In This Article
A website can cost sixty dollars a year or six figures, and both answers are true. That spread is why almost every guide on this question feels useless. They're all technically correct, and not one of them tells you what you're going to pay.
The more useful thing to know: Your price is mostly decided before anyone sends you a quote, by choices you're probably making without realizing you're making them. Once you can see those choices, the number stops being mysterious and starts being something you can steer.
I've been building websites since 2007, for one-person shops and for companies like Disney and Owens Corning. The gap between those projects is enormous. The reasons for the gap are the same handful of things every single time.
The short answer
For a small business in 2026, here's the honest lay of the land:
- Do it yourself on a builder like Squarespace or Wix: roughly $200 to $600 a year once you count the plan, the domain and any add-ons you end up needing.
- A template site set up by a freelancer: commonly $1,000 to $5,000, one time.
- A custom-designed small business site: commonly $5,000 to $15,000.
- A site with real machinery behind it (bookings, payments, a customer login, a connection to software you already run): $15,000 and up, sometimes well up.
Those are market ranges, not my rates, and they line up with what GoDaddy, Forbes and the rest will tell you. The ranges have never been the problem. The problem is that nobody explains why a project lands where it lands, so you're left holding a number with no way to judge whether it's fair.
Everyone answering this question is selling something
Search "how much does a website cost" and look closely at who's answering. Wix, Squarespace, GoDaddy and Elementor all sell website builders. Forbes Advisor earns affiliate revenue on the tools it recommends. Most of the rest of page one is Reddit threads where a hobbyist quoting $300 and an agency owner quoting $30,000 sit side by side with equal confidence.
None of them are lying. They're answering from where they sit. But a company selling a $16 a month builder has a structural reason to make this sound small, and an agency has a structural reason to make it sound big. When every source has a stake in the answer, weight the answers accordingly.
So let me declare mine. I build websites for a living, which makes me exactly as non-neutral as everybody else on this page. What I can do is show you the mechanics, so you can size the thing yourself and know when someone's number doesn't add up.
The number on a website quote is mostly a measure of how many decisions are still unmade when the project starts.
The price is set before anyone quotes you
By the time a developer gives you a figure, most of that figure is already locked in by facts about your project. A quote isn't a price list. It's an estimate of unknowns, and whoever writes it has to cover the risk of the ones that bite.
That's why two shops can quote wildly different prices for what sounds like the same website. They aren't pricing the same project. They're pricing different assumptions about what you'll need, how often you'll change your mind, and who's writing the words.
What moves the number
Six things, in rough order of how much damage they do to a budget.
1. Unique layouts, not page count
A forty page site where thirty eight pages share one template is cheaper to build than a six page site where every page is designed from scratch. Page count is the number everyone asks about and it's close to irrelevant. Ask how many unique layouts are being designed, and you'll learn more in one question than in an hour of discussion about pages.
2. Who writes the words
Copy is the single most common thing that stalls a project, and a stalled project costs money. If you hand over finished text, the build moves. If you want the developer to write it, that's a real service with a real price. If nobody is clear on whose job it is, the site sits at ninety percent done for four months while everyone waits on somebody else. I've watched that happen more times than I can count, and it's never the technology's fault.
3. Whether it has to talk to anything else
A brochure site stands on its own. A site that pushes leads into your CRM, syncs inventory, takes bookings against a real calendar, or logs customers into a portal is a different category of work. Every connection to another system is a place where somebody else's software can change under you, and that maintenance risk is priced in whether or not anyone names it. This is usually the line between a website and custom software, and it's worth knowing which side of it you're on.
4. How many people have to approve it
Cost scales with approvers, and not in a straight line. One decision maker is fast. Three is manageable. A committee of seven who all see the design for the first time in week six will double your budget, and nobody will be able to point at the line item where it happened.
5. What happens to what you already have
Carrying your existing URLs, rankings and content across to a new site is real work. Throwing it all away is cheaper on day one and can be brutally expensive later, especially if you currently rank for anything worth having. If your site brings in business today, protecting that is part of the job, not an upsell. That's most of what SEO work is.
