One dashboard. The same KPI definitions, rolling up site by site. The same leadership team reviewing planned work, backlog, PM compliance, and cost for every facility on one screen.

From 30,000 feet, multi-site maintenance looks almost simple. Sort the strong sites from the laggards, then push everyone toward the same standard. Job done.

Then you visit the plants.

One is an aging brownfield facility, congested, decades of modifications layered on top of each other, patchy documentation, and a shrinking bench of experienced craft. Another is newer, cleaner, more modular, sitting in a completely different contractor market. A third has decent systems on paper and chronic turnover in planning and supervision. A fourth is carrying years of thin history and inconsistent coding in the EAM.

So now the real question. Are you comparing performance? Or are you measuring four different realities with the same ruler?

That's the actual challenge of multi-site maintenance.

Most corporate maintenance strategy treats multi-site execution as a replication problem. Pick the right process. Train everyone on it. Install the same system. Watch the same KPIs. Hold people accountable. All of that matters. None of it is enough.

Copying a model from one location to the next ignores the thing that actually makes multi-site hard. Each plant lives inside its own mix of inherited conditions and local constraints. You're not managing one system repeated five times. You're managing five different worlds at once, and they happen to share a logo.

What shapes those worlds? Asset age and physical condition. Production process and operating rhythm. Layout, congestion, and access. The local labor market and contractor capability. Leadership stability and maintenance culture. Digital maturity and the quality of years of historical data. Those variables don't just move the performance numbers. They change what "good" 

even means at a given site, and how fast that site can realistically improve.

The easy mistake is to treat every site like it started from the same place. It shows up in a few familiar ways.

Uniform KPI pressure, where every plant has to hit the same metric on the same timeline, even though one is fighting old assets and bad documentation and another isn't. Copied staffing models, where an org chart that works in one environment gets lifted and dropped into a site with different scale, different skills, and different execution demands. Same process, different data reality, where corporate rolls out one set of planning and scheduling expectations everywhere, but some sites have usable job plans and trustworthy history while others are starting from scratch. And the worst one, misreading poor fit as poor discipline, where leaders see a site struggling with a system that doesn't suit it and conclude the people are resisting change or lack accountability.

Sometimes that last read is fair. Often it isn't. And when the diagnosis is wrong, the result is predictable: frustrated sites, deep skepticism about anything labeled a corporate program, and a lot of quiet KPI gaming to make things look more aligned than they are.

The phrase I keep coming back to is one company, many maintenance worlds.

A corporate standard can fairly say every site should improve backlog health, planned-work percentage, and PM effectiveness. Reasonable goals, all of them. But the road to get there won't be the same anywhere. For one plant, the first priority is job-plan quality. For another, supervisor capability. For another, contractor coordination during outages. For another, something more basic: cleaning up years of garbage master data so the site can finally trust its own work history.

That's the point where multi-site complexity quietly becomes leadership complexity. The job isn't to force identical action across the fleet. It's to hold enough standardization that the company can align and learn from itself, while leaving enough room that each site improves from where it actually stands. That's a far harder job than publishing a playbook and waiting.

The healthiest multi-site organizations get clear about what should be common and what has to be local.

They standardize the things that let an enterprise behave like one enterprise: the core maintenance and turnaround language, the KPI definitions and governance cadence, the planning and scheduling principles, role clarity for planners, schedulers, supervisors, and operations, and a floor on critical data and work-order history. Then they tailor the things that genuinely differ site to site: the pace of rollout, the level of process rigor, where the support resources go, the order they tackle improvements in, and how deeply they embed hands-on execution help at a given 

location.

That mix is what lets a company act as one without pretending every plant is equally ready, equally constrained, or equally complex. Standardize the principles. Tailor the execution. Skip either half and you get a fleet that's either fragmented or faking alignment.

A few moves make this far more manageable.

Profile each site's complexity and maturity before you push the same initiative everywhere, so you've got a baseline grounded in reality rather than in the dashboard. Segment sites instead of averaging them, comparing an aging high-complexity plant to its real peers, not to a newer and simpler facility that never faced the same conditions. Deploy support where the complexity is highest, because some sites need hands-on execution help, not another binder of standards, especially where planning quality, outage execution, or data discipline are weak. And build local champions, because change sticks faster when site leaders own the approach instead of feeling it land on them from corporate.

This is the seam where execution support earns its place. A strategy gets designed centrally. Execution always happens locally, on a specific plant floor, with a specific crew, under specific constraints. Bridging that gap, turning a good central plan into something that actually fits the plant in front of you, is one of the hardest parts of multi-site improvement, and it's a lot of what we do.

Multi-site maintenance leadership isn't about making every plant look identical on paper. It's about helping very different plants improve under very different conditions while still moving toward a shared standard.

That starts with admitting a hard truth. You're not running one maintenance system copied five times. You're running multiple realities that share a logo. The organizations that handle it well standardize the principles, tailor the execution, and put their help where the complexity is worst. The ones that handle it badly keep treating the differences between sites as excuses instead of the real design constraints they are.

Next in the series, we go back to the debt side and the single event where all of this comes to a head. The turnaround. Done right, it's your best chance to pay down years of maintenance debt. Done wrong, it's where you pile on more than you retire.