How to Price Your First Freelance Project: The Math, the Research, and Four Real Quotes

A first quote is usually a feeling dressed up as a number. Here is how to replace the feeling with something you can defend.

Last updated: 6 November 2026

Jessica Hislop had never pitched a project on her own when a client asked for a two-minute animation with a crowd of twenty characters walking around. Crowds frightened her, so she pulled a number out of thin air and sent it. It was far too high, and the client never answered. Months later she worked out why: they didn’t need twenty distinct characters, only a scene that felt busy. A simpler, cheaper version would likely have fitted the brief. She had priced her fear instead of the project (Creative Lives in Progress, July 2026).

Hislop’s error pointed upward. Most beginners worry about the opposite one. Both have the same root, which is that no calculation sits behind the number.

This guide is built from documented first-quote stories, published research, and arithmetic you can check yourself. Where the evidence is thin, I say so. Tax and fee figures were last checked on 3 November 2026, and they change, so confirm them before you rely on them.

The short version

• Price from the work, not from fear or from a competitor’s number. Start with an honest hour estimate, add a buffer for what you’ll forget, add the unpaid hours, then add costs.
• Ask for the budget before you quote. Two of the freelancers below say this single habit took most of the guesswork out of pricing.
• A quote is a scope document with a price on it. Deliverables, revision rounds, what costs extra, deposit, payment terms, all in writing.
• Your headline hourly rate is not your real hourly rate. In the worked example below, a rate that feels like $80 an hour ends up near $31 to $35 once real hours, costs and self-employment tax are counted.
• If you have to give something up, give up scope before price. If you do discount, make it visible, finite and traded for something (a testimonial, a referral).

Why you can’t just look up the right price

The standard advice is to research what others charge. It’s good advice and harder than it sounds, because freelance rates are invisible by default. A job listing shows a salary band. Nobody publishes what a logo, a landing page or a 3,000-word article actually sold for, and clients often won’t share their budgets.

What happens when people set a price alone, with nothing to anchor on? The best large-scale evidence comes from Upwork, one of the few places where freelancers publish their own rates.

Why does a gender gap belong in an article about first projects? Because it shows what self-set prices look like when skill and reviews are held constant: they still diverge by a wide margin, for reasons the portfolio doesn’t explain. That is a warning about pricing by instinct, whoever you are.

I looked for a study that measures first-project undercharging directly and didn’t find one. These papers show a broader pattern, and the first project is where that pattern starts.

Sources: Foong et al. 2018, Revelio Labs, Sept 2026, Revelio Labs, April 2026.

The first number sets the frame

Negotiation research explains why the opening quote matters so much. In three experiments, Adam Galinsky and Thomas Mussweiler found that whichever side made the first offer, buyer or seller, got the better outcome, and that first offers strongly predicted the final price. It held for face-to-face and email negotiations (Journal of Personality and Social Psychology, 2001).

The useful part is the exception. The effect vanished when the other side concentrated on information that contradicted the anchor: the first-offer maker’s alternatives, their reservation price, or the receiver’s own target. In practice, the defence against someone else’s anchor is dull preparation. Know your target before the call, and know what else the client could do with their money.

A companion finding is worth knowing too. Negotiators whose first offer was accepted immediately were less satisfied afterwards, even when the outcome was as good as or better than for those who haggled (Galinsky et al., 2002). So the small pang you feel when a client says yes at once is documented, and it is not proof you undercharged. These were lab negotiations, not freelance quotes, so treat them as a prompt to prepare and not as a law.

Four first quotes, and what they teach

Each case below comes from a public interview. I note where a story is self-reported or where the storyteller sells something.

Owen Harvey, photographer: the quote that felt great

Harvey’s first paid job arrived sideways. He worked at a photo lab that kept being asked to shoot antiques for collectors, had no photography department, and started handing customers his own number as “a freelance photographer”. His first quote, about £250 to photograph a collector’s antiques at home, beat his day rate at the lab, so he was pleased.

This is not an undercharging story, and I won’t bend it into one. What matters is what came next: he had to work out what people would pay against what he’d accept, and he describes the early period as largely guesswork. Today he tries to learn the client’s budget before quoting, so he’s neither far over nor embarrassingly under. For commercial work, he separates a day rate from a usage fee, and he urges newer freelancers to talk fees openly with peers.

