You can afford a hire when the forecast still clears your cash floor in every month after the person starts, including the months before anything they bring in has reached the bank. The test is the full monthly cost of the role, dated to when each part leaves the account, set against the receipt or contract that pays for it, dated to when that money arrives, read on the same forecast with and without the hire. The output is a month, a headroom figure and a condition, not a yes or no.

This page works that test end to end for one decision, the next role, in a finance team of three at a business of 11 to 50 staff. It gives the 2026/27 UK employment costs and the date each one is paid, a method for tying the role to the money that funds it, a six-step scenario, and a worked example run at two start dates. It does not repeat the general method for cash flow scenarios for big decisions, which covers a hire alongside a purchase and a late payer, and it is not the quarterly what-if set, which tests shocks the business does not choose.

What "afford" means when the answer is a bank balance

Start with the word most people reach for. Runway, in the sense of cash divided by monthly burn, is a term built for a business that is losing money by design and counting the months until it must raise again. No UK professional body defines it for a business that trades profitably, and it does not fit one, because a trading business does not burn at a steady rate. It has a low week in every month and a low month in every quarter, and the question is whether a new fixed cost pushes one of those low points through the floor.

Headroom is the better word, and it has a definition. The Association of Corporate Treasurers uses it for the undrawn amount of a borrowing facility, and in practice finance teams count available cash alongside it. This page uses it that way: cash plus any undrawn facility, minus the balance the business has decided never to go below, read in each month of the forecast. A hire is affordable when the headroom stays positive in every month with the hire in, and the decision is taken on the tightest month, not the average.

No professional body prescribes how many months of headroom a business should hold before adding a person. What the guidance does ask for is a forecast that is re-run when a material assumption changes and that shows the effect of changing one assumption at a time. That is what the scenario below does, with the hire as the assumption.

What a hire costs, and when each part leaves the bank

The salary is the smallest surprise. The rest of the cost is set by statute, and each part has its own payment date. The figures below are for the UK 2026/27 tax year; the rates page on gov.uk is the source and is updated each year.

Three points on the table decide the cash timing. The National Insurance on a salary paid on the 25th leaves the bank around four weeks later, because the tax month containing that payday ends on the 5th and the remittance is due on the 22nd after that. If the 22nd is a weekend or bank holiday, the money must reach HMRC on the last working day before it, unless your payment method clears on the day. Employers whose whole PAYE remittance averages under £1,500 a month can pay quarterly, and a new hire is often what takes a small employer over that line.

The Employment Allowance is worth stating carefully because it is often modelled as a discount on the new person. It is set against the employer's National Insurance bill for the whole year, so a business already running a payroll of a dozen people has normally used it well before the new person's first payday. Model the marginal hire at the full 15% unless you know there is allowance left.

Pension timing can move by up to three months. An employer can postpone automatic enrolment for up to three months from the start date, in which case the first contribution is calculated from the deferral date rather than backdated, unless the employee opts in during the postponement. The example below shows both.

Holiday pay for a salaried employee sits inside the salary rather than on top of it. Bonuses, a car allowance, a better-than-minimum pension and any benefits all sit on top, and each carries its own National Insurance.

The multiple that is not a fact. The figure that circulates, that an employee costs 1.25 to 1.4 times salary, is a US rule of thumb from an MIT column that bundles American payroll taxes and health benefits. No UK statistics body or professional body publishes a per-hire multiple, and the one economy-wide measure that existed was discontinued in 2021. Build the number from the rates above instead. On the statutory items alone, a £42,000 UK salary costs about £48,600 a year, or 1.16 times salary. Add the one-off costs of the first year and it lands near 1.4. By the second year it is back at 1.16. A multiple without dates attached is not a cash number.

