What a small business website actually costs in 2026

Search this question and you'll get answers ranging from $0 to $48,000. That spread isn't because nobody knows. It's because of who is answering.

We went through the first page of results for the obvious searches. Every single article was published by a company that sells one of the options being priced — agencies quoting agency rates, website builders quoting their own subscription, affiliate sites earning a commission on whichever builder you pick. There is no neutral source ranking for this question.

We sell one of the options too, so treat this the same way. What we can do is show our work: real prices, sourced and dated, including the parts that don't appear in anyone's quote.

The three routes, honestly priced

There are really only three ways to get a website: build it yourself on a subscription platform, pay a freelancer, or hire an agency. Here's where each actually lands.

1. Do it yourself on a builder

Advertised from about $10–17/month. This is the number in every headline, and for most businesses it is the wrong number — see the tier trap below.

2. Hire a freelancer

This is where published estimates diverge most sharply, and the divergence is itself informative. Agency-published guides put freelancers at $50–150/hour. But Upwork's own platform data — from the marketplace where a lot of this hiring actually happens — puts the median web designer rate at about $21/hour, with a typical range of $15–30.

Both can be true. A US-based senior designer really does charge $100+/hour; the global median on a freelance marketplace really is around $21. What you should take from it is that “a freelancer costs $75/hour” is not a fact about freelancers, it's a fact about which freelancers the author has in mind. Project pricing for a small business site generally lands somewhere between $500 and $8,000 depending on the same variables.

3. Hire an agency

Roughly $6,000 at the boutique end, $12,000–35,000 and up for a full-service shop. For a café or a salon this is usually more website than the business needs, and the quote is what sends most owners back to option one.

The tier trap: the plan you need isn't the plan they advertise

This is the single biggest gap between what's published and what you'll pay, and almost nobody writes it down.

Builders advertise their cheapest paid tier. That tier reliably excludes the one feature a real business needs. If you take bookings, sell anything, or want to avoid a cut of every sale, you are on a higher plan than the one in the ad.

Follow that through for a hair salon that wants online booking. The advertised price is $16–17/month. The real one is Wix Core at $29/month, or Squarespace Core at $23 plus Scheduling at $16 — $39/month, about $468 a year, before a domain or an email address.

The costs that don't appear in the quote

Whichever route you take, these are yours forever. They're the reason “how much does a website cost” has no single answer: it isn't a purchase, it's a subscription to several things at once.

Domain: the renewal is the real price

The first year is marketing. GoDaddy will sell you a .com for a dollar or two and renew it at $21.99. Namecheap sits around $14–15. Cloudflare Registrar sells at cost — about $9.77 — and charges the same every year, which is unusual enough to be worth knowing.

Builders bundle “free domain for year one”. Year two, it renews at their rate. Whatever you register, the number that matters is the renewal, because you'll pay it every year you stay in business.

SSL: free, and you should not pay for it

The padlock in the address bar costs nothing. Let's Encrypt issues certificates free, and essentially every host and builder includes one automatically. Paid certificates exist for organisations needing identity validation; a café does not. If a quote has a line item for SSL, ask what it's for.

Email, booking, and the rest of the stack

• Business email at your own domain — roughly $6–8 per user per month on Google Workspace or Microsoft 365, so about $70–100 a year for one inbox. Note that Microsoft raised Business Basic to $7/user/month from July 2026.
• Booking or scheduling — either bundled into a higher builder tier, or a separate $16–49/month product.
• Premium themes and plugins — a one-off $19–69 for a theme, but page builders like Elementor Pro are annual ($59+/year), which surprises people.
• Payment processing — Stripe is 2.9% + 30¢ online. Square quietly raised its free-plan online rate to 3.3% + 30¢ in October 2025; getting back to 2.9% costs $49/month. Most cost guides still quote the old number.
• Stock photos — $8–10 per image if you buy them. For a local business, usually skip this: photographs of your shop taken on a phone beat stock images of a generic one.

The number nobody publishes: five years

Every cost guide stops at year one, which is where the promotional pricing lives. A business doesn't. Here's the same salon, over a realistic horizon.

The interesting part isn't which row wins. It's that the gap between the cheapest and most expensive route narrows a lot once you stop looking at the upfront number — and that every route has a floor of a few hundred dollars a year that never goes away.

The maintenance question, answered specifically

“You'll need maintenance” is usually where an upsell starts, so it's worth being precise about what the actual risk is. It is mostly a WordPress-and-plugins risk.

Patchstack's State of WordPress Security in 2026 catalogued 11,334 new vulnerabilities across the WordPress ecosystem during 2025, up 42% year on year. 91% of them were in plugins — only six were in WordPress core. Nearly half (46%) had no fix available from the developer at the moment they became public, and the median gap between public disclosure and the first observed exploitation attempt was about five hours.

That reframes the decision. If you run WordPress with a stack of plugins, ongoing maintenance isn't an upsell, it's a requirement. If you're on a hosted builder, the platform patches itself — that is a genuine and rarely-stated part of what the monthly fee buys.

What nobody can tell you honestly

We went looking for data on how many hours a small business owner actually spends building their own website, and what share never finish. We checked survey publishers, the builders themselves, and the usual industry research. That data does not appear to exist. Any article giving you a confident figure is guessing.

What does exist is suggestive. Clutch's 2025 survey of small businesses found that 41% of small business websites run on DIY builders — but that only 8% were built by the business owner personally. In-house staff built 37%, and 45% were outsourced.

Those are different questions with the same population, so read it carefully: it means most DIY-builder sites were not, in the end, built by the owner who bought the subscription. The tool gets purchased on the promise that anyone can use it, and then the work gets handed to a staff member, a relative, or a cheap freelancer. That handoff is the real cost of DIY, and nobody has measured it.

How to actually decide

• Price the tier you need, not the one advertised. Open the pricing page and find the plan that includes bookings or selling if you need them. That's your real number.
• Add the stack before comparing. Domain at renewal price, business email, booking tool. Comparing a builder's headline against an agency quote is comparing two different things.
• Multiply by five. Then ask whether the upfront difference still matters.
• Be honest about who will do the work. If the answer is “me, on a weekend”, the survey data suggests it's more likely to end up with someone else — or unfinished.
• Ask what happens when it needs changing. Not launch day: the Tuesday six months later when your hours change. Whether that's a five-minute job or a $150 invoice matters more over five years than the build price.

The short answer

For a small local business in 2026: expect $400–900 a year all-in if you build it yourself on a platform that actually supports what you need, $1,500– 5,000 upfront plus a few hundred a year for a freelancer, and $6,000+ for an agency. Whatever route you take, the recurring floor doesn't go away, and the first-year price is the least useful number in the decision.