The Sixty-Thousand-Pound Chair: What a UK Hire Really Costs Before It Earns
Photo: Francisclarke, CC BY-SA 4.0, via Wikimedia Commons
A vacancy appears, the team is stretched, and the answer feels obvious: recruit. Yet few UK SMEs apply the same scrutiny to a hire that they would to a three-year lease or a piece of plant machinery. The salary is visible; almost everything else the decision costs is not. Assembled in full, a mid-range appointment can easily represent a £55,000 to £60,000 commitment in the first year — and considerably more if it does not work out.
The gap between the salary and the cost
Begin with the statutory layer. Employer National Insurance contributions are charged at 15% on earnings above the secondary threshold of £5,000 a year, so a £35,000 salary carries roughly £4,500 of employer NIC. The Employment Allowance can reduce that bill, but it is not available to employers whose NIC liability exceeds £100,000 in the prior tax year, which rules out a good number of established SMEs.
Pension follows. Auto-enrolment sets a floor of 3% of qualifying earnings, but most employers seeking to attract capable people in technical, commercial or professional roles offer 5% or 6% as a matter of course. On a £35,000 salary the statutory minimum lands just under £900; a 5% scheme is closer to £1,450.
Then come the costs that sit outside payroll but belong squarely to the hire: recruitment, equipment and software licences, training, an allocation of payroll and HR administration, employer's liability insurance, and any benefits extended to the wider team.
A worked example, honestly costed
| Item | Year-one cost |
|---|---|
| Salary | £35,000 |
| Employer NICs (15% above £5,000) | £4,500 |
| Pension at 5% of qualifying earnings | £1,440 |
| Recruitment at 15% of salary | £5,250 |
| Equipment, licences, initial training | £1,500 |
| Payroll, HR and insurance allocation | £600 |
| Subtotal before productivity | £48,290 |
The subtotal is the number most business cases stop at. It is not the cost.
The two costs nobody puts in the spreadsheet
The first is the ramp. A new starter in most roles is fully productive somewhere between three and six months in, and heavily supervised for the first eight to ten weeks. Model ten weeks at half output and you have absorbed roughly five weeks of salary — around £2,900 on this example — plus the manager's time spent on induction, review and correction.
The second is management overhead, which persists well beyond onboarding. Two hours a week of supervision, checking and coordination at a fully loaded internal rate of £45 an hour comes to £4,680 across the year. Some of that is genuine investment; some is simply the administrative drag of employing another person.
Add those two items and the year-one figure sits near £56,000, before any pay review, any benefit uplift or any cost of a failed appointment — an event that typically consumes three to six months of salary in recruitment, notice and lost output.
The number that actually governs the decision
Headcount should be priced per productive hour, not per annum. A full-time employee working 37.5 hours a week nominally provides around 1,950 hours a year. Remove statutory and contractual holiday, an allowance for sickness, training days and the ramp period, and 1,400 to 1,500 genuinely productive hours is a realistic first-year figure.
At £56,000, that is close to £38 an hour — and the role must generate contribution above that rate, loaded with whatever margin your overheads require, simply to break even. Setting that threshold down before the advert goes out changes the conversation. It converts a sentiment ("we need help") into a test ("this role must generate £X of contribution within twelve months").
The alternative that rarely gets a fair hearing
Capacity can be bought in three ways: hire, outsource, or release. The third option is chronically under-explored because it produces no new colleague and no visible sign of growth.
Ask what the prospective hire would actually do in their first six months. In many SMEs the honest answer is a list of tasks — rekeying purchase invoices, chasing approvals by email, reconciling two systems, rebuilding the same report each month, rescheduling jobs by hand. Those tasks have a market rate. Often it is a fraction of £56,000, and it carries none of the employment risk: no notice period, no redundancy exposure, no tribunal exposure, and no commitment stretching into the following financial year.
The comparison is not always favourable to the process fix. Client-facing work, judgement-heavy roles, regulated positions and anything requiring relationship continuity still belong with people. Nor is automation free: someone must own the system, and poorly specified tooling simply creates a second set of problems. But the comparison should at least be run, in writing, before headcount becomes the default.
Four tests worth applying before the contract is signed
Durability of demand. If the workload driving the hire would need to persist for at least eighteen months to justify the commitment, is there evidence it will? A single large contract is not a trend.
Subtraction before addition. Which of the role's intended duties could be removed rather than resourced? A fortnight spent documenting the workflow frequently eliminates a quarter of the perceived requirement.
Contribution threshold. What revenue or cost saving must the role deliver, and by when, for the hire to be value-accretive rather than merely comforting?
Timing. Recruitment typically takes eight to twelve weeks, with three months of ramp behind it. Capacity requested in November tends to arrive in the second quarter of the following year — by which point the original pressure may have changed shape entirely.
None of this argues against hiring. It argues against hiring as an unexamined reflex. An owner who can state the fully loaded cost, the productivity ramp, the contribution threshold and the process alternative has made a commercial decision. One who has only approved a salary has made a guess.
