Use the same measure at every location: PM services completed on or before their due point at that depot, divided by PM services due there in the period, times 100. Hold every depot to at least 95%, review the numbers monthly, and make each depot’s Shop Supervisor the accountable owner. The fleet-wide average is not enough — a fleet sitting at 95% overall can be hiding one yard at 82%, and that yard is where your next breakdown, downtime spike, and budget overrun is coming from.
Most fleets that run multiple depots report one PM compliance number. It looks clean on a leadership slide, and it hides exactly the thing you need to see. Averages bury the outlier. If four locations run at 97% and one runs at 82%, the fleet number still reads in the low 90s — close enough to target that nobody asks questions. Meanwhile the lagging depot is quietly building a backlog of overdue services, and overdue PMs turn into unplanned repairs, comebacks, and lost availability.
By the time a single depot drags the fleet-wide number below target, the problem is months old. Tracking compliance per location is how you catch it in week one instead.
Don’t invent a new metric for multi-site tracking. Take the PM compliance measure you already run and calculate it separately for each location:
| Element | Per-Depot PM Compliance |
|---|---|
| Definition | The share of scheduled PM services completed on or before their due point at that location |
| Formula | PMs completed on or before due at that location ÷ PMs due there in the period × 100 |
| Target | At least 95% — at every depot, not just on average |
| Review cadence | Monthly, every depot in the same review |
| Accountable owner | That depot’s Shop Supervisor |
This is the same definition, target, and owner we recommend for fleet-wide PM compliance — the only change is that you calculate and review it one location at a time. One real example: a Colorado school district fleet that started managing PM compliance this way pushed compliance and vehicle availability up together, because the two metrics move as a pair.
Per-depot numbers are only comparable if every depot plays by the same rules. Two things have to be locked down fleet-wide:
Mixed fleets are the usual objection — one depot runs refuse trucks, another runs sedans. That’s a reason to set due-point rules by vehicle class, not a reason to excuse a depot from the target. Heavier-duty classes get their own PM schedules; the 95% expectation stays the same.
The number tells you where to look, not who to blame. For any location under 95%, pull its overdue PMs and look at how late they are. A depot full of services one or two days past due has a scheduling problem. A depot with PMs weeks overdue has a capacity problem — usually technician coverage, parts delays, or bay availability.
Then follow up with that depot’s supervisor on the specific pattern. The monthly review should end with the lagging depot having a short list of named fixes, and next month’s number tells you whether they worked. Fleets using Fleet360’s maintenance and repair tools get this rollup automatically — compliance by location, overdue lists, and how late each service is — without anyone building spreadsheets.
See Fleet360 in Action
Per-depot PM compliance rollups built from your real work orders — no spreadsheets.
Keep the target at 95% everywhere. What changes by depot is the PM schedule itself — heavier-duty vehicle classes get more frequent due points. Lowering the target for a struggling depot just hides the problem the metric exists to surface.
That depot’s Shop Supervisor. The Fleet Manager reviews all depots together monthly, but accountability for hitting 95% sits with the person who controls that location’s schedule and bays.
Monthly, with every depot in the same review so locations are compared on the same period. Supervisors at depots below target should be watching their overdue list weekly in between.
For leadership reporting, a weighted fleet-wide number is fine. For managing the fleet, the per-depot numbers are what matter — a small depot at 80% is a real problem even if it barely moves the weighted average.