Value-based pricing for freelancers: how to do it, including the hard part

You are not supposed to hear the case against value-based pricing from a company that builds it into its software. North drafts tiered, outcome-priced proposals; that is a large part of what it is for. I still think the way the method gets sold to freelancers is broken in one specific place, and that place happens to be the entire difficulty.

The method itself is four moves. Establish what the finished work is worth to this client. Quote a share of that. Show the arithmetic where they can check it. Give them more than one way to say yes. Nothing there is controversial, and if you have ever read a pricing guide you have read those four moves before, generally as a numbered list.

Here is where it goes wrong. The most-repeated number in favour of all this is that freelancers who price on value earn forty-two per cent more per project than freelancers who bill hourly. We went looking for the study behind it in August 2026. There is no study. The figure appears on a guide published in April 2026 attributed to "a 2026 data study," and every other appearance we could find cites that guide. The citation is circular. The number sits near the top of a page that goes on to tell you to build your business on being able to quantify things.

That is not a reason to bill hourly. It is a reason to notice that the people selling this method skip the only part that is actually difficult: getting a real figure out of a real client, when the client has never thought about it in those terms and has no obligation to start.

The conversation, which is the whole job

Dan Mall publishes the sentence he says to buyers, which almost nobody does:

The opportunity seems to be generating $4M in additional annual revenue for you, and I typically start at 10% as I thought a 9× return on investment would be great for you.

Read it twice, because two things are happening and only one of them is the percentage. He says the client's own number back to them before he says his. And he states the ratio from the buyer's side of the table, as their return, not his fee. By the time the price arrives it is the third thing in the sentence and the least surprising.

Now notice what the sentence requires. It requires the client to have already told him that the opportunity is worth four million a year. That is the load-bearing element, it arrives before any of the pricing craft, and it is the part every guide waves at. So: how do you get it?

Ask what happens if it works, then stop talking. Not "what are your goals," which produces a paragraph of website language. What happens. Who does something differently on the Monday after launch. The silence after this question is long and you have to let it be long.

Ask for the unit, not the total. Most people cannot tell you what their brand is worth and can tell you instantly what one new wholesale account is worth, or one booked consultation, or one enterprise seat. Then ask roughly how many they do in a year. You have now built the total out of two numbers they were happy to give you, and they will believe it more than a figure they had to produce on the spot.

Ask what they are doing instead right now, and what that costs. Every problem you get hired for has an incumbent workaround, usually somebody's evening. The workaround has a price and the client knows it, because they are paying it.

Ask who else has to be pleased about this. This is the one that finds the actual buyer. A founder answers "me." A marketing director says "our CFO, if it moves trial signups," and has just told you the metric the money is attached to.

Then do the arithmetic out loud and get it slightly wrong. This is the move that works when nothing else has. "So if a new account is worth about nine thousand a year and you are adding four or five, we are talking somewhere in the region of forty thousand — is that the right shape?" People who will not volunteer a figure will correct one instantly. Correction is a much smaller social act than disclosure, and either way you leave the call with a number that came from them.

Two failure modes worth naming, because the guides pretend they do not exist. Some clients genuinely do not know: an internal team spending a budget, a pre-revenue startup, a grant-funded programme where the value is real and not denominated in money. And some clients know exactly and will not say, because a previous consultant did the four-million-times-ten-per-cent thing to them and they have learned that volunteering a figure is how the quote gets bigger. That second group is not being difficult. They have met this method before, badly done, and adjusted.

When the number does not exist

You have had the conversation. There is no figure. This is where value-based pricing quietly becomes something else, and the useful question is what.

Price the proxy. Not revenue, but a countable thing the client already tracks and already values: patients seen per week, applications received, hours the office manager spends re-keying orders. It is weaker evidence than money and it is real evidence, and it lets the proposal say what the work is for.

Price the status quo. What is the current situation costing, including the parts nobody has costed? The founder's Sunday. The two enquiries a month that go nowhere because the site does not say what the company does. You cannot invoice against a Sunday, but you can hold it up next to your fee and let the buyer make the comparison themselves.

Or price the scope and keep the framing. Deliverables, rounds, a schedule, a number you derived from your own floor. Then present it against the outcome anyway: here is what this is for, here is what it costs, here is when you have it.

That third one is where most value-based pricing ends up in practice, and the evangelists would rather not say so. It is worth saying so. A defined project at a defined price, quoted against a purpose the client recognises, beats hourly billing on the two dimensions that matter most for creative work: it stops paying you less for getting faster, and it ends the conversation about whether that Tuesday afternoon really took four hours. The method degrading into project pricing is not the method failing. It is the floor of the method, and the floor is well above where hourly leaves you.