Recruitment. The CIPD's 2024 resourcing survey, fielded in April 2024 with 1,016 UK respondents, put the median cost per hire at £2,000 for senior managers and directors and £1,500 for other employees, from the 246 and 207 respondents able to estimate it to within a fifth, and the CIPD itself cautions against reading much into the figures. Agency fees are set by contract and are usually a percentage of first-year salary; government guidance on the conduct regulations treats the level as a commercial matter between agency and hirer, so there is no standard rate to quote. Take the percentage, the invoice date and the rebate period from the agency's terms, and put the fee in the month it will be paid.

Equipment and onboarding. No independent UK figure exists for the one-off cost of equipping and inducting an office-based employee. Every number in circulation comes from a supplier. Price it from your own quotes: the laptop, the licences, the desk if there is one, and the first month of a colleague's time.

Ramp to productivity. The best known UK study, from 2014 across more than 500 firms in five sectors, found that a new hire took anywhere from about 15 weeks to a year to reach full productivity depending on where the person came from, with smaller firms reporting faster ramps. It is the only dated UK source, it is twelve years old, and it does not give a figure for your role. Ramp belongs in the scenario as a labelled assumption about the person's output in the early months, which is what the example does.

Other markets. The method is the same in the US, Australia and New Zealand; the on-costs and their dates are not. In Australia, superannuation guarantee is 12% of ordinary time earnings and, since 1 July 2026, is paid on payday and must reach the fund within seven business days, with payroll tax set by each state and territory. In New Zealand, the compulsory employer KiwiSaver contribution is 3.5% from 1 April 2026, rising to 4% in April 2028, and ACC work levies apply on liable payroll. In the US, the employer pays Social Security and Medicare on wages, federal unemployment tax on the first $7,000, and state unemployment insurance at a rate and wage base set by the state. Check the tax authority's own page for the year before the numbers go in the forecast.

The receipt or contract that pays for the role

Most hiring scenarios stop at the cost. The half that decides the answer is the other side: which cash, on which dates, the role is being hired against.

No professional body sets out a method for matching a hire to a named contract or receipt, and a strict match would be false anyway, since the bank account does not know which pound paid which salary. What the guidance does support is stating the assumption explicitly, dated, and testing it on its own. So write it down in one line: this role exists because of contract X, worth Y a month, invoiced on these dates, paid on these terms, and here is how much of Y depends on the new person being in post. If the line cannot be written, the role is being hired on a belief about revenue, and the scenario should say so and test it harder.

The reason to be strict about it is overtrading. The British Business Bank defines it as taking on work the business cannot fulfil for want of working capital, materials or workforce. A hire made to deliver a contract is the workforce half of that definition, and the cash gap between the person's first payday and the contract's first receipt is the working capital half. Both go in the scenario.

How to run the hiring scenario

Step 1: Copy the live base case. Build the hire as a scenario on the current forecast, never as an edit to it, so the only differences between the two versions are the hire's own lines. Confirm the cash floor before you start; every reading below is measured against it.

Step 2: Enter the role as dated cost lines, not a monthly number. Salary from the start date on the payday. Employer National Insurance on the 22nd after each tax month. The pension contribution on the 22nd of the month after each deduction, from the enrolment date, with or without postponement. The recruitment fee, equipment and onboarding in the months they will be paid. Benefits and their Class 1A charge if the role has them.

Step 3: Enter the funding line on its own dates. The receipt or contract the role is hired against, invoiced when the contract says and paid on the customer's observed terms rather than the stated ones. If the person's output builds over the first months, scale the early invoices down and label the ramp as an assumption.

Step 4: Read the tightest month and the headroom in it. Find the first month the scenario crosses the floor, or the month it comes closest, and record the gap. The gap is the answer to "can we afford this": we can from this date, with this much room.

Step 5: Run the same hire at a second start date. Move the start by one or two months and re-read. The cost of waiting is the delayed contract cash and whatever the role would have brought in; the benefit is the base forecast's own low points being past before the new cost lands. Most "no" answers become "yes from" answers here.

Step 6: Attach a condition and a trigger, then merge or retire. Write down what has to be true for the chosen start date to hold, the figure that would tell you it is not, and who watches it. When the offer is accepted, move the scenario's lines into the base forecast. If the hire does not happen, delete the layer.

