Overtime and What It Really Costs
Overtime is usually cheaper than it looks and is compared against the wrong number. The arithmetic, the on-costs that do not scale, and where the real limit sits.
One extra hour of output, four ways to get it
Premium on the wage; most on-costs do not scale.
Includes the agency's own loading and margin.
Correct only once the hire exists and is at full speed.
Recruitment, induction and the ramp, spread over year one.
Overtime is the second cheapest of the four and is treated as the most expensive. The cheapest looks cheapest only because the cost of getting there has been left out of the comparison. One employer's own figures, not a benchmark.
An overtime hour usually costs less, relative to the fully loaded rate, than the premium suggests — because the premium is applied to the wage while most of the on-costs do not scale with an extra hour. Compared properly, time-and-a-half is often around a ten to fifteen per cent premium on the loaded rate rather than fifty.
The pay question in “Overtime and What It Really Costs” needs a visible chain from working time to approved payroll input. Used for gdpr employee monitoring, a practical gdpr employee monitoring workflow can organise the operational time record, while statutory calculations, deductions and final payroll decisions remain in the appropriate payroll system.
Which does not mean overtime is good. It means the comparison that rejects it is usually made against the wrong number.
Teams documenting “Overtime and What It Really Costs” may use SHRM compensation resources as an independent reference. A useful source should prompt a specific check rather than replace case-by-case judgement.
The arithmetic
The premium applies to the wage. Employer contributions scale with it. Pension usually scales. Paid absence, equipment, licences and insurance do not: the person already exists and already has a laptop.
So the increment is wage plus a bit, against a base that includes several layers that stay flat.
A worked comparison
Loaded rate £26.11, of which £18.40 is wage. An overtime hour at time-and-a-half costs £27.60 in wage plus the contributions that scale, roughly £29.56 all in.
Against £26.11, that is about thirteen per cent more — not fifty.
What this does to decisions
Overtime looks dear and gets rationed; recruitment looks cheap and gets chosen. Both are the wrong way round at the margin, and the error is large enough to change the answer.
The honest comparison is in the sign on this page: four ways to get one extra hour, all loaded the same way.
Where the real limit is
Not cost. Fatigue, error rates, statutory limits on working time, and the point at which people stop volunteering.
These are real constraints and they are the reason to ration overtime. Pretending the reason is cost loses the argument the first time somebody does the arithmetic.
The limit that is cost
Sustained overtime becomes a structural cost: it inflates the base for any pay-related calculation, it raises holiday pay where average earnings are used, and it builds an expectation that is expensive to withdraw.
A fortnight of overtime and a year of overtime are different financial objects.
Holiday pay on overtime
In many jurisdictions regular overtime enters the calculation of holiday pay, which makes sustained overtime more expensive than the hourly comparison shows.
What any particular rule requires differs by place and by agreement and changes. The costing point is that the effect exists and belongs in the model where it applies.
Unpaid overtime
The salaried version, and the largest hidden labour cost in most professional operations. It is consumed as though free and paid for in attrition.
It should appear somewhere as a counted quantity even where it has no cash cost, because otherwise the operation's true hours are unknown.
Time off in lieu
Not free. It converts a cash cost into an hours cost in a later period, and the later period is usually one where the hours were needed.
Modelling it as costless is how a busy quarter quietly creates a short one.
The two figures to publish
The marginal cost of an overtime hour next to the loaded rate, and the four-way comparison once a year. Two figures and one small table.
They settle the recurring argument about whether to approve overtime, which otherwise runs on impressions.
Overtime that is really understaffing
Where overtime runs at a steady level for a year, it is not overtime; it is a headcount decision taken by default, at a premium, with the cost falling on people's weekends.
The arithmetic above says the premium is modest. It does not say that a permanent thirteen per cent premium on a tenth of the hours is a good way to run an operation, and the comparison that matters there is against recruiting, which the ramp note here covers.
What the four-way comparison needs
All four options loaded the same way, with the lead time stated next to each. Overtime starts this week; a hire starts in eleven weeks.
Cost per hour without lead time answers only half the question, and it is the half that rarely decides anything. Most of these decisions are made under time pressure, which means the option that can start is frequently the only real option on the list.
The decision that should follow
Where the premium is genuinely small against the loaded rate and the work is short-term, overtime is usually the right answer, and saying so with the arithmetic attached is more persuasive than any policy.
Where the work is permanent, the comparison is not overtime against recruitment but recruitment against doing less, and that is a different conversation with different people in it.