The share, and the floor underneath it

Mall starts at ten per cent. The ten-to-twenty-per-cent band that circulates everywhere else has no survey behind it that anyone will name, and it has the tell of an invented range: round at both ends, no sample, no source, endlessly repeated.

The share is also the last thing to decide, not the first, and it moves. Ten per cent of forty thousand is a sensible fee. Ten per cent of four million is a fee nobody has ever paid a freelancer for a brand refresh, and quoting it is how a defensible method acquires a reputation for greed. As the underlying figure grows the fraction shrinks, because at some point the client is no longer buying a share of an outcome, they are buying a piece of work that helps produce it.

Underneath all of it sits a floor, and the floor is arithmetic. Take the hours the job will really take, multiply by the rate you need, and refuse to go below it however good the value story sounds.

Which brings us to benchmarks, briefly, because this is the least interesting section in the post and it should be. In a 2025 survey of web designers, the median billed $92.75 an hour and the average was $100, with revenue tiers running $76 to $131 (State of Web Designer Pricing, Web Designer Academy, 2025, n=208, around sixty per cent US). Broad aggregators looking at roughly the same job title land nearer thirty-five dollars an hour. Same title, three times apart, both published with a straight face. Use whichever is closer to your market to check that your floor is not absurd, then put it away. It is a sanity check on the bottom of your number and it has nothing to say about the top.

Say a studio brand refresh where the client has told you a new positioning is worth about $120,000 a year in enterprise deals they currently lose at the shortlist stage. A tenth is twelve thousand. Your floor on the work — roughly six weeks, at the rate you need — is fifty-two hundred. So the room is between those two figures, and the proposal lives there, at three tiers inside it rather than one number on top of it. That is a hypothetical and the figures are made up; the shape is not. Almost every one of these ends with a floor, a ceiling, and a decision about where in between you are comfortable. (This is the part North drafts for you from the notes you took on the call, tiers and sums included, as something you edit before anyone else sees it.)

The objections that hold up

Hourly billing is defensible and the people defending it are not fools. On genuinely murky scope it aligns incentives better than any fixed bid, because a fixed bid on unknown work is a bet you will usually lose. It makes leaving a bad engagement easy, which fixed-price work does not. And it removes the temptation, real in fixed-bid work, to do exactly enough to get paid and no more.

The strongest version of the argument I know comes from a developer on Hacker News, who pointed out that you do not really have an hourly rate at all: you have a day rate, used for jobs so small that a week would be silly. That is right, and it reframes the fight. The disagreement is not hourly versus value. It is about the size of the unit you sell, and every step up that ladder — hour to day to week to project to outcome — moves the risk from the client to you and the upside along with it.

There is a smaller, funnier piece of evidence for the same point. A freelancer on the same site started charging double for meetings, and reported the result in one line: he was suddenly invited to a lot fewer meetings. Price is not only compensation. It is also how people decide how much of you they need, and it works whether or not you meant it to.

Which is also the answer to "will raising my prices lose me clients." Yes. The community's standard reply to that worry is arithmetic and it is sound: scare off half your clients at double the rate and you have halved your workload for the same money. What the confident version leaves out is that you do not get to choose which half leaves, and some of the good ones will go. The trade is real and so is its cost. The fuller version of that argument is a post of its own, and a logo derived end to end, sums on the page is what the method looks like on one small job.

The thing that actually caps your income

"Charge more" has been the community's standing advice since Patrick McKenzie started giving it, and it is correct. It is also insufficient in a way that rarely gets said out loud.

Rate courage without lead flow is just anxiety with better posture. If you have two enquiries a quarter, doubling your price does not double your income; it changes which of the two you can afford to take. The binding constraint for most freelancers is pipeline, and pipeline is harder, less flattering and much less fun to write about than pricing psychology. Nobody builds an audience explaining how to find clients, because the honest answer involves a lot of unglamorous months.

So: price properly, and separately, get more at-bats. They are different problems and only one of them is solved by a better proposal. If you want the mechanical part of the pricing side handled, the project pricing calculator will do the floor arithmetic while you do the conversation.

One last thing, and it is Jessica Hische's argument rather than mine. Every logo sold for two hundred dollars makes the next designer's two-thousand-dollar quote sound insane to the person hearing it. Pricing sensibly is partly a favour to a stranger who will quote after you. That is not a reason to charge more than the work is worth. It is a reason not to charge less out of embarrassment.

Get the number from the client. Everything above only tells you what to do with it once you have.