More Than One Depot
Fleet data becomes comparative, which is useful and is the point at which comparisons start being made badly.
Running it · Analysis
A single-site fleet has a manager who knows every vehicle. Several sites produce a table, and tables invite conclusions.
What genuinely differs between sites
The work: urban and rural rounds produce different consumption, different wear and different utilisation.
The vehicle mix, frequently accumulated rather than chosen.
Local suppliers, whose labour rates and parts pricing differ substantially.
Terrain and climate, which affect consumption and battery performance more than most comparisons allow.
Which means a raw cost-per-mile league table compares geography, not management.
Comparing fairly
Like vehicles on like work, which usually means comparing within a vehicle type and round type across sites rather than site totals.
Normalise for the obvious: urban against trunk, laden against light.
And state what is not normalised, because something never is.
A comparison with its caveats stated is used; one presented as definitive is disputed and then ignored.
What is worth comparing
Unscheduled maintenance proportion, which is management rather than geography.
Time from fault code to workshop, which is process.
Daily check completion, which is discipline.
Utilisation, which is planning.
These four are within a site's control and are the ones a comparison can honestly inform.
Redistribution
The most common finding across depots is imbalance: one site short and another with idle vehicles.
Which is a transfer rather than a purchase, and the transfer is usually cheaper than either site's preferred solution.
Check the work first, because a vehicle idle at one site may be specialist and useless at the other.
Standardising what to standardise
Device configuration and alert thresholds, so the data means the same thing everywhere.
Vehicle and device registers, with one owner.
Maintenance intervals for the same model, which drift when each site books its own.
Not: supplier choice, which is frequently better local.
The measure
The four comparable figures, by site, quarterly.
With the vehicle mix and work type stated alongside.
And the redistribution opportunity, counted, which is the finding that pays for the reporting.
Compare the four controllable figures
And nothing else.
Unscheduled maintenance proportion. Time from code to workshop. Daily check completion. Utilisation.
These are management.
Cost per mile is geography and vehicle mix, and a league table on it produces a depot performance conversation about terrain.
Follow the difficult record
Use open the record-flow reference to frame one representative case. The useful evidence is the record created when a value is corrected, approved and exported.