Hourly, Fixed or Value-Based: One Project, Three Quotes

A designer I know quoted the same six-page marketing site three times in one month. Same scope, same client profile, same two weeks of work. The first quote was an hourly rate. The second was a flat fee. The third was a number tied to what the site was supposed to earn. All three were honest. All three were defensible. And the money they produced was not remotely the same.

That is the part the usual advice skips. “Stop charging hourly” is a slogan, not a decision. What actually separates the three pricing models is not how sophisticated you sound in a kickoff call — it is who is holding the bag when the estimate turns out to be wrong. Once you see it that way, choosing between them stops being a personality test and becomes arithmetic plus one honest question about the client.

The estimate is always wrong; the model decides who pays for that

Every project has a number nobody knows at the start: how many hours it will really take. You guess. The client accepts your guess. Then reality adjusts it, usually upward, because of a stakeholder who appears in week two, a brand asset that does not exist, or a round of revisions nobody scoped.

Hourly, fixed and value-based pricing are three different answers to that single uncertainty:

• Hourly hands the risk to the client. The meter runs; overruns are billed.
• Fixed price moves the risk to you. You quoted a number and you live inside it. This is the same structure as a fixed-price contract in procurement, where the seller absorbs cost overruns by design.
• Value-based prices the outcome instead of the effort, which is why the classic definition ties the price to the buyer’s perceived value rather than to your cost of delivery. You still carry the delivery risk, but your upside is no longer capped by the clock.

Nothing about that ranking says one is more professional than another. It says they distribute the same uncertainty differently.

One project, three quotes, run through the numbers

Here is the designer’s project made concrete. Working rate: $85/hour. Honest estimate: 80 hours. The fixed quote is that estimate at that rate ($6,800). The value-based quote is $12,000, pitched against a site the client expects to carry a product launch.

Now watch what happens when the estimate is wrong in each direction. The figures below are straight arithmetic on those three quotes — not survey data, just what each model pays out at 80, 112 and 60 hours of real work.

What actually happens	Hourly at $85/h	Fixed at $6,800	Value-based at $12,000
Lands on the estimate (80 h)	$6,800 — $85/h	$6,800 — $85/h	$12,000 — $150/h
Runs 40% long (112 h)	$9,520 — $85/h	$6,800 — $61/h	$12,000 — $107/h
Comes in 25% short (60 h)	$5,100 — $85/h	$6,800 — $113/h	$12,000 — $200/h
Your reward for getting faster	None — you earn less	Every hour saved is yours	Every hour saved is yours, at a higher base
Who absorbs the overrun	The client	You	You

Two things jump out of that table, and both of them matter more than the headline number.

The first: hourly is the only model where your effective rate never moves. That is its whole appeal and its whole ceiling. You cannot have a good month by being good at your job; you can only have a good month by working more hours.

The second: fixed price and value-based are mechanically identical — both pay a flat sum, both punish overruns, both reward speed. The only difference between $6,800 and $12,000 is the story you told to justify the number. That is worth sitting with, because most freelancers who say they want to “move to value-based pricing” are actually describing a fixed-price quote with a better argument attached.

Hourly earns its keep in exactly two situations

Hourly gets dismissed too quickly by people selling pricing courses. It is the correct choice when the scope genuinely cannot be pinned down: ongoing maintenance, open-ended consulting, the first month with a client whose internal chaos is still an unknown quantity. Billing the meter is not a failure of nerve there. It is the only honest way to sell work whose size nobody can state.

It is also the right call when you are new to a category. If you have built two Shopify stores, your estimate for the third is a guess with a confidence interval the size of a barn. Hourly lets you learn the shape of that work on the client’s clock instead of eating a 60-hour miss on a flat fee.

The trap is staying there. Hourly billing pushes you toward tools and habits that measure the wrong thing — every hour tracked, every hour justified — and it makes every efficiency you gain a pay cut. If you are billing by the hour, at least make the tracking and invoicing cheap and automatic; our rundown of invoicing software for freelancers covers the tools that turn tracked time into a sent invoice without a manual step.

Fixed price is a bet on your own estimating

Flat fees are where most freelancers should be for most defined project work, and where a lot of them get quietly hurt. The bet is simple: you are wagering that your estimate is closer to reality than the client’s expectations are to infinity.