Being happy with a quote tells you nothing about whether it was right. A budget conversation and a few honest peers do.

Nicola West, copywriter: the quote that did the heavy lifting

West’s first client came from a startup workshop: a new graphic-design company that needed a social media plan. She started from the average London freelance day rate of the time, estimated the hours for each stage, and put a fee on every stage. She also wrote down how many rounds of changes she’d allow and her late-payment policy.

It was a lot of work before the work, she says, and she believes it made the client more committed. Then comes the candid part. As her workload grew, she got casual, agreeing day rates without a quote or contract, and for the first time ran into payment problems and shifting expectations. She now takes a 50% deposit, because clients who had paid nothing sometimes walked away mid-project. She also avoids naming prices on the phone, since a spoken number arrives with no description of what it buys. She prefers a call to understand the need, then a written quote.

The written quote is the product. Skipping it feels efficient until the first dispute.

Jessica Hislop, animator: the fix after the panic

After the crowd-scene misfire, Hislop built a system. She sends detailed cost breakdowns, makes extras such as sound effects optional, quotes later amendments separately, and warns clients early when a timeline is hard to estimate. She sets both a day rate and an hourly rate so short jobs aren’t underpriced, and she accepts that some requests will simply be too low to meet.

Itemised, optional pricing lets the client shape the scope instead of walking away from one frightening figure. It also stops you giving away the extras.

Jonathan Stark, consultant: stop selling hours

Stark is the best-known voice for pricing the outcome. A former software developer, he stopped billing hourly when he went solo in 2006 and has said he roughly doubled his earnings in his first year, compared with his previous salary (Agency Management Institute podcast, Cloudways interview). He later wrote Hourly Billing Is Nuts and now teaches the method.

Three flags. The income claim is his own and unaudited. He sells books and coaching built on the idea. And by his own account, his first solo clients came with him from his old firm (Being Freelance), so a ready-made client base may explain part of that first year. None of that makes him wrong. It means you should test value pricing on a small project before betting a month’s income on it.

What the four have in common

None of them found a magic number. Each changed a process: learn the budget, define the scope, itemise, take a deposit, put it in writing. Two of the four first quotes weren’t low at all. That is why this article builds its case on method and not on a claim that beginners always undercharge.

Do the math you’re skipping

Most pricing guides wave their hands here, so let’s build a worked example. This is an illustrative model, not a documented case, and every assumption is labelled so you can swap in your own.

The scenario. A designer estimates a project at 15 hours and quotes $1,200. That feels like a respectable $80 an hour.

What actually happens. The project takes 24 hours, because there were extra calls, revision rounds, and file wrangling. Another 4 unpaid hours go on pitching, emailing, invoicing, and chasing payment. That’s 28 hours in total. The table runs the same quote through two routes: a direct client (a referral, say) and a platform that takes 10%.

A feeling of $80 an hour became roughly $31 to $35, and that is before federal and state income tax, health insurance, software subscriptions, or a single unpaid sick day. Change the assumptions, and the figures move, but the shape stays.

Where the assumptions come from.

• Self-employment tax is 15.3% (12.4% Social Security plus 2.9% Medicare) applied to 92.35% of net earnings, and you cover both the “employee” and “employer” halves (NerdWallet). The Social Security part stops at an annual cap, $184,500 for 2026 (Xero), which rarely matters in a first year. Half of the tax is deductible when you work out income tax; this simple model ignores that.
• Expenses at 6% sit inside a research range. A 2026 NBER working paper found self-employed workers spend an extra 5 to 8 percentage points of gross pay on unreimbursed expenses compared with employees. The same paper found something more useful for first-time pricers: people given no numbers expected freelancing to net them more than a job with the same gross pay, while people shown customised estimates of costs and taxes expected it to net less (Freeland, Garin & Koustas, NBER Working Paper 34843). Most of us underestimate the burden until someone does the arithmetic. Now you’ve seen it.
• The platform fee is a placeholder. Fee trackers report that Upwork has charged a variable freelancer fee of 0 to 15% per contract since May 2025, with about 10% commonly cited. It is shown to you before you submit a proposal. Confirm the current rate on the platform, and set it to 0% if your client comes direct.
• The 24-hour overrun and 4 unpaid hours are my assumptions, not research findings. The next section covers how to replace them with something better.

