The four numbers behind a freelance rate (and why 40 billable hours a week is fiction)

Kayeni — Wed, 11 Nov 2026 02:38:54 +0000

Almost everyone sets their first freelance rate the same way — take the old salary, divide by 2,080, round up. It is a reasonable-looking number, and it falls apart the first time you check it against a real calendar.

Not because it is slightly low. Because it leaves out tax, leaves out the cost of running the business, and leaves out the fact that only part of a working week can be billed — and then assumes take-home and gross pay are the same figure.

Here is the full calculation, with real numbers, in the order the arithmetic requires.

The four numbers you need

Take-home target — what you want to actually keep in a year, after tax and after business costs. Not revenue. Not "what a salary would be". The money that lands in your account and stays there.
Business costs — the yearly total of everything you spend in order to be able to work: software, insurance, hardware replacement, accounting, hosting, bank fees, and the part of rent and internet that the work consumes.
Tax rate — the share of gross income that goes to income tax and self-employment tax combined, as a percentage.
Real billable hours — the hours you can actually put on an invoice, per week, times the weeks you actually work in a year.

Then there is a fifth input, which is not a fact about your situation but a decision: the buffer.

Step 1: start from take-home, not from salary

Take $60,000 as the target — the amount you want to keep.

A $60,000 salary and a $60,000 take-home target are not the same number, and this is where the shortcut breaks first. When you were employed, tax came out before you ever saw the money, and your employer carried costs that you now carry yourself. So $60,000 is the floor of the calculation, not the ceiling.

Step 2: add the business costs

Say $6,000 a year covers software, insurance, accounting, hardware and the home-office share of utilities.

These are not optional expenses you can skip in a lean month. They are the cost of being able to send an invoice at all, and they are spent out of revenue before anything reaches you — so they get added on top of your take-home target, not taken out of it.

Step 3: gross up for tax

Tax is charged on gross income, so to keep $60,000 after a 30% tax rate you have to earn more than $60,000. The gross-up is a division, not a multiplication:

$60,000 ÷ (1 − 0.30) = $85,714

Multiplying by 1.30 instead gives $78,000 — and $78,000 taxed at 30% leaves $54,600, which is $5,400 short of the target. The difference between dividing by 0.7 and multiplying by 1.3 is the difference between hitting the number and missing it.

Now add the business costs, because they are paid from gross income too:

$85,714 + $6,000 = $91,714

$91,714 is the revenue the year has to produce. Hold on to that figure — everything after this is about how many hours you have to sell in order to produce it.

(One honest note: the model gives business costs no tax deduction, so it is slightly pessimistic. If your costs are deductible where you live, the real requirement is a little lower than $91,714.)

Step 4: count the hours you can actually invoice

This is the step that moves the answer most, and the one people guess at.

A 40-hour working week is not 40 billable hours. It cannot be, by construction: billing requires work that is not billed. Writing proposals, answering email, invoicing, chasing late payments, bookkeeping, learning the thing you are about to charge for, and the revisions that were not in scope all happen inside the same week. None of them can go on an invoice as a line item.

So the input is not "how many hours do I want to work". It is "how many hours per week can I realistically put on an invoice". Twenty is a conservative figure for someone doing their own sales and admin — not a rule and not a statistic, but an estimate to replace with your own and then argue with yourself about.

The second half of the input is weeks. Fifty-two weeks, minus holidays, minus sick days, minus the quiet stretch at the start of the year: 46 working weeks is a reasonable starting assumption.

20 billable hours/week × 46 weeks = 920 billable hours a year

The minimum rate

Divide the revenue the year needs by the hours the year can sell:

$91,714 ÷ 920 = $99.69 per hour

That is the minimum. At $99.69, on those assumptions, the year produces $60,000 of take-home and pays for itself. Not a comfortable number — the break-even one.

The buffer, and why it is not padding

$99.69 is break-even against a set of assumptions that will not all hold. The client who pays late. The project that runs a week over. The month with no work in it. The laptop that dies. The buffer is the line item for the fact that the inputs are estimates.

15% is a reasonable starting point because it is large enough to absorb an ordinary bad month and small enough that it does not price you out of the conversation:

$99.69 × 1.15 = $114.64 per hour

$114.64 is the number to quote. $99.69 is the number below which the year does not work. Keeping both in your head is the useful part: one is a negotiating position, the other is a floor.

A day rate follows from the same figure — the recommended hourly rate times 7 gives $802.50 for a standard day — but it is only as good as the assumption behind it, and seven billable hours in a day is already an optimistic day.

Why "40 billable hours a week" is fiction

Run the same revenue through 40 billable hours a week and watch what happens:

40 × 46 = 1,840 hours
$91,714 ÷ 1,840 = $49.84 per hour

Half the rate, for the same take-home. The only thing that changed is the assumption about how much of the week can be invoiced.

That is why a rate set from "40 hours a week, 52 weeks a year" comes out looking generous and then fails in practice: the denominator was never 2,080. A working week has a fixed amount of non-billable work in it, and a rate calculation that leaves that work out has to recover it somewhere — usually by working evenings, or by quietly earning less than intended.

What moves the answer

The inputs are all yours, so the output moves with them. Here are three runs, holding take-home at $60,000, costs at $6,000 and tax at 30%:

Billable hours/week	Hours a year	Minimum	Recommended
10	460	$199.38	$229.29
20	920	$99.69	$114.64
30	1,380	$66.46	$76.43

The take-home target is fixed in all three rows, and the rate changes by a factor of three. That is the honest way to use the calculation: change one input at a time and watch which one moves the answer. The hours input moves it more than anything else, which is the opposite of what people expect when they start out arguing about the rate itself.

Using it

None of this is proprietary. It is four inputs and three operations. The value is in doing it with your own numbers instead of someone else's, and in noticing that the rate you were about to quote was set by an assumption you never checked.

If it helps to have the sums done as you type, there is a free calculator that runs exactly this method in the browser — six inputs, no signup, no email, nothing tracked:

https://4127a18b6ac2441468bb63f2150b219d.cto.site/calculator

Change the billable-hours field first. It is the one that moves the answer most, and the one most often left at 40.

Disclosure: I built the free calculator linked at the end of this article. I have not worked as a freelancer myself — everything above is arithmetic, not experience.

AI disclosure: an AI assistant helped draft and edit this article. Every figure in it was recalculated and checked before publishing.

DEV Community: Kayeni