What a website actually costs in 2026 — an honest breakdown
Why quotes for the same site swing so wildly, what you're really paying for, and how to tell a fair price from a bad one — without needing to be technical.
The honest answer to "how much does a website cost" is the one nobody likes: it depends. But that's a cop-out, and you came here for something useful. So instead of a single number, let me show you what actually drives the price — and why the same brief can be quoted at wildly different figures depending on who you ask. Once you see what moves the number, you can tell a fair price from a rip-off in either direction.
The relative picture
This is the relative cost of the build — a comparison between project types, not a quote. Every project is scoped and priced individually; the domain (~$12/year) and hosting (often $0) sit outside it either way.
| What you're getting | Relative cost |
|---|---|
| Template site you assemble yourself (Wix, Tilda, Squarespace) | Subscription only |
| One-page / small business site, custom-built | $ |
| Multi-page marketing site, multilingual, SEO-ready | $$ |
| Web application (auth, dashboards, real data) | $$$ |
| SaaS MVP (multi-tenant, billing, the works) | $$$$ |
The trap isn't the absolute number — it's a mismatch between the price and the row. A marketing site quoted like a SaaS build is a red flag; so is a SaaS quoted like a one-pager, because nobody builds multi-tenant software that cheaply, and the corners they cut will cost you more later.
Why the same brief gets wildly different quotes
Four things move the number more than anything else.
1. Custom vs. template. A template is cheap because the work is already done — you're renting someone else's design and paying monthly. A custom build costs more up front but you own it, it's faster, it ranks better, and it looks like your business, not a theme 10,000 others use. Neither is "right" — it depends on whether the site is a formality or a sales channel.
2. Who's doing it. An agency has account managers, designers, project managers and overhead — you pay for all of them. A solo engineer who does the whole thing has none of that markup. The trade-off is capacity, not quality: an agency can run ten projects at once; one person takes a few at a time but you talk directly to the person writing the code.
3. Scope clarity. The single biggest cost driver isn't features — it's uncertainty. A vague brief ("a site for my business, modern, like Apple") gets padded with risk. A clear one ("five pages, these sections, in English and Ukrainian, contact form to my email") gets a tight number. The more precisely you can say what you want, the less you pay for the unknown.
4. The invisible 60%. The page you see is maybe 40% of the work. The rest is responsive layout, accessibility, SEO, performance, forms that actually deliver mail, analytics, security headers, and the boring infrastructure that keeps it fast and online. Cheap quotes usually skip this — and you find out when the site is slow, invisible to Google, or the contact form silently drops leads.
What a fair quote includes
When you compare offers, you're not really comparing prices — you're comparing what's inside them. A fair one spells out:
- Fixed scope, in writing. Exactly what gets built, before any money moves. No "we'll see as we go" — that's how a fixed-price site quietly grows into an open-ended invoice.
- Mobile + SEO + performance, not as paid extras but as table stakes.
- Who owns the code and the domain at the end. The answer should be you.
- What happens after launch — handoff, small fixes, who you call when something breaks.
- Payment terms. I work the way I'd want to be treated as a client: scope and price agreed up front, and you pay after the site is live and you're happy with it — no upfront deposit. Not everyone does this, but it tells you how confident someone is in their own work.
"Can't I just use AI / a builder for free?"
Sometimes, yes — and you should. If you need a one-page placeholder so you exist online, a builder is genuinely the right call. Don't pay someone thousands for what a Saturday on Tilda would do.
Where it breaks down is the moment the site has a job: rank for the searches your buyers use, integrate with something, handle real data, load fast on a phone in a bad signal, and not embarrass you next to a competitor. Builders hit a ceiling there, and AI-generated code without someone who understands it becomes a liability the first time it needs to change. The cost of "free" shows up later, in leads you never see.
How to actually decide
Forget the absolute number for a second and ask three questions:
- What is this site's job? A digital business card and a lead-generating sales channel are different products at different prices. Be honest about which you need.
- What does it earn or save? A site that brings two clients a month pays for itself fast; pure cost-cutting on it is a false economy. A site nobody will ever search for shouldn't cost much at all.
- Do I trust this person to still be there in six months? The cheapest quote from someone who vanishes after launch is the most expensive one.
If you're weighing a project and want a straight answer on what it'd actually take — scope, range, and whether it's even worth building — that's exactly the conversation I like having. You can see what I build or just tell me what you're working on; I usually reply within a few hours.
Building something similar?
Tell me what you're working on. I take on a small number of projects at a time.