Pricing is the highest-leverage skill in freelancing and the one people spend the least time on. A 20% rate increase applied to the same workload is a bigger raise than most people get in five years of employment, and it takes an afternoon of thinking rather than a year of skill-building.
There are three pricing models in common use. They are not a maturity ladder where value-based is the enlightened endpoint — each one fits a different kind of work, and using the wrong one is what makes projects unprofitable.
Start with your floor rate
Before you can price anything you need to know the number below which work costs you money. Almost every freelancer computes this wrong by assuming a 40-hour billable week.
Realistically, sales calls, proposals, invoicing, chasing payment, admin, learning and dead time consume 40–50% of your working hours. If you bill 22 hours in a 40-hour week you are doing well.
| Step | Calculation | Result |
|---|---|---|
| Target take-home | — | $72,000 |
| Add tax + business costs (~35%) | 72,000 × 1.35 | $97,200 |
| Working weeks per year | 52 − 6 (holiday, sick, dead) | 46 |
| Billable hours per week | — | 22 |
| Billable hours per year | 46 × 22 | 1,012 |
| Floor rate | 97,200 ÷ 1,012 | $96/hour |
Your floor rate is a diagnostic instrument, not a price. You may never quote it to a client. You use it to check, after a project ends, whether that project was worth doing — and to notice when a “great” $6,000 retainer is quietly paying you $38 an hour.
Model 1: Hourly
Hourly billing is the default for a reason: it is easy to explain, easy to adjust, and it protects you when the scope is genuinely unknowable. It also has one structural flaw that everyone eventually runs into.
Hourly billing punishes you for getting better. The developer who can fix a broken payment integration in 40 minutes because they have seen it nine times earns less than the one who flounders for six hours. Your growing expertise reduces your income. That is not a sustainable structure for a career.
Use hourly when: the work is open-ended maintenance, ongoing support, or exploratory work where nobody can define “done” up front. Avoid hourly when: the deliverable is clear and your speed is an advantage.
Model 2: Fixed fee
A fixed fee prices the outcome rather than the clock. It is what most clients actually want, because it converts an unknown expense into a line item they can approve.
Quoting fixed fees safely comes down to three habits:
- Estimate the hours, then add 40%. Not because you are slow — because discovery, revisions and the client’s internal chaos are real work you always forget to count.
- Write down what is excluded. A fixed fee without a written exclusion list is an unlimited-liability contract with a number on it.
- Bound the revisions. “Two rounds of revisions; additional rounds billed at $X/hour” ends more disputes than any other sentence in freelancing.
Fixed fees also let you profit from efficiency. Build a component library, a deployment script, a set of templates, and the same deliverable takes half as long next time — at the same price. That gap is where a freelance business becomes something other than a job with worse benefits.
Model 3: Value-based pricing
Value-based pricing sets the fee as a fraction of the financial outcome the work produces. If a checkout fix recovers $200,000 in annually abandoned orders, charging $2,500 because “it only took two days” is leaving an enormous amount of money on the table for both parties.
It works only when three conditions hold at once:
- The outcome is measurable in money — revenue gained, cost removed, risk avoided.
- You are talking to someone who owns that number, not someone forwarding a brief.
- Your contribution is attributable — you can point at the change and the result.
The conversation, not the spreadsheet, is the hard part. It sounds like: “Before we discuss cost — if this works, what does it change for the business over the next twelve months?” Then stop talking. Whatever number they say is now the anchor, and it is nearly always larger than the one you were about to invent.
Price the problem, not the hours. Clients are not buying your afternoon; they are buying the version of their business where the problem is gone.
Which model, when
| Situation | Best model | Why |
|---|---|---|
| Ongoing maintenance / retainer support | Hourly or capped monthly | Scope genuinely unbounded |
| Defined deliverable (site build, migration, brand) | Fixed fee | Rewards your efficiency |
| Revenue-critical fix or growth work | Value-based | Fee tracks the payoff |
| New client, unclear requirements | Paid discovery, then fixed | De-risks both sides |
| Rush work, hard deadline | Fixed + 25–50% rush premium | Prices the disruption |
Paid discovery deserves special mention. Charging $1,500–$3,000 to produce a specification, technical plan and firm quote solves the estimation problem entirely: you are paid to learn enough to quote accurately, and the client owns a useful document whether or not they hire you for delivery. It also filters out tyre-kickers with near-perfect accuracy.
Retainers: the pricing model that fixes cash flow
The fourth model nobody lists is the monthly retainer, and it solves the problem that makes freelancing stressful: income that arrives in unpredictable lumps.
There are two kinds, and confusing them is how retainers go wrong.
- A block of hours. “20 hours a month, unused hours do not roll over.” Simple, easy to sell, and it re-creates the hourly problem — you are still selling time.
- An availability and outcomes agreement. “I own your site’s uptime, performance and monthly release cycle, and I respond within four hours.” You are selling a standing capability, and the price is not tied to hours consumed.
The second is better for both parties, but only with a written boundary — what is included, what counts as a project, and what the response commitment actually means. Without that, a retainer becomes an unlimited-support subscription priced for occasional help.
Price retainers so a client who uses the agreement heavily still leaves you at or above your floor rate, and expect light months to subsidise heavy ones. Two or three retainers covering your baseline expenses changes how you negotiate everything else, because you are no longer quoting from a position of need.
How to talk about money without flinching
Most underpricing is not an arithmetic failure. It is a conversation failure — the number was fine until it had to be said out loud to a person who paused before responding.
Three habits fix most of it:
- State the price and stop talking. The instinct to fill the silence with justification is where discounts are born. Say the number, then wait — the pause is the client thinking, not disapproving.
- Ask about budget early and plainly. “Do you have a range in mind for this?” is a normal professional question. A client who genuinely will not answer in either direction is telling you the conversation is not serious.
- Never apologise for a rate. “It’s a bit expensive, but…” invites negotiation you did not have to have. The rate is the rate.
And when a client pushes back on price, resist the reflex to cut. Ask what they had in mind, then reshape the scope to fit it. “I can do the core migration for that, and we can add the redesign next quarter” preserves your rate, keeps the relationship, and often results in both pieces being bought anyway.
Three quoting mistakes that cost real money
Quoting a single number. A single price is a yes/no question, and clients under budget pressure answer no. Offer three options — a lean version, the recommended version, and an expanded version with extras. The conversation changes from “should we?” to “which one?” A meaningful share of clients pick the middle or top tier, which they would never have reached from a single quote.
Discounting to win. Discounts teach clients that your prices are fiction and attract the buyers most likely to grind you on scope later. If a price needs to come down, remove something from the scope so the discount buys them less rather than costing you more.
Never raising rates on existing clients. Long-term clients quietly become your lowest-paid work through inflation alone. Review rates annually, give 60 days’ notice, and expect the reaction to be far milder than you fear. Clients who value your work rarely leave over 10%; clients who leave over 10% were freeing up your calendar for better ones.
The uncomfortable part
If nobody ever flinches at your price, it is too low. A healthy freelance practice loses somewhere around a quarter of its quotes — that ratio means you are pricing at the top of your market rather than the middle of it. Winning every single job is not a sign of excellence; it is a sign that you are the cheap option.
Rates are also downstream of where clients come from. Bidding platforms compress prices structurally because buyers see ten comparable quotes side by side — worth understanding before you conclude your rate is the problem. The real cost of platform fees and finding clients outside bidding sites both change what you are able to charge far more than any negotiation script will.