The sum almost everyone does first
You earned £60,000 as an employee. You are going freelance. What should you charge?
The instinctive calculation is to divide by the number of working hours in a year: 52 weeks × 40 hours = 2,080. That gives about £29 an hour. Round it up to £35 to be safe, and you are away.
This is wrong by roughly a factor of three, and the error is not a matter of degree. It is structural, and it comes from two things the sum silently assumes: that every working hour is a billable hour, and that the £60,000 was the full cost of employing you. Neither is close to true.
Here is what an employer was paying for that an invoice now has to cover:
- Employer pension contributions and payroll taxes — typically 15–20% on top of gross salary before anything else.
- Paid holiday, public holidays and sick leave — between five and seven weeks a year during which you were paid and produced nothing billable.
- Equipment, software licences, phone, insurance, workspace.
- Training, and the hours spent in meetings, admin and internal work that no client would ever have paid for.
- The absorption of gaps — a quiet fortnight cost your employer money, not you.
The £60,000 salary probably cost the employer £75,000–£85,000 all in. That is the number the freelance rate has to replace, out of a much smaller number of billable hours.
The number that ruins the arithmetic
The single most damaging assumption in freelance pricing is that a working year contains 2,080 billable hours. It does not, and it is not close.
| Item | Hours | Remaining |
|---|---|---|
| Nominal working year (52 × 40) | 2,080 | 2,080 |
| Holiday (5 weeks) | −200 | 1,880 |
| Public holidays (8 days) | −64 | 1,816 |
| Illness (5 days) | −40 | 1,776 |
| Sales, proposals, pitching | −250 | 1,526 |
| Admin, invoicing, bookkeeping, chasing payment | −180 | 1,346 |
| Training, tooling, keeping current | −100 | 1,246 |
| Gaps between contracts | −150 | 1,096 |
About 1,100 billable hours — roughly 53% of the nominal year. That is a normal, healthy freelance year, not a bad one. Established freelancers with steady repeat clients might reach 1,300–1,400. New ones, still building a pipeline, are often closer to 800.
Every unbillable hour is paid for by a billable one. At 53% utilisation, each hour you invoice has to carry itself and very nearly one other hour. That alone roughly doubles the naive rate, before tax or expenses enter the picture.
The formula, run backwards
Start from what you need to keep and work outwards, adding each layer that sits between the invoice and your bank account:
1. take-home needed (what you live on) 2. ÷ (1 − effective tax rate) → pre-tax profit needed 3. + business expenses → revenue needed 4. ÷ billable hours → hourly rate
Each step deserves a sentence.
Effective tax rate, not marginal. This is total tax and social contributions divided by total profit — income tax plus self-employment or national insurance contributions, after any allowance. For a mid-income sole trader it commonly lands between 25% and 35%. Using your top marginal band overstates it; using the basic rate understates it badly.
Expenses are the business's, not yours. Software, insurance, accountancy, equipment, professional subscriptions, workspace, travel. Add them up honestly — annual licences renewing in a quiet month are the ones people forget.
Billable hours, not working hours. The number from the section above. Be pessimistic; nothing in freelance pricing punishes optimism faster.
Worked example: from £45,000 take-home to a day rate
An experienced contractor, working alone
Target take-home: £45,000. Effective tax and NI rate: 29%. Business expenses: £7,200 a year. Realistic billable hours: 1,100.
pre-tax profit = 45,000 ÷ (1 − 0.29) = £63,380 revenue = 63,380 + 7,200 = £70,580 hourly rate = 70,580 ÷ 1,100 = £64 per hour day rate (7h) = 64 × 7 = £448 per day
Compare that with the naive sum. A £60,000-salary equivalent divided by 2,080 hours gave £29. The real answer is £64 — more than double, for a lower headline income than the salary it replaces.
Now the sensitivity, which is the part worth internalising. Hold everything else constant and vary only the billable hours:
| Billable hours | Required hourly | Day rate |
|---|---|---|
| 800 (building up) | £88 | £616 |
| 1,100 (steady) | £64 | £448 |
| 1,400 (busy, repeat clients) | £50 | £350 |
The same person, same expenses, same take-home — and the defensible rate ranges from £50 to £88 purely on how full the calendar is. This is why "what do others charge?" is such a poor guide. Their rate encodes their utilisation, not yours.
Day rates, retainers and fixed price
The hourly figure is the foundation. What you quote is usually something else.
Day rates should be built on a real working day — 6 to 7 productive hours, not 8. Quoting an 8-hour day at your hourly rate means either working an unpaid ninth hour or under-delivering, and the second one costs you the client.
Retainers are usually discounted, and the discount should be earned by something real: guaranteed hours, a paid notice period, or priority access. Discounting a retainer for the feeling of security, with no commitment attached, is just a lower rate. Where a retainer genuinely lifts your utilisation from 1,100 hours to 1,400, the table above shows a 10–20% discount is defensible; beyond that it is not.
Fixed price is the one that quietly destroys rates. A fixed fee is an hourly rate with the hours hidden and the overrun risk transferred to you. Estimate the hours, apply your rate, then add a contingency for the scope that always appears — 20–30% is normal, and the projects that overrun by more than that are the ones where the scope was never really defined. If you cannot estimate the hours within about 25%, the project is not ready to be priced fixed.
Raising a rate you have already set too low
Most freelancers discover this arithmetic after eighteen months of charging half what they should. The gap does not close by itself, and inflation makes a static rate a shrinking one. Some things that work:
- Price new clients at the new rate immediately. There is no negotiation and no history to explain. Within a year, most of your book has re-priced itself simply through turnover.
- Give existing clients notice, in writing, with a date. Sixty to ninety days is courteous and is what a supplier would do. It is a business communicating a price change, not a favour being withdrawn.
- Do not justify it with your costs. A client is buying an outcome, and your expenses are not their problem. "My rate from January is £X" is a complete sentence.
- Expect to lose some. Losing the bottom 20% of a client list on price usually raises total income, because those clients consume disproportionate unbillable time. Run the arithmetic before assuming you cannot afford it.
What this does not tell you
This calculation produces a floor — the rate below which the work is not worth doing on your own terms. It is not a market price and it says nothing about what anyone will pay.
- It ignores value. If your work saves a client £400,000, your cost base is irrelevant to what it is worth. Cost-plus pricing systematically underprices high-value specialist work.
- It ignores the market. If the floor comes out well above what your field pays, the problem is not the rate. It is the cost base, the utilisation, or the positioning.
- It assumes you get paid. Late payment and bad debt are not in the formula, and both are ordinary features of freelance life.
- The tax figure is an estimate. Effective rates depend on jurisdiction, structure, allowances and reliefs. Get the number from an accountant rather than a rule of thumb, and revisit it when the business changes shape.
The other half of this problem is what happens to the money after it arrives. Irregular income makes tax feel optional right up until it is not — see reserving tax on irregular income.