6. Your deadline
Rush work costs more in every trade and this one is no exception. A date tied to something real, a trade show, a funding round, a lease starting, is worth paying for. A date that exists because it sounded good in a meeting is not, and softening it is the cheapest change you can make to a quote.
Is $1,500 a good price for a website?
It can be excellent, and it can be the most expensive fifteen hundred dollars you ever spend. It depends entirely on what's inside it.
At that price you're realistically getting a template, lightly customized, with your logo and your text dropped in. For a lot of businesses that's the right call. A two person landscaping company that needs to look legitimate and get phone calls does not need a custom build, and anybody telling them otherwise is selling.
What turns fifteen hundred into a bad deal is buying a site you don't own or can't change. Three questions sort this out before you sign:
- If we part ways, who owns the site, and can I take it with me?
- What does it cost me to change a phone number, a price or a photo next year?
- What happens if it breaks in month four?
If the answers are "you do," "nothing, you can do it yourself" and "I'll fix it," then fifteen hundred is a fine deal and you should take it. If the answers are vague, the low build price is the bait and the recurring fee is the actual business model.
The costs that never show up in the quote
The build price is the part everyone compares. The ownership cost is the part that decides whether the site was worth it.
- Hosting and the domain. Modest and predictable. Budget somewhere around $15 to $50 a month for a small business site, plus about $20 a year for the domain.
- Updates and security. On WordPress especially, core and plugins need patching or the site eventually gets compromised. That's a genuine recurring cost, paid either in money or in your Saturdays.
- Theme and plugin licenses. The $60 theme with four premium plugins bundled in quietly becomes $400 a year in renewals, and pieces of the site stop working if you let them lapse.
- Changes. Every site needs them. What matters isn't whether changes cost something, it's whether a small change is easy or turns into a support ticket and a two week wait.
- Your own time. The biggest hidden cost and the one nobody invoices you for. If a cheap build means twenty hours of you reviewing, rewriting and chasing, price your hours honestly and add them in. Sometimes the cheap option still wins that math. Often it doesn't.
How to compare two quotes that look nothing alike
When two numbers are far apart, the gap is almost always information rather than greed. Ask both people the same six questions and the difference will surface fast:
- How many unique layouts are you designing?
- Who's writing the copy, and is it in this price?
- What system am I getting, and can someone non-technical edit a page in it?
- What's included after launch, and for how long?
- What are the recurring costs, all of them?
- Who owns the code, the domain, the hosting account and the analytics?
Run both quotes through those and they usually stop resembling each other. That's the point. Once you can see that one is a template refresh and the other is a rebuild with content and integrations, you're not choosing between two prices anymore. You're choosing between two projects, which is a much easier decision to make well.
If you're weighing this because your current site isn't pulling its weight, it's worth separating "needs replacing" from "needs fixing." I wrote about the signs that justify a redesign, and a fair number of sites don't need one. And if you've already built something with AI and want to know whether it's shippable, that's a different question with a different answer.
Why I don't put a price on this page
You'll have noticed I haven't given you my number. The reason isn't coyness, and I'd rather say it plainly than let you wonder.
A rate card works when what you're selling is standardized. A dentist can publish the price of a cleaning because a cleaning is a cleaning. Websites aren't like that. Two clients ask for "a new website" in identical words, and one needs six pages and their photos organized while the other needs an inventory feed, three integrations and a customer login. Quoting both off the same table means overcharging one and underdelivering to the other, and I'd rather do neither.
What I will do is tell you on a first call roughly where your project lands and why, in enough detail that you could take it to somebody else and get a sane comparison. That includes the times the honest answer is that you don't need me. Some businesses belong on Squarespace, and I'll say so rather than sell you something bigger.
If you want that conversation, here's what I do on the web side, and the custom website work is where most of these questions end up getting answered properly.
