Your Best Maintenance Crew Is Already Outperforming Your Worst. Here's How to Measure the Gap.

Every external benchmark gets dismissed eventually.

Different ore body. Different contractor model. Different shift structure. Different everything. The asterisk appears before the conversation is finished, and whatever the benchmark revealed gets filed away with a note about context.

Internal benchmarking doesn't have an asterisk.

When you compare your own crews against each other, on the same site, with the same budget and the same supply chain, the context is identical. The gap is undeniable. And in almost every maintenance assessment we run, that gap is larger than anyone expects.

What the data usually shows

Planning compliance ranging from 47% on the worst-performing crew to 83% on the best. Both crews. Same site. Same planning system.

Repeat failure rates on the same equipment differing by a factor of three between shifts. Not because the equipment is different. Because the disciplines around it are.

PM task completion that looks identical across crews on paper, with failure history that tells a completely different story underneath.

The gap between a site's best and worst maintenance crews frequently exceeds the gap between the best crew and genuine industry best practice. That's a confronting finding. It's also a useful one, because it means the improvement you're looking for is already happening somewhere inside your own fence line.

Why internal benchmarking works where external doesn't

External benchmarking serves a purpose. It gives you a strategic reference point, helps you make the case to leadership that investment is warranted, and connects your performance to an industry framework like the GFMAM Maintenance Management Landscape or ISO 55001 maturity criteria.

But it's a second conversation, not a first one.

The first conversation is internal because it's the one that can't be argued with. When the gap is between your own crews, on your own site, the response isn't "our context is different." The response is: "What is Team A doing that Team B isn't?"

That's a question that leads somewhere. It leads to coaching, to process standardisation, to planning rhythm changes, to the specific behaviours that produce reliability. And it leads there without capital spend, without new technology, and without waiting for an external assessment to land.

The five dimensions that matter

Across maintenance assessments in mining and heavy industry, five performance dimensions consistently separate high-performing crews from average ones on the same site.

Planning and scheduling maturity. How far ahead is work planned? Are jobs kitted and resourced before the shift starts? Planning compliance above 80% is achievable. Teams below 50% are spending significant time reacting to work that was predictable.

PM effectiveness. Not PM completion, PM effectiveness. Are the tasks actually preventing the failures they were designed to prevent? Are intervals based on failure data or inherited from commissioning? The distinction matters more than the completion rate.

Work order quality. What does a closed work order actually contain? "Fixed pump" is a data entry, not a record. Teams that capture equipment ID, failure description, what was found, what was done, parts used, and time to repair are building an asset history that supports better decisions. Teams that don't are running blind.

CMMS data integrity. Can you trust what the system tells you? An asset register with gaps, incomplete failure codes, and unmaintained equipment hierarchies produces analysis you can't rely on. Teams that maintain data integrity get better information back out of the system. It compounds over time.

Execution discipline. When work is planned, does it get done the way it was planned? Scope reductions that go undocumented, jobs that are completed to close the work order rather than to resolve the failure, quality shortcuts under time pressure. The gap between planned and actual execution is often where reliability is lost.

Using the gap

The point of internal benchmarking isn't to embarrass the worst-performing crew. It's to identify what the best-performing crew is doing differently, and transfer that capability across the site.

Your benchmark crew is your proof of concept. They're demonstrating, right now, that the performance level you're targeting is achievable with your resources and your operating environment. The question isn't whether improvement is possible. It's which specific behaviours need to change, and what support the other crews need to change them.

This is also the structure that makes the second conversation easier. Once you've closed half the internal gap and have results to point to, the case for broader investment in technology, systems, or an external maturity assessment becomes credible. You're not asking for a leap of faith. You're asking to build on demonstrated progress.

Score your crews

We've built a free Internal Benchmarking Scorecard that covers all five dimensions with 1-5 scoring rubrics written so a maintenance superintendent can self-assess without external support. It scores up to four crews simultaneously, produces a gap analysis, and includes a capability transfer plan template to structure the improvement work that follows.

Running it takes a few hours. It requires access to CMMS data and an honest conversation with the supervisors who know the crews best. The output is a clear picture of where the performance variation sits, what it's worth to close it, and where to start.

[Download the Internal Benchmarking Scorecard]

The bottom line

The improvement potential sitting inside your own fence line is almost always larger than the improvement potential sitting in an external benchmark.

Find the gap. Use your best crew as proof it's closable. Build from there.

The external benchmark will still be there when you need it for the second conversation.

Shane Scriven is the Managing Director of SAS Asset Management. SAS-AM helps asset-intensive organisations measure what matters, model what's coming, and make the right call.