What to charge for a website

Pricing · By the StellarReach team · 12 min read · August 2026

"What should I charge for a website" has no single answer, and any article that gives you one number is selling something. The same five-page brochure site is a $900 job and a $9,000 job depending on who's buying, what it replaces, and what else you're carrying.

What can be answered is the structure: which pricing model to use, what each one does to you over a year, and how to arrive at a number you won't resent. That's what this is.

The four models

Hourly

The default for most people starting out, and the one with the worst incentives in the business. Three specific problems:

It punishes competence. The faster and better you get, the less you earn for identical output. Ten years of experience makes you cheaper, which is precisely backwards.

It caps you at your calendar. There are only so many billable hours, and you can't work more of them than exist. Every income increase has to come from the rate, which you can only raise so often.

It makes every conversation adversarial. The client's interest is fewer hours; yours is more. Every change request becomes a negotiation about the meter rather than a discussion about the work.

Hourly is defensible for genuinely open-ended work — ongoing maintenance, an unclear brief you agree to explore. For "build me a website," it's usually the wrong instrument.

Fixed price for a defined scope

You name a number, the scope is written down, and efficiency becomes yours to keep. This is where most studios end up, for good reason: the client knows what they're spending, you know what you're earning, and nobody watches a clock.

The failure mode is scope creep, and it's entirely solvable by writing down what the price includes — number of pages, how many rounds of revisions, who supplies copy and photography, what happens if they want a booking system halfway through. Every fixed-price disaster traces back to a scope that lived in someone's head.

Value-based

Price against what the site is worth to the business rather than what it costs you to make. A booking site that adds twenty appointments a month to a clinic is worth a multiple of what the same effort is worth to a hobby blog.

This is the highest-earning model and the hardest to run. It requires a real discovery conversation, a client willing to discuss their numbers, and enough standing to hold the price. It doesn't work cold, and it doesn't work at the bottom of the market. Grow into it.

Retainer

Monthly recurring for ongoing work — updates, hosting, content changes, performance. Individually small, collectively the thing that ends the feast-and-famine cycle, because it's the only model here that pays you in a month where you win nothing.

Attach one to every project you deliver, even a modest one. Three or four retainers cover most solo operators' fixed costs, which changes how it feels to turn down bad work.

Roughly where the market sits

Useful as orientation, not as a quote. These are the tiers you'll encounter rather than a survey result, and they vary enormously by country and city:

Template-based small business site ($500–$2,500). An existing theme, their content, light customisation. High volume, fast turnaround, competing partly with DIY builders. The flat-price end of the market.

Custom small business site ($2,500–$10,000). Designed for them, built properly, some bespoke functionality. Where most competent freelancers and small studios live.

Custom with real functionality ($10,000–$40,000+). E-commerce, booking and payment flows, integrations, multi-stakeholder sign-off. Agency territory, and the point where discovery becomes a paid phase of its own.

If you're consistently at the bottom of your tier, the problem is usually positioning rather than the number — a generalist competes on price because nothing else distinguishes them.

The hidden cost nobody prices in

Here's the line item most freelancers leave out of every quote: the unpaid hours it took to win the job.

Count them honestly. The discovery call. The follow-up call. The proposal. The revised proposal. Then multiply by every prospect who didn't buy, because those hours were spent too. It's common for a closed project to carry four to eight hours of unpaid sales effort once you include the deals that went nowhere.

On a $3,000 project, that quietly removes a meaningful chunk of your effective rate. And it's why two otherwise identical designers can charge the same and earn very differently: one of them spends a day per deal winning it, the other spends twenty minutes.

Your effective rate is what you earned divided by every hour you spent, including the ones nobody paid for. That's the only number that matters.

Setting a flat price you won't regret

If you want to cut the sales cost, publishing a fixed price for a defined package is the strongest lever available — the price does the qualifying before anyone speaks to you. The people for whom it's wildly wrong self-select out, and the ones who remain have already accepted the number.

To set one you can live with:

Price the median, not the best case. Take your last five similar projects, find the one in the middle, and price that. Pricing the smoothest one guarantees you lose money on average.

Define the edges explicitly. Page count, revision rounds, who provides copy and images, and what is definitively not included. The three things that most reliably break a flat price are more pages than agreed, a logo that has to be designed from scratch, and anything transactional — bookings, payments, logins. Name those as separate conversations.

Take a deposit. Half up front, half on delivery is standard and filters out people who were never going to proceed.

Say "from." A starting price with an honest note about what moves it lets you quote a bigger job without having advertised a number you can't honour.

This is the mechanic behind the flat-price offer: a fixed price on every demo, three scope questions that decide whether the job still lands at it, and an invoice raised from your own Stripe account when it does — a consultation only for the ones that genuinely need one.

When to raise your prices

Two signals, and they point the same way. If you're winning nearly every job you quote, you're too cheap — a healthy close rate leaves some business behind. And if you resent a project while doing it, it was underpriced, whatever the spreadsheet says.

Raise on new quotes only, don't announce it, and don't justify it. Most people who try discover the objection they were braced for doesn't arrive.

For where the leads to quote come from in the first place, see how to get web design clients and warm leads vs. cold leads.