Estimate hours like someone who has been burned

The weakest number in any first quote is the hour estimate, and there is a well-documented reason. In a classic study of the planning fallacy, 37 psychology students predicted they’d finish their senior theses in 33.9 days on average. They actually took 55.5, and only about 30% finished by their own prediction. Even the average “worst case” guess, 48.6 days, came up short (Buehler, Griffin & Ross, 1994). A thesis isn’t a freelance project, and that is a 64% overrun in a lab sample, not a benchmark for your trade. My example’s 24-against-15 happens to be a similar size, which is a coincidence and not a finding.

A method that doesn’t depend on optimism:

1. List every task, including the ones clients don’t see: briefing calls, research, revision rounds, file preparation, delivery, handover.
2. Estimate each task in hours and add them up. That’s your base.
3. Add a buffer. If you’ve never done this kind of project, start high. My rule of thumb is 50%, and it is a rule of thumb, not a research result. After three similar jobs, replace it with your own measured overrun.
4. Add the unpaid hours (pitching, admin, invoicing, chasing). Track them on your first project and you’ll know the number for the second.
5. Borrow an outside view. Ask two or three peers how long a comparable job took them. People estimate other people’s tasks more accurately than their own, and Harvey’s advice to talk fees openly with peers applies to hours as well.

Treat your first project as your first dataset. Run a timer on everything, because that record is what makes your second quote better than your first.

Work backwards to a real quote

Pick the take-home rate you’d actually be happy with, multiply by total hours, then reverse the deductions.

The US formula (pre-income-tax take-home):

Quote = (target hourly take-home × total hours) ÷ [(1 − platform fee − expense share) × (1 − 0.9235 × 0.153)]

The second bracket works out to about 0.8587. Using a $50-an-hour target across 28 hours (a $1,400 goal, chosen as an example, not a recommendation):

If that feels like too much, notice the feeling. The math isn’t greedy. It is complete. And the number you send doesn’t have to equal the number you calculate. Before you write the email, decide three figures:

• Your floor: the lowest take-home rate you’d accept, from the math above.
• Your target: what you’d be glad to receive.
• Your opener: the quote you send, usually at or a little above target.

Creative Boom’s pricing guidance makes the same point about knowing your minimum before you negotiate (Creative Boom).

If you’re in the UK

The principle is identical, and the numbers differ. For 2026/27, self-employed people pay Class 4 National Insurance at 6% on annual profits between £12,570 and £50,270 and 2% above that. Compulsory Class 2 contributions were abolished in April 2024 (LITRG). Income tax comes on top: nothing on the first £12,570 of the tax year’s income (the personal allowance), then 20% in the basic-rate band, with higher bands above £50,270. Scotland sets its own income tax bands.

Because UK tax depends on your whole year’s profit, think in set-aside percentages for the band your profit falls into, not in a per-project formula:

Take the same example in pounds: a £1,000 quote, no platform, 6% expenses (£60), leaving £940 profit that lands in the basic-rate band. Class 4 NI is £56.40 and income tax is £188, so you keep £695.60, which is £24.84 an hour over 28 hours. Before income tax, it’s £31.56. If you’re just starting alongside a job or other income, the allowance may shield much of this, which is another reason to ask an accountant.

VAT matters only above a turnover of £90,000 for 2026/27 (Which?); if you register, quote ex-VAT and say so. Elsewhere, the structure holds (fee, expenses, social contributions, income tax) but the percentages won’t, so check yours locally. I’m not a tax advisor.

Hourly, fixed, milestone or value-based?

The advice you’ll find contradicts itself partly because the right answer depends on how well you can estimate, which is exactly what you know least on a first project.

If you’re nervous, a hybrid is reasonable: a fixed price for a firm scope, hourly for anything outside it. Hislop’s optional add-ons are a version of the same idea. If the scope is truly foggy, you can also propose a small paid first phase to define it, and quote the rest once you both know what you’re building.

Should you discount your first project?

You’ll meet this advice constantly: with no idea of your speed, find the market price and charge 20 to 30% less for your first one or two projects. One widely shared designer’s guide on UX Collective says exactly that (UX Collective).

The logic is easy to see. A discount lowers the stakes and can win a first review. Three things still bother me:

• I couldn’t find evidence that it works. It’s a common recommendation, not a measured one.
• It stacks on every hidden cost above. A 25% cut to a quote that already nets about $31 an hour leaves roughly $23.
• It sets the anchor discussed earlier, and first clients are often your first referral source.

