How much does it cost to build a website? The honest answer

Ask what a website costs and you will get answers ranging from nothing to fifty thousand, which is useless when you are trying to budget. The honest reason is that “a website” can mean a one-page site you build yourself in an afternoon or a custom platform a team builds over months, and those are not the same product any more than a bicycle and a car are the same vehicle.

I build websites and tools, increasingly with AI, and I am not selling you a plan or an agency retainer, so I can give you the version with no agenda. This guide lays out the real price ranges for every route, what actually drives the cost up or down, the ongoing fees most articles conveniently skip, and how to spend the right amount for your situation rather than the most.

Why there is no single price (and that is the point)

The reason nobody can answer the question with one number is that the same word covers wildly different things. A website might be a simple presence, a few pages that tell people who you are and how to contact you. It might be a content site with a blog, or a full online store taking payments, or a custom web application with logins and databases. Each of those is a different amount of work, so each is a different price, and lumping them together is what makes the answers feel random.

The other half of the answer is who builds it. The exact same five-page small-business site costs almost nothing if you build it yourself on a builder, a few thousand if a freelancer does it, and five figures if an agency does. You are not paying for different websites, you are paying for different amounts of someone else’s time and expertise. Once you separate those two questions, what kind of site, and who builds it, the price stops being a mystery and becomes a choice.

So before you can budget, you have to know which route you are on. The interactive below lays the routes out side by side, with the real numbers behind each one.

What a website really costs, by route

Here are the five realistic routes, from cheapest to most expensive. The price is hidden on each one, tap a card to reveal what it actually costs, how long it takes, and who it is for. Reveal all five and you will have a clearer picture of the market than most quotes will give you.

The jump that surprises people is between the freelancer and the agency, because the website itself is often similar. What changes is everything around it, the strategy, the project management, the team, the accountability, the support afterwards. That can be worth it for a large business, and pure waste for a small one. The skill in buying a website is matching the route to what you actually need, not to what sounds most professional.

What actually drives the price up or down

Within any route, a handful of factors move the number more than anything else. Understanding them lets you see exactly what you are paying for, and what you can cut.

The biggest is complexity, meaning what the site has to do. A handful of informational pages is cheap. Add a blog and it grows a little. Add an online store with payments, inventory, and accounts and it grows a lot, because each of those is real functionality that has to be built and tested. Be honest about what you truly need on day one, because paying for an online store you will not use for a year is money frozen.

Design is the next lever. A clean site built on a good template costs a fraction of a fully custom design drawn from scratch, and for most businesses the template-based result is more than good enough. Custom design is worth paying for when your brand genuinely depends on standing out visually, and a needless luxury when it does not.

Then there is content, the words and images. Someone has to write the copy and source the photos, and if you do not do it yourself, you pay someone who will. This is the cost people forget to budget for and then scramble over at the end. Finally there is the number of pages and any special integrations, booking systems, CRMs, payment flows, each of which adds work. None of this is mysterious once you see it as a list of decisions, each with a price attached.

What an online store and special features add

The single biggest jump in price comes from turning a website into something that does a job beyond informing, and an online store is the clearest example. Selling products online means payments, a product catalogue, a checkout, order handling, and often accounts and inventory, and each of those is real machinery that has to be built, connected, and tested. That is why an ecommerce site reliably costs more than a brochure site on the same route, sometimes several times more.

The good news is that the cheap routes have caught up here too. A modern builder or an AI-assisted setup can give you a perfectly capable small store for a low monthly cost, with the payments and checkout handled for you, which a few years ago would have meant a custom build. You only really need the expensive custom route when your store does something unusual, complex inventory rules, subscriptions, integrations with other systems, or scale that the standard platforms cannot handle.

The same logic applies to any special feature, a booking system, a members area, a multi-language setup, a connection to your CRM or accounting. Each one is a discrete piece of work with its own price, and each is worth paying for only if it earns its keep. The discipline is to start with what you need to open for business and add features when there is a real reason, not to commission every possible feature upfront because it might be useful one day. Functionality you are not using is just frozen money that also makes the site slower to build and harder to maintain.

The ongoing costs nobody mentions

Here is where the headline price misleads people most. A website is not a one-time purchase, it has running costs, and ignoring them is how a “cheap” site turns expensive. There are three you will always meet.

The first is the domain name, your address on the web, which is a small annual fee, usually the price of a couple of coffees per year. The second is hosting, the service that actually keeps your site online. On builder and AI routes this is bundled into the subscription, on a custom or freelance site it is a separate monthly or yearly cost that varies with how much traffic and power you need. The third, and the one most underestimated, is maintenance, keeping the site updated, secure, and working. On a simple builder site this is essentially nothing, the platform handles it. On a custom site someone has to do it, either you or someone you pay, and a managed maintenance arrangement can run from a small monthly fee to a serious one for a complex site.

The honest way to budget is to think in two numbers, the upfront cost to build it and the yearly cost to keep it alive. A site that is cheap to build but expensive to maintain can cost more over three years than one that was pricier upfront and runs almost free. Ask anyone quoting you to spell out both, and be suspicious of a quote that only mentions the first.

