How Executives Who Sustain Great Reliability Programs Think About Industry Comparisons — and What They Measure Instead
The programs that keep getting better aren’t chasing someone else’s numbers. They’re tracking what matters in their own operating context.
The Idea Worth Examining
Most executives who fund reliability programs have, at some point, reached for an industry benchmark. It makes sense. When a board asks whether the maintenance spend is appropriate, or whether OEE is where it should be, the instinct is to find a number that answers the question. What does a good program look like? What are strong performers achieving?
That instinct is reasonable. The executives running the most durable programs I’ve seen share it. Where they’ve ended up, though, is somewhere more interesting than the benchmark table itself.
I want to explore the layer they’ve added to this — because it changes how they invest, how they evaluate their programs, and how they make the case to the rest of the leadership team.
Why the Benchmark View Makes Sense
Industry benchmarks on reliability exist for good reasons. Metrics like OEE, mean time between failures, maintenance cost as a percentage of replacement asset value, and planned maintenance compliance give executives a common language. They allow comparisons across sites, years, and peer companies. When an operation is in early stages of building reliability capability, knowing that world-class plants achieve 85 percent or better OEE, or that mature programs run planned maintenance rates above 80 percent, provides a useful orientation point.
Benchmarks also serve a governance function. A board or investment committee asked to fund a multi-year reliability program needs an external reference. What are we funding this toward? What does good look like? The benchmark provides an anchor for that conversation.
For these purposes, benchmarks are legitimate tools. The executives I’ve worked alongside have used them this way. They haven’t abandoned them. They’ve just learned not to confuse the benchmark for the target.
What the Best Operators Add to This View
The shift I’ve observed in the programs that keep improving is relatively straightforward, but its implications run deep. The executives running those programs understand that a benchmark represents an average — or occasionally a top-quartile figure — drawn from a population of plants that do not share their specific operating context.
That context matters enormously. Two plants reporting the same OEE number may be in entirely different positions. One may be running mature, well-understood equipment in a stable production environment, where that 82 percent reflects years of disciplined work. Another may be running a complex, high-mix process with older assets, where the same number represents a significant achievement given the constraints. The benchmark cannot see any of this.
Nowlan and Heap’s foundational work on reliability-centered maintenance made this point at the engineering level: the appropriate maintenance strategy for any asset depends on its specific operating context, failure mode profile, and consequence of failure — not on what other plants do with nominally similar equipment. The same logic applies at the investment level. The right level of reliability investment for a given facility is determined by its own risk profile, its production criticality, its asset age and condition, and the cost of failure in its specific market context.
The executives who have internalized this don’t stop using benchmarks. They use them to open conversations, and then move quickly to the question that actually drives investment decisions: What does our program need to do, given what we’re running and what failure costs us?
The Fuller Mental Model — Four Additions That Change How You Think About This
The best programs I’ve seen add four layers to the standard benchmark view. None of these replace the benchmark. They just give it the context it needs to be useful.
What This Changes About How You Operate
For executives making investment decisions, the fuller mental model shifts a few things in practice.
The Board Conversation Becomes More Specific
When you can explain that your reliability investment is calibrated to the cost of failure in your specific operation, and that you’re tracking improvement against your own baseline, the governance conversation gets more traction. You’re not defending a number against a benchmark. You’re showing a trajectory against a risk-adjusted standard set by the business itself.
The Capex Case Gets Easier to Make
Sustaining capex for reliability is one of the harder investment categories to justify, particularly in years when operating budgets are under pressure. The programs that do this consistently well have learned to frame it in terms of failure consequence avoided — not maintenance spend normalized. What does it cost us when this type of failure occurs? What has our frequency of that failure type been? What investment changes that trajectory? This is fundamentally different from ‘our maintenance cost as a percentage of RAV is above the industry average.’
The Team’s Work Gets Properly Recognized
One consequence of benchmark-chasing that I’ve seen play out repeatedly is that it can make a strong program invisible. A plant running difficult equipment in a demanding environment may be performing exceptionally well by any reasonable measure, but show up in the middle of the benchmark table because the industry average was built on easier operations. The executives who understand this can recognize and reward the actual quality of their team’s work — rather than a percentile ranking that doesn’t account for the challenges they face.
The Investment Decisions Get More Durable
When reliability investment is anchored to the organization’s own risk profile and improvement trajectory, it tends to survive leadership transitions and budget cycles better. The programs that go backwards are often those where the investment rationale was always external: we need to match benchmark. When leadership changes or the benchmark shifts, the foundation isn’t there. Programs anchored to consequence, trajectory, and asset-level criticality have a rationale that survives those transitions.
If you set your industry benchmark table aside for a moment and looked only at your own program — at the trend line on your critical assets, at the cost of your most frequent failure modes, at the gap between planned and reactive work — what would you see?
That picture is the most honest read on where your program stands, and where the next investment creates the most value.
The executives who have navigated this well have found that the benchmark is most useful as a starting point for that internal conversation — not as its conclusion.
I’ve had versions of this conversation on plant floors and in boardrooms across a range of industries. The details vary, but the pattern is consistent: the programs that keep improving are the ones that have built their own standard of excellence, one calibrated to their operation, their assets, and what failure actually costs them. That’s a more demanding target than a benchmark table. It’s also a more honest one.
Alain Pellegrino
Alain is President of Reliability Solutions and brings over 25 years of expertise in industrial reliability and predictive maintenance. He began his career as a PDM Consultant and predictive maintenance coordinator before spending nearly two decades at Laurentide Controls, where he rose to Vice President of Industrial Reliability Solutions. He now leads Reliability Solutions with a focus on helping organizations build world-class reliability programs.