A worked example: one role, two start dates

This example is illustrative. The business, the figures and the contract are invented; the statutory arithmetic is real for 2026/27. Figures are net of VAT.

A services business with 28 staff has signed a time-and-materials contract worth up to £8,000 a month at full delivery, invoiced on the last day of each month on 45-day terms. Delivering it needs one more project manager on £42,000, paid on the 25th. The cash floor is £75,000. The finance team's base forecast already shows a low point in February, when the December-quarter VAT payment falls due.

The role's cost, built from the table above and two labelled assumptions:

The Employment Allowance is assumed already used by the existing payroll. Ramp is assumed as the new person billing half of the contract's value in their first month, three-quarters in the second and the full £8,000 from the third, so the invoices are £4,000, £6,000 and £8,000, each paid 45 days after month end.

Now the forecast, with pension postponed to month three, at a 1 December start:

The breach comes from three things landing on the base forecast's own low month: the recruitment fee in January, the third month's salary before any contract cash has arrived, and the VAT quarter. Without postponing the pension the February figure is £179 lower, which shows how little the pension timing matters against the fee and the receipt dates.

The same hire at a 1 February start, with the contract's delivery moved back two months to match:

Read the two together. In December the answer is no, by £8,625 in February. In February the answer is yes, with £3,700 of headroom in the tightest month, and that month is the one before any of the new person's cost has really landed. The condition attached to the February start is that the base forecast's January collections arrive as expected, because a £3,700 shortfall there is the whole margin. The trigger is the January receipts run against expected dates, watched by whoever owns the forecast, with the fallback agreed in advance: a March start, or a mobilisation invoice on the contract paid before the person starts.

Reading the result

Three readings come out of the example and apply to any hire.

The start date is a lever with a price. Moving the start to February bought £12,325 of February headroom, and the price was two months of contract revenue delayed and the risk that the customer would not wait. Run the second date before deciding rather than after, and put the cost of waiting next to the headroom it buys.

The floor month matters more than the average. Across the six months the December start still leaves an average of nearly £16,000 of headroom a month, which sounds affordable. The February month alone is what says no. Read every hire on its worst month.

The funding line is the other half of the answer. In the example the contract pays £8,000 a month from month three, covering the role nearly twice over, which is why the answer flips to yes so quickly once the timing works. Had the role been hired against pipeline rather than a signed contract, the same scenario with the receipts removed breaches the floor in February and clears it by about £400 in March, and keeps that shape until the receipts are real. The scenario does not judge the pipeline; it shows what has to be true.

How Float fits

Float connects to Xero or QuickBooks Online through a one-way connection, imports reconciled bank balances, invoices and bills once a day with a manual sync on demand, and holds a rolling 13-week and monthly forecast that the scenario above is copied from, so the base case is live rather than last month's export. A Sage Intacct connection is in development, with a waitlist open.

A new hire is entered on the People costs tab: the salary, start date and on-cost settings, from which Float calculates the monthly cost lines and shows them under People costs in the cash flow table, with a toggle to switch the hire on and off and see the effect on the closing balance. The on-cost settings are editable defaults for the countries the template covers; check them against the current year's rates before relying on them. Once the hire is confirmed, its costs are entered as budgets in the forecast.

The scenario itself is a layer. Scenarios in Float stack budgets on top of the base forecast, which is made up of your budgets, invoices, bills and reconciled bank transactions, and the graph shows the base line and the scenario line together. A scenario changes budgets only; it cannot alter an invoice, a bill or a bank transaction, which is the reason the base stays intact underneath it. The contract's receipts go in as a repeating budget from the first expected payment date, stepped up over the ramp months; once the first invoice is raised, the invoice's expected date takes over. Duplicate the layer to run the second start date, compare the two on one view, and when the offer is accepted move the scenario's budgets into the base. Every plan includes a set allowance of scenario layers, listed on our pricing page, with more available at additional cost.