Do it yourself, or hire someone?

This is the decision under all the prices, and it comes down to a simple trade, money against time and control. Building it yourself, on a builder or with AI, costs the least in money and the most in your own hours, and the result is as good as your patience and taste. Hiring someone costs more money and almost none of your time, and you get the benefit of their skill, though you give up some control and have to communicate what you want clearly.

The honest way to choose is to value your own time. If you have more time than money, and your site is fairly simple, building it yourself is often the right call, and modern tools have made the result genuinely respectable rather than the clunky DIY sites of a decade ago. If your time is worth more spent on your actual business, or the site needs to be genuinely custom, paying a freelancer or agency buys back those hours and usually produces a better result for anything beyond the basics.

There is also a middle path that more people should consider, build the first version yourself to get online cheaply and learn what you actually need, then hire someone later to level it up once the business can justify it and you know exactly what to ask for. That order often saves money twice, you avoid overpaying at the start when your needs are vague, and you give a professional a precise brief later instead of a guess. The worst outcome is paying a lot upfront for someone to build the wrong thing because neither of you yet knew what the business needed.

How AI changed what a website costs

This is the genuinely new part, and it is why some of the numbers above are lower than they would have been a couple of years ago. AI has collapsed the cost and time of the parts that used to be expensive, generating a layout, writing the first draft of the copy, producing images, even scaffolding the code. Work that used to mean a brief, a quote, and weeks of back and forth can now produce a clean, working first version in an afternoon.

What this means in practice is that the floor has dropped. A capable person can now build, with AI help, a site that would have cost a few thousand from a freelancer just a short time ago, and do it in a fraction of the time. I build sites and small tools this way myself, and it has changed what is reasonable to pay, especially for a small business that does not need a custom platform. The expensive routes still exist and still make sense for complex needs, but the cheap end has become far more capable.

What AI has not changed is the judgment. It will happily generate a site that looks fine and says nothing, or that misses the one thing your customer needs to see. The value now sits in knowing what a good site should do and steering the tools toward it, which is the same point I make in my guide to using AI in business. The tools got cheap. Knowing what to build with them did not, and that is where the real money is either saved or wasted.

How much should you actually spend?

The right number is not the biggest you can afford, it is the smallest that does the job for where your business is now. A brand-new business testing an idea has no reason to spend thousands, a clean builder or AI-built site that looks credible and explains the offer is plenty, and the money is better kept for actually getting customers. You can always upgrade once the business has proven itself.

An established small business that relies on its website to bring in real work can justify more, a freelancer-built custom site in the low thousands is often the sweet spot, professional and tailored without agency overhead. The agency and custom routes earn their cost only when the website is genuinely central to a larger operation, complex functionality, many stakeholders, real traffic, where the strategy and support around the build matter as much as the build itself.

The mistake in both directions is real. Underspending on a site your business depends on costs you customers who judged you by a weak first impression. Overspending on a site your business does not yet need freezes money you could have used to grow. Match the spend to the stage, and revisit it as you grow rather than trying to buy the final version on day one.

What about redesigning an existing site?

A lot of people are not building from scratch, they already have a site and it feels dated or is not bringing in work. A redesign sits on the same routes and the same price logic as a new build, but with one useful difference, you usually have content and a clearer idea of what you need, which can make it faster and cheaper than starting cold.

Before paying for a full redesign, it is worth asking whether you need one. Sometimes the problem is not the design at all but the content, the site looks fine and simply does not say the right things, in which case rewriting the copy costs far less than rebuilding. Sometimes a light refresh, updated images, a cleaner homepage, a faster load, fixes the dated feeling without a ground-up rebuild. A full redesign earns its cost when the site is genuinely broken on mobile, painfully slow, impossible to update, or so off-brand that it actively loses you customers.

When a redesign is justified, the AI-assisted route is especially strong, because rebuilding a known site with content you already have is exactly the kind of work modern tools do fast. The same caution applies as with a new build, get the redesign matched to what is actually wrong rather than rebuilding everything by reflex, and separate the one-off redesign cost from the ongoing running costs so you know the true picture over the next few years.

How not to overpay

A few simple habits protect you from paying too much. Get more than one quote, because prices for the same work vary enormously and a single quote gives you no reference. Be precise about what you need, a vague brief invites a padded quote, while a clear list of pages and features gets you a tighter one. Ask every quote to separate the upfront cost from the ongoing costs, so you are comparing like with like and not getting surprised later.

Be wary of two opposite traps. The suspiciously cheap offer often hides its real cost in the maintenance, the add-ons, or the quality, and the prestigious agency quote often bundles in overhead and process you simply do not need. And before you pay anyone for a simple site, genuinely consider whether a modern builder or an AI-assisted build gets you most of the way for a fraction of the cost. For many small businesses in 2026, it does, and the honest professional will tell you so rather than sell you something larger.

If you want a hand working out which route actually fits your situation, or you want a site built sensibly with modern tools rather than overbuilt, that is the kind of work I do. You can see how I work on my site, and the broader thinking behind it in my guide to digital for entrepreneurs.