions.  The important point is that the recommendation should follow from the insight, not from preconceived notions. 

Prioritize the Opportunities

Analysis will often uncover more opportunities than an organization can realistically pursue.  The goal isn’t to fix everything at once – it’s to identify the actions that have the greatest potential business impact and create a practical path forward.  A useful way to think about each opportunity is:

Opportunity – Cost = Net Benefit

This is a fundamental business case and answers three questions:

1. What’s the potential improvement on the bottom line?
2. What will it take to achieve it?
3. Does the expected benefit justify the investment?

For our example, eliminating a recurring equipment failure might create significant additional production capacity.  But the solution could range from a relatively simple maintenance strategy change to a major equipment modification.  The right decision depends on the expected benefit, cost, and risk.

Close the Loop

The process doesn’t end when the decision is made.

Once actions are implemented, the organization should return to the original business problem and ask: “Did the action produce the expected result?”

If conditions improve the analysis has been validated.  If the expected improvement doesn’t occur, that’s information too. Perhaps the assumed cause wasn’t the primary driver, the solution wasn’t effective, or another constraint is limiting the result.  This creates a continuous learning cycle.

The KPI returns at the end of the process, not as the answer, but as a way to determine whether the decision produced the intended outcome.

Moving beyond KPIs doesn’t mean abandoning them. It means putting them in their proper place.

KPIs → Insights → Decisions → Actions

Start with the business problem. Ask the questions that matter.  Gather the evidence needed to answer them.  Challenge what you think you know.  Then translate what you learn into decisions and actions that can improve the business.

The result isn’t simply a better understanding of maintenance performance – it’s the confidence to act.