The cash floor is a threshold. Set it and Float shows the date the balance is due to cross it, so comparing two start dates is a matter of comparing two dates. Four user roles, including a read-only viewer, keep a scenario with a planned salary in it visible to the finance director without being editable outside the team, and the forecast exports to PDF or CSV whenever someone asks for it; nothing is sent on a schedule.

Frequently asked questions

How do I work out the full cost of a new employee in the UK?

Start from the salary and add the statutory items at the current year's rates: employer National Insurance at 15% on earnings above £5,000, an employer pension contribution of at least 3% of earnings between £6,240 and £50,270, and statutory sick pay from the first full day of any absence. Then add the one-off costs in the month they are paid, the recruitment fee, equipment and onboarding, and any benefits with their Class 1A charge. On the statutory items alone a £42,000 salary costs about £48,600 a year; the one-offs decide the first-year figure.

When does employer National Insurance actually leave the bank?

With the PAYE remittance for the tax month in which the salary was paid. Tax months run from the 6th to the 5th, and the remittance is due by the 22nd of the following month if paid electronically, the 19th by post, so National Insurance on a salary paid on the 25th leaves the bank about four weeks later. If the 22nd falls on a weekend or bank holiday, the payment must reach HMRC on the last working day before it unless the method clears on the day.

Does the Employment Allowance reduce the cost of a new hire?

Only if there is allowance left. The Employment Allowance is £10,500 for 2026/27 and is set against the employer's total employer National Insurance for the year, not against each employee, so a business already running a payroll of a dozen or more has usually used it before a new person's first payday. Model the marginal hire at the full 15% unless you know otherwise.

Is it true that an employee costs 1.25 to 1.4 times their salary?

Not as a UK fact. The range is a US rule of thumb that bundles American payroll taxes and health benefits, and no UK statistics body or professional body publishes a per-hire multiple. Built from the 2026/27 statutory rates, a UK salary costs about 1.16 times itself on the ongoing items; the first year lands higher because the recruitment fee and equipment are paid once. Use the dated cost lines rather than a multiple.

How much cash runway should we have before hiring?

No professional body prescribes a number of months of runway or headroom before adding staff, and runway in the cash-divided-by-burn sense is a start-up measure that does not fit a trading business. The test that works is whether the forecast clears your cash floor in every month with the hire in it, read on the tightest month rather than the average, and whether the receipt or contract the role is hired against is dated and real.

How do I model a new hire's ramp-up in a cash flow forecast?

As a labelled assumption about output, not as a cost line. Scale down the receipts the role is expected to generate in its first months, for instance half in month one and three-quarters in month two, and state the assumption next to the table. The only dated UK study, from 2014, found ramps of anywhere from about 15 weeks to a year depending on the person's background, so there is no benchmark to borrow; your own view of the role is the input.

Should we delay the start date if the forecast is tight?

Run the same hire at a second start date before deciding. Moving the start by a month or two lets the base forecast's own low points pass before the new cost lands, and often turns a breach into a clearance with a small margin. The price is the delayed contract cash and whatever the role would have brought in, and the second date should carry a condition and a trigger of its own, because a thin margin is a margin that a single late receipt removes.

How does Float model a new hire?

On the People costs tab: enter the salary, start date and on-cost settings and Float calculates the monthly cost lines, shows them under People costs in the cash flow table, and lets you toggle the hire on and off to see the effect on the closing balance. The on-cost settings are editable defaults and should be checked against the current year's rates. Run the decision as a scenario layer on the base forecast, duplicate the layer for a second start date, set a cash threshold to see the date each version crosses it, and once the hire is confirmed, enter its costs as budgets and move the scenario into the base.

The value of the hiring scenario is the month it shows you before the bank does: the one where the third salary, the agency invoice and the VAT quarter land together, with most of the contract's cash still a month or more away. Start a free 14-day trial of Float and run the role you are deciding on now as a layer on a live forecast of your own numbers.