Fractional against full-time, with the real numbers on both sides
A salary is not what an employee costs. Employer national insurance, the pension minimum and the fee to find them sit on top of it, and none of them appear in the figure anybody quotes. Below is the full-time side worked out properly, against the same seat at two days a week.
Every assumption here is set against this site's own interest, and the reason is in the page. The method is published, including where a figure is a judgement rather than a statute.
What sits on top of a salary
15 per cent of everything above the £5,000 secondary threshold. On a £140,000 salary that is £20,250, and it is the largest single item most comparisons leave out.
3 per cent of qualifying earnings, the band from £6,240 to £50,270. It caps at £1,321 and does not rise with salary. Modelled at the statutory minimum deliberately, because that is what is compulsory.
20 per cent of first-year salary, so £28,000. That sits inside the published contingency band of 15 to 25 per cent and below the retained band of 25 to 33, which means the full-time side of this comparison is never made to look worse than it is.
Equipment, desk, private medical, employer liability cover, the cost of the notice period if it does not work out, and the management time of running a search. All real, all left out, all pushing the same way.
Year one, side by side
Full-time salaries are mid-market assumptions rather than London top end. Fractional is the published band at two days a week, on 46 billable weeks a year.
Where full-time wins, said plainly
Three things, and any one of them settles it.
- The function has a team that needs managing every day. A part-time leader becomes the bottleneck, and no arithmetic rescues that.
- The work is volume rather than decisions. Running a process is a five-day job whatever the seniority of the person doing it.
- The days keep creeping past three. At that point the comparison has already flipped and usually nobody has rerun it.
Run the specific case in the calculator, which computes the break-even day count rather than asserting one, and says so when full-time is the cheaper answer. Where it is, full-time search is a service sold here too, priced on the page.
Common questions
What does a full-time CFO actually cost beyond the salary?
On a £140,000 base, employer national insurance adds £20,250, the statutory pension minimum adds £1,321, and a search fee at 20 per cent adds £28,000. Year one lands at £189,571, which is £15,798 a month.
Is a fractional executive cheaper than a full-time hire?
At two days a week, usually by a wide margin. A fractional CFO at two days runs £6,133 to £11,500 a month against £15,798 for the full-time equivalent in year one. At four days or more the comparison reverses, and it reverses before most companies expect it to.
How much is employer national insurance in 2026?
15 per cent on earnings above the £5,000 secondary threshold. On an executive salary it is the largest single addition to the headline number and the one most often left out of a comparison.
Why is the pension figure so small in this comparison?
Because it models the statutory minimum rather than what a generous employer pays. Auto-enrolment requires 3 per cent of qualifying earnings, the band between £6,240 and £50,270, so it caps at £1,321 and does not rise with salary. Most companies pay more than that for an executive, but more is a choice and a comparison should model what is compulsory.
At how many days a week does full-time become cheaper?
It depends on the rate and the salary, which is why the calculator computes it rather than this page asserting a number. As a rule of thumb it lands between three and four days for most seats, and the moment a function needs daily management the answer is full-time regardless of where the arithmetic falls.
Whichever the arithmetic says
Browse the bench for the fractional route, with the rate on the profile. Retained search for the permanent one, with the fee on the page. Same place, so nobody here has an interest in selling you the wrong shape.