There is a real argument on the other side, and it deserves a fair hearing. In a field experiment on the online marketplace oDesk, Amanda Pallais hired 952 randomly chosen inexperienced workers for a short data-entry job and gave them public ratings. Being hired, and being rated in detail, substantially improved their later employment outcomes (American Economic Review, 2014). That study didn’t test discounts, but it supports the idea that a first completed, publicly rated job has real value.

So the defensible version is this: if you want to pay for a first review, pay for it visibly and once. List the full price, show the discount as its own line, state that it applies to this project only, and ask for something in return: a testimonial, permission to show the work, an introduction. Otherwise, if a client can’t afford your quote, cut scope. Fewer pages, one fewer revision round, no rush delivery, a smaller first phase.

What goes in the quote

A quote is a shared picture of the work with a price attached. A skeleton to adapt:

1. Deliverables, specific and countable. “Five-page website,” not “website help.”
2. What’s not included. West’s advice is to overstate the obvious, because people don’t always imagine the same thing.
3. Revisions: how many rounds, and what a further round costs.
4. Timeline, and what you need from the client by when. Delay on their side shouldn’t be free.
5. Price, itemised where the project has natural parts, with extras marked optional.
6. Deposit and payment terms. West takes 50% upfront. Choose your own number, but choose one.
7. Late-payment policy, in plain words.

If you can, ask early: “Do you have a budget in mind for this?” A client who won’t answer still tells you something. A client who does has just removed most of your guesswork. If they won’t say, offer two or three scoped options at different prices and see which one they pick. That’s my suggestion, built on Hislop’s optional add-ons, and not a tested technique.

A quote email that doesn’t grovel

Hi Sam,

Thanks for the call. Here’s what I’d suggest, based on what you described.

Scope: [countable deliverables]. Includes two rounds of revisions. Not included: [list]. I’m happy to quote any of these separately.

Price: $X, with 50% due to start and 50% on delivery. Timeline is [dates], assuming I have your [materials] by [date].

If it helps, I can also put together a smaller first phase at $Y that gets you [outcome] sooner.

Happy to talk through any of it.

State the price once, plainly, and stop. Leave out “I know it might be a lot, but.” The price should be the shortest sentence in the email.

When they say “that’s too high”

Resist filling the silence with a discount. Work through these in order:

1. Ask what they were expecting, or what’s driving the concern. It’s often scope, not price.
2. Offer a smaller version instead of a cheaper version of the same thing.
3. Say no, politely. Hislop’s view applies: some requests are too low to meet, and that’s fine.

Pricing when AI speeds up your work

If AI tools cut a task from six hours to four, should your price drop by a third? Practitioner guides published in October 2026 mostly say no: change the price when the scope or value changes, not because your workflow got quicker. One argues that a faster process is a reason to protect your price or take on more work, not to discount it (Speckyboy, 4 Oct 2026). Another says the price should come from scope, difficulty, expertise, responsibility, turnaround, and the quality of the result, with no universal “AI rate” (Fueler, 3 Oct 2026).

These are opinions from practitioner blogs, not research, and the market may push back, especially in commodity categories where clients can generate a draft themselves. Two practical points hold either way. If a client asks directly whether you use AI tools, answer honestly. And check the client’s contract or policies for any rules on AI use before you start. Both are easier when your quote describes deliverables and outcomes instead of hours.

Frequently asked questions

How much should I charge for my first freelance project?

There is no honest universal figure. It depends on your discipline, location, the client and the scope. Work it out in three steps: estimate your real hours with a buffer, find the client’s budget and a few peers’ rates as reference points, then reverse the math above to confirm your take-home rate is one you can live with.

Should my first freelance project be hourly or fixed price?

Hourly protects you when the scope is fuzzy, and a fixed price rewards you when the scope is clear and you can estimate it. If you’ve never estimated this kind of work, consider a fixed price for a tightly defined scope with limited revisions, or hourly for anything outside it.

Should I discount my first project?

Not by reflex. If you do, make it a visible, one-time line item and ask for a testimonial or referral in return. If the client’s budget is tight, reduce the scope instead of the price.

How big a deposit should I ask for?

There’s no universal standard. Nicola West, a copywriter, takes 50% upfront after seeing clients leave mid-project when they’d paid nothing. For a first project with an unknown client, some deposit is better than none.