You win that bet with two documents, not with confidence. The first is a scope definition tight enough that “one more page” is visibly a new line item — which is really a question of how you open the engagement at all, and the onboarding sequence you put in front of a new client decides how much of that gets settled before the work starts. The second is a change-order clause with a price in it, because unpriced changes are how scope creep converts a 40% profit margin into an apology.

The padding question

Should you add a buffer to a fixed quote? Yes, and it should be a number you chose rather than a feeling. Look back at the last five projects you delivered, calculate how far over your estimate each one landed, take the worst of the five, and add that percentage. If your misses run 20-30%, a 25% buffer is not padding. It is the actual cost of the work, and you have been donating it.

Value-based pricing needs a number the client says out loud

Value-based pricing is not “charge more because you are worth it.” It is charging against an economic figure the client has already accepted as true. That figure has to come from them, in a discovery conversation, before you quote anything.

Workable: a client tells you their current landing page converts at 1.4%, that a point of conversion is worth roughly $40,000 a year to them, and that the rebuild is meant to move exactly that. A $12,000 quote is now 30% of a first-year gain the client named. Not workable: you believe the site is worth $12,000 because it is good work. It may well be. But without their number, you are just quoting a higher fixed fee and hoping nobody asks how you got there.

Which means the requirement is brutally practical. Value-based pricing needs a client with a measurable outcome, a person in the room who owns that metric, and a project whose contribution to it is plausible. Sales pages, lead funnels, checkout flows, systems that visibly cut hours of someone’s payroll — those qualify. A brand refresh, a blog redesign, a set of icons: usually not, and forcing the frame there reads as bluster.

The three questions that pick the model for you

Run a prospective project through these in order. The first “no” is your answer.

1. Can I state the deliverable in one sentence a stranger could check? If no, price hourly. An unscopeable project on a flat fee is a slow-motion loss.
2. Have I done this specific thing at least three times? If no, price hourly or add a buffer you would be embarrassed to say out loud, then quote fixed.
3. Did the client name a number this work is supposed to move? If no, quote fixed. If yes, and they own that metric, quote against it.

That is the whole framework. Notice that two of the three questions are about you and your track record, not about the client’s budget or your confidence level.

What each model quietly costs you

Every model has a failure mode that does not show up in the quote.

Hourly leaks at the edges. The unbilled fifteen minutes, the “quick call,” the revision you did not log because logging it felt petty. On an $85 rate, four unlogged hours a week is over $17,000 a year. If you bill hourly, the tracking has to be ambient rather than a thing you remember to do.

Fixed price leaks at the end. The project is delivered, the client has one more small thing, and you do it because the invoice has not cleared. That is why the final-payment terms and the handover belong in the same conversation — how you close out and hand over an engagement is a pricing decision, not an admin one, and the gap between “done” and “paid” is governed by the terms you set, which is the whole argument in net 15 versus net 30.

Value-based leaks at the start. It requires discovery work — sometimes hours of it — before you can quote at all. Do that for free often enough and your effective rate across won and lost pitches drops below what you would have made billing hourly. The fix is to charge for the discovery as a small paid engagement, which also tells you fast whether this client will ever pay the big number.

What I would actually do, and who should ignore it

If I were rebuilding a freelance practice from scratch, I would price hourly for the first three engagements in any new service line, switch to fixed the moment I had three comparable projects to estimate from, and reserve value-based pricing for the two or three clients a year who arrive with a number attached. Most of the revenue would be flat fees. That is the boring answer and it is the one that compounds, because flat fees are the only structure where getting better at your craft makes you more money per hour.

Ignore that if your work is genuinely open-ended — fractional roles, retainers, ongoing ops support. Those are hourly or monthly by nature, and dressing them as fixed-price projects just means renegotiating every six weeks. Ignore it too if you are early enough that your estimates are still wild. Underquoting a flat fee twice in a row does more damage to a new practice than a year of billing by the hour.

One last thing worth saying plainly: none of this requires new software. The model lives in your proposal and your contract, not in your stack. Whatever you charge, keep the cost of running the business low enough that a bad quarter is survivable — a full freelance stack can run under $30 a month, and the rest of the freelance business and money material here assumes exactly that. Pricing is the lever with the most torque. Start there, and change one model on one client before you change all of them.