Cost per operating hour is the most useful number in fleet management and the easiest one to calculate badly. Divide last month's repair invoices by the hours on the meter and you get a figure that looks precise and tells you almost nothing, because it excludes the costs that do not arrive as invoices and includes hours the machine spent idling. Done properly it becomes the number that settles arguments — which machine to send where, whether to hire or own, when to replace, and what to charge a project. Done carelessly it produces confident decisions built on a bad denominator. You can review how your own figure is currently derived in a free 30-minute session.
Know What Each Machine Actually Costs Per Hour
A per-asset cost figure built from real maintenance, parts, labour, and downtime records rather than from a fleet-wide average divided by meter readings.
What Has to Be in the Number
Leave any of these out and the figure is not wrong so much as incomparable, which is worse.
Parts, attributed to the machine
Issued against a work order, not against a site, or the cost never finds the asset that consumed it.
Labour, booked to the job
In-house technician time counts even though no invoice is raised for it.
External repair invoices
Including callout premiums, which are part of what that machine cost you rather than a separate category.
Consumables and wear items
Tyres, undercarriage, and fluids, which distort comparisons badly when omitted from some machines and not others.
Productive hours, not engine hours
The denominator matters as much as the numerator. Idle hours inflate the meter and flatter the figure.
Four Decisions the Number Actually Settles
Two nominally identical excavators rarely cost the same to run. Sending the cheaper one to the long project and the more expensive one to short work is a decision you can only make if you know which is which.
A hire rate is a known number. Comparing it against a guess is not a comparison. Cost per hour turns this from a preference into a calculation.
Rising cost per hour on one machine, tracked over time, is the clearest replacement signal available, and it appears well before the machine becomes obviously unreliable.
Internal cross-charging built on real figures stops projects arguing about whether they were fairly loaded, because the basis is visible.
See the Figure for Your Own Machines
Cost attributed per asset as work orders close, so the number builds continuously rather than being assembled when someone asks.
When the Number Misleads
Low utilisation inflates it
A machine that barely worked still incurred fixed costs, so its cost per hour looks terrible. That is a utilisation finding, not a machine finding.
New machines look expensive early
First-year costs include commissioning and early adjustments spread over few hours. Judging a new asset on its first quarter is unfair to it.
One major repair skews a short period
A single engine rebuild dominates a monthly figure. The metric is a trend line, not a monthly verdict.
Different duty cycles are not comparable
A machine on rock costs more per hour than the same model on soft ground, and that difference is the site rather than the asset.
Frequently Asked Questions
Should ownership cost be included alongside maintenance?
It depends on the decision. For deployment and maintenance questions, running cost alone is usually the right comparison. For hire-versus-own or replacement decisions, EMI or depreciation needs to be in there. You can work through which version suits each question on a call with our team.
How long before the figure is trustworthy?
Usually a few months, because short periods are dominated by whatever happened to break. The trend becomes reliable before any single month's number does.
What if our hour meters are inaccurate?
Then fix that first, since the denominator affects every figure downstream. Meter drift and inconsistent reading cadence are common and both distort cost per hour significantly.
Can we compare across different equipment types?
Not usefully. Cost per hour compares like with like — the same model doing similar work. Across categories it produces a number without a meaning.
What is the most common mistake?
Dividing by engine hours rather than productive hours, which makes heavily idling machines look more efficient than they are. Start with a free trial to see both figures side by side.
Build the Number Once, Then Use It Everywhere
Attribute parts, labour, and external invoices to the machine that incurred them, divide by productive hours rather than meter hours, and read it as a trend rather than a monthly verdict.