Do I need a contract for my first project?

At minimum, a written quote that both sides accept, covering scope, revisions, price, and payment terms. For larger or riskier work, a contract is wise. I’m not a lawyer, so for anything substantial, get legal advice.

What if the client won’t share their budget?

Offer two or three scoped options at different prices, and see which one they choose. You’ll learn more from that choice than from a guess.

What the evidence can’t tell you

I’d rather end with the gaps than pretend they aren’t there.

• No study I found measures first-project undercharging directly. The best research on self-set rates is about a platform, not about beginners, and it is observational.
• Most success stories are self-reported, and some come from people who sell coaching or courses. Survivorship is real: you hear from the people it worked for.
• The gender-gap studies show a pattern, not a cause. Revelio infers gender from names, and neither study can say why the gap exists.
• My worked examples are models. The 10% fee, 6% expenses, 24-hour overrun, and 4 unpaid hours are labelled assumptions. Plug in your own.
• Tax rules vary, and they change each year. The US and UK figures here were last checked in November 2026. I’m not a tax advisor, so ask a local accountant what you can deduct.
• Rates vary by field and location. I deliberately haven’t given you a “correct” number for your discipline, and anyone who does without knowing your niche is guessing.

Before you hit send

• ☐ Did I ask for the client’s budget, or at least their goal and deadline?
• ☐ Is my hour estimate padded for revisions, calls, and the things I can’t foresee?
• ☐ Have I added the hours nobody pays for (pitching, invoicing, chasing)?
• ☐ Have I run the numbers through fees, expenses and tax, and does my real hourly figure still feel acceptable?
• ☐ Do I know my floor, target and opener before the call?
• ☐ Is the scope written down: what’s in, what’s out, how many revisions?
• ☐ Is there a deposit and a payment date?
• ☐ If they push back, do I have a smaller version ready instead of a lower price?

Your first quote won’t be perfect. Every freelancer quoted here says as much, and all of them have adjusted since. The aim is a price you can explain, defend, and learn from.

Sources

Documented cases

• Nicole Fan, “What to charge for your first freelance project”, Creative Lives in Progress, 15 July 2026 (Jessica Hislop, Owen Harvey, Nicola West)
• Jonathan Stark: Cloudways interview; Agency Management Institute podcast; Being Freelance podcast

Research and data

• Foong, Vincent, Hecht & Gerber (2018), “Women (Still) Ask For Less,” Proc. ACM Hum.-Comput. Interact. 2, CSCW, Article 53
• Revelio Labs, “The Gig Work Gender Gap” (April 2026) and “The Gender Pay Gap Followed Women Into the Gig Economy” (September 2026)
• Freeland, Garin & Koustas (2026), “How Do Workers Think About The Costs and Benefits of Freelance Work?” NBER Working Paper 34843
• Galinsky & Mussweiler (2001), “First offers as anchors,” Journal of Personality and Social Psychology 81(4), 657–669
• Galinsky, Seiden, Kim & Medvec (2002), “The dissatisfaction of having your first offer accepted,” Personality and Social Psychology Bulletin
• Buehler, Griffin & Ross (1994), “Exploring the planning fallacy,” Journal of Personality and Social Psychology 67(3), 366–381
• Pallais (2014), “Inefficient Hiring in Entry-Level Labor Markets,” American Economic Review 104(11), 3565–3599

Tax and fees

• NerdWallet, “Self-employment tax” (15.3% rate and 92.35% rule); Xero, “Self-employment tax” (2026 Social Security wage base)
• Low Incomes Tax Reform Group, “What tax rates apply to me?” (Class 4 NI and thresholds, 2026/27)
• Which?, “Self-employed income tax, National Insurance and VAT” (VAT threshold, 2026/27)
• Upwork’s freelancer fee: confirm the current rate on Upwork’s own fee page before quoting through the platform

Practitioner guides

• UX Collective, “Freelance designer 101: pricing strategies” (first-project discount advice)
• Creative Boom, “Essential tips on how much to charge for your freelance work”
• Speckyboy, “Should Your Agency Charge Less for AI-Assisted Solutions?” (4 Oct 2026); Fueler, “How to Price Work When the Client Knows AI Can Do Part of It” (3 Oct 2026)

This article is general information, not tax, legal, or financial advice.