You ask three plant managers for their PM compliance rate and you get three numbers that mean three different things. One counts a PM as complete when the work order closes. The next counts it when the technician signs off, which might be a week later. The third doesn’t track it at all and just gives you a guess. You can’t roll those up or compare them, and you can’t tell which site is actually doing better, because everyone is measuring a different thing.
That’s the real cost of letting every site run maintenance its own way. The chaos itself is survivable. What gets you is going blind the moment you try to manage more than one location as a system.
Why every site drifts apart on its own
Nobody sits down and decides to do it differently. It happens because each site got its CMMS, its spreadsheet, or its whiteboard at a different time, set up by a different person, to solve whatever problem was on fire that week.
Site A names assets by the maintenance lead’s own logic: “Big Compressor,” “Compressor #2,” “the one by the dock.” Site B uses the manufacturer’s serial numbers. Site C inherited a numbering scheme from a system it stopped using back in 2019. Each one works fine inside its own building. None of them survive contact with a second building.
Priority levels drift the same way. At one site, “urgent” means production is down. At another, it means a manager asked for it personally. So when a report says one site has four urgent work orders open and another has nineteen, you can’t tell whether the second site has a bigger problem or just a looser definition of the word.
The pattern repeats across PM schedules, work order categories, and what “closed” even means. Each is a small local decision, reasonable on its own. Stack them together and the portfolio becomes impossible to read.
What to standardize, and what to leave alone
Most directors overreach here and try to standardize everything, which is exactly what triggers the revolt. You’re not after uniformity for its own sake. The things you need to compare across sites have to mean the same thing everywhere, and everything else can stay local.
Standardize these four:
Asset naming and hierarchy. This is the foundation everything else hangs off. Pick one convention covering site code, area, asset type, and a sequence number, then apply it everywhere. Something like STL-HVAC-RTU-04 (St. Louis, HVAC, rooftop unit, number four) tells you where it is and what it is before you open anything. You don’t need every site to love the format. You need it to be the same format, so a part number, a failure history, and a cost report all line up across locations.
Priority levels. Three or four levels, each with a written definition tied to something you can observe: safety risk, production impact, a response-time target. “Urgent = safety hazard or production stoppage, respond within 2 hours” leaves far less room for local interpretation than a bare word does. Once priority means the same thing everywhere, an open-work-order report finally tells you something true.
PM templates for shared equipment. If five sites run the same model of rooftop unit, the preventive maintenance procedure for that unit should be one template, written once, deployed to all five. The local technician still does the work their own way, but the task list, the interval, and the completion criteria stay identical. That’s what makes PM compliance comparable.
Reporting definitions. Agree on what each metric counts before you ask anyone for it. “PM compliance” = scheduled PMs completed on or before the due date, divided by scheduled PMs, measured monthly. Write it down. A rolled-up number is only as good as the weakest local definition feeding it.
Leave these local:
Staffing models, shift schedules, and who does what. The specific vendors a site uses. Local safety procedures shaped by that building’s hazards, the exact wording of internal notes, how a team physically organizes its storeroom. None of these need to match across sites, and forcing them to buys you nothing but resentment. Standardize the data you roll up, and leave the daily craft alone.
Rolling out a standard without a revolt
The fastest way to get a standard rejected is to mail it out as a finished mandate from headquarters. The people who have to live with it had no say, so they’ll find every reason it doesn’t fit their site. And at first they’ll be right, because you designed it without them.
A rollout that survives looks different.
Start with the site that’s already best. Find the location with the cleanest practices and build the first draft around what they already do. Now the standard isn’t theoretical. It’s a real working system from a real peer, which is much easier for other sites to accept than something invented in a conference room.
Pull two or three sites into the draft. Get a working maintenance lead from each major site type into the room before anything is final. They’ll catch the things that don’t generalize, like the asset category one site needs that another has never heard of, or the priority level that makes sense in a 24/7 plant but not in an office building. Catching those early is the whole point.
Roll out one element at a time. Asset naming first, since everything else hangs off it. Get every site renamed and verified before you touch PM templates. Try to change naming, priorities, PMs, and reporting all at once and you’ll guarantee that all four land badly. Done one at a time, each gets done properly.
Convert the back catalog instead of just starting fresh. A standard that applies only to new work orders leaves you with years of history in the old format and nothing to compare against. Map the old asset names to the new scheme and migrate the records. It’s tedious work, but it’s the difference between a standard that produces trend data and one that quietly resets your history to zero.
For the operational side of rolling a standard across locations, our multi-site maintenance overview covers how the pieces fit together.
But what about the site that genuinely is different?
Sometimes a site pushes back and it isn’t resistance, it’s a real difference. A food-processing plant under FDA oversight has documentation requirements an office building never will. A facility on the coast fights corrosion a desert site doesn’t. A 24/7 operation measures “urgent” against a clock the day-shift-only site simply doesn’t have.
Don’t flatten those differences just to force a clean spreadsheet. The fix is layering, not erasing. Keep the shared core identical, the asset naming, the base priority scale, the reporting definitions, so everything still rolls up. Then let the genuinely different site add what it needs on top: an extra PM step for compliance, a tighter inspection interval, an additional work order category. The core stays comparable while the local reality gets respected. You only lose comparability if you let a site change that shared core, and that’s the one place you hold the line.
Rolling metrics up across sites
Once the data means the same thing everywhere, the portfolio view becomes worth having. You finally see real things instead of artifacts of inconsistent definitions.
You can rank sites on the same PM compliance metric and trust that the ranking reflects performance rather than bookkeeping. You can notice that one location’s emergency repair rate is double the others’ and go ask why, instead of wondering whether they just log it differently. You can spot the same pump model failing twice as often at one site and chase the root cause, because the asset names finally line up across locations.
This is where consistent data pays back the effort of getting it. Cross-site reporting and analytics only tells the truth when the inputs are defined the same way, which is the whole reason the standardization work comes first. TeamWork keeps every site in one tenant with shared asset conventions and PM templates, so a portfolio-wide compliance or cost report reads as one coherent picture rather than four spreadsheets you have to reconcile by hand.
Standardizing across sites is mostly organizational work rather than software work, but it’s far easier on a platform built to hold every location in one place. If you’re weighing how to bring your sites onto a common system, the 30-day free trial at teamworkcmms.com gives you room to model your naming convention and PM templates against real assets before you commit to a rollout.