How to price freelance work without guessing

Two questions get welded together here and they should not be. 'What must I earn per hour to survive?' is arithmetic and produces a floor. 'What is this work worth to this client?' is judgement and produces a price. Charge the first number and you will spend years working hard, hitting your targets, and quietly wondering why it does not feel like enough. It does not feel like enough because you have been invoicing your costs.

Work out the floor first

Target income plus business expenses, grossed up for self-employment tax, divided by billable hours. Billable hours are typically 50–65% of hours worked once sales, admin and rework are counted — using 2,080 is the single most common reason freelance rates come out too low.

Then price the project, not the hours

Estimate the hours honestly, add the revision rounds you will include and a risk buffer, and quote a fixed number. Getting faster then makes you more money instead of less.

Cap revisions in writing

Uncapped revisions turn profitable quotes into unprofitable ones within two rounds. Two included rounds with a stated price for additional ones is the standard that protects both sides.

Raise prices on new clients first

It tests the number without risking existing relationships, and it gives you evidence before the conversation with clients you already have.

Know your walk-away number

Below your cost floor, work makes you poorer than doing nothing. Knowing that number is what lets you decline quickly and without agonising.

Why the salary-divided-by-2080 method fails

Then price the project, not the hours

The hourly number is a floor for deciding what to accept. The price should come from scope and value, so that getting faster makes you more money rather than less — the entire problem with billing by the hour is that it charges you for improving.

The two questions people merge, and should not

“What must I earn per hour to survive?” and “What is this piece of work worth?” are unrelated questions with unrelated answers. The first is arithmetic on your costs and produces a floor. The second is a judgement about the client's alternative and the value of the outcome, and produces a price. Merging them is why so many freelancers quote their survival number and then feel underpaid for years — they have been charging their cost, not their price.

Where the billable-hours assumption breaks

Ask a freelancer what share of their week is billable and most say 80%. Ask them to track it for a fortnight and the answer comes back between 50 and 65%. The missing time is not idleness: it is scoping calls that do not convert, invoicing, chasing payment, tooling, admin, and the rework that follows unclear briefs. Because the divisor sits on the bottom of the rate calculation, a 30-point error there produces a rate roughly a third too low — and it compounds with the other optimistic assumptions in the same direction.

What each assumption costs you

Moving to project pricing without losing the argument

The objection you will hear is “how do I know it will not take you an hour?” The answer is that they are not buying hours, and that the risk transfer runs the other way too: if the work takes three times as long, the price does not move. Framed as risk rather than as time, fixed pricing is an easier sell than most freelancers expect — and it is the client's preference in almost every category where the outcome matters more than the process.

The revision clause that decides profitability

Two included rounds, a stated price per additional round, and a definition of what counts as a round. That last part is where the money leaks: without it, eleven separate emails each containing one change are eleven rounds delivered as one. Define a round as a consolidated set of feedback delivered at once, and both sides behave better.

Raising rates on existing clients

Apply the new number to new clients first and let it run for a quarter, so the conversation with existing clients starts from evidence rather than nerve. When you do have it, give notice, give a date, and do not apologise or over-explain — a rate change presented as an operational fact is accepted far more often than one presented as a request. Expect to lose someone. The arithmetic usually survives losing one client at the old rate and gaining none.