Same Company. Same Budget. Four Sites. Wildly Different Results. Here's Why.
Same company, same commodity, four Australian gold mining sites. A structured maturity assessment revealed why maintenance results varied so dramatically across the portfolio and what to do about it.

When a gold mining group asked us to benchmark maintenance management across four of their Australian operating sites, they expected to find some variation. What they did not expect was the scale of it.
Same parent company. Same enterprise systems. Same maintenance budget structure. Same leadership expectations. And yet, when we applied a consistent maturity framework across all four sites and assessed both maintenance crews at each, the gap between the best and worst performing crew in the portfolio was larger than the gap between their best crew and industry best practice.
That finding tends to stop people in their tracks. It stopped this group's executive team too.
Here is what we found, and more importantly, why it happened.
The Four Sites
The portfolio covered four operating gold mines, each with its own orebody characteristics, workforce profile, and maintenance history. Same commodity. Same basic equipment mix. Same company standards on paper.
So Why Did the Results Differ So Much?
The honest answer is that it almost never comes down to budget, equipment, or technology. Those factors matter at the margins. The differences that drive a 1.4 point maturity gap between two crews on the same site come from somewhere else entirely.
Leadership at the crew level, not just the site level
At Site B and Site D, the gap between crews was small because frontline supervisors actively transferred knowledge, held consistent standards, and treated planning discipline as non-negotiable. At Site A and Site C, crew performance was heavily personality-dependent. When a strong supervisor moved on, standards drifted because the processes were not embedded deeply enough to survive the transition. Good practice left with the person rather than staying in the system.
Work order quality compounds over time
The difference between a CMMS that captures failure modes, causes, and corrective actions, and one full of fixed pump entries, seems minor on any individual job. Across five years and thousands of work orders, it becomes the difference between a data-driven maintenance strategy and guesswork. Site C's data problem did not happen overnight. It accumulated across years of inconsistent standards, and by the time we assessed it, the downstream effects were visible in every dimension from PM effectiveness to spares management.
Contractor integration is either a force multiplier or a drag
Site B's contractor management score reflected a deliberate decision to stop treating contractors as a separate workforce and start treating them as an extension of the team. Shared planning visibility, joint KPIs, and contractors attending the same toolbox talks as internal crews. The result was a contractor workforce that actively contributed to improvement rather than just executing scope. Site C's approach was almost the opposite: contractors received work packages and were expected to deliver independently, with minimal planning integration and performance reviewed only at contract renewal. Same contractors, very different outcomes.
Reliability engineering needs a seat at the table, not just a desk
The difference between Site D's reliability engineer and the equivalent role at Site C was not skill or experience. It was integration. At Site D, the reliability engineer was involved in shutdown planning, PM strategy, and capital recommendations. At Site C, the role existed on the organisational chart but operated largely in isolation, producing analysis that did not consistently feed into maintenance decisions. The work was being done. It just was not being used.
What the Data Actually Tells You
A maturity assessment across multiple crews and sites produces two distinct stories, and both are valuable.
The first story is about quick wins. When your best crew is already demonstrating what good looks like on the same site, with the same systems and the same conditions, bringing your weaker crews up to that standard does not require capital investment or new technology. It requires structured knowledge transfer, consistent standards, and someone holding the line on process discipline. The proof it works is already on site.
The second story is about where the ceiling sits. Once you know where your best crew stands against an external framework, you know how much headroom exists above your current best practice. That is the conversation that builds the case for broader investment in systems, training, or technology.
For this group, the immediate priority was Site C. Not because the gaps were the largest, though they were, but because the downstream effects of poor work order quality were cascading into every other dimension. Fixing the data discipline problem first unlocked improvement across spares management, PM effectiveness, and shutdown planning simultaneously. One root cause, multiple gains.
Key Takeaways
The question is not whether performance varies across your sites and crews. It does, in every organisation we have assessed. The question is whether you have a clear, objective picture of where it varies and why.
Internal benchmarking using a consistent framework removes the different context excuse that derails external comparisons. Nobody can argue the comparison is unfair when it is the same company, same commodity, same systems, and same operating environment.
External benchmarking then tells you where your entire portfolio sits relative to what good looks like, and gives you the strategic ammunition to make the case for improvement at the leadership level.
The most powerful position is having both pictures at once.
If you are curious about where your maintenance crews actually sit, we run structured maturity assessments across all eight dimensions of maintenance management, calibrated against the GFMAM Maintenance Management Landscape and tailored to your operational context. The process takes two to three weeks per site, and you walk away with a clear heatmap, a gap analysis, and a prioritised improvement roadmap.
Book a scoping call to talk through what that looks like for your portfolio.

Anomaly Detection vs Classification: Which ML Approach When?

.jpg)