The price is the smaller number

Cost is a poor way to judge a website.

Not because money doesn't matter. It does. But the invoice is a number you see once. The other number, what the site makes you or loses you, you see every single day.

For a business that sells online, the website isn't a line item. It's the front door. It's the shop window. It's the salesperson, the checkout and the person who answers questions at 2am. It's all of it.

So the better questions aren't "how much?" They're these. How much will it make? How much is the current one costing? How quickly will the new one pay back?

It's an investment. Treat it like one.

Why the quotes are all over the place

Ask around and you'll get quotes of $4,000, $20,000 and $100,000 for what sounds like the same job. Nobody's lying. They're quoting different things.

At one end, someone working from their bedroom with a template. At the other, a 200-person agency with a strategist for every page. There are good reasons to hire both. A side project with no traffic doesn't need the agency. A global brand with millions of visitors doesn't need the bedroom.

Most businesses sit somewhere in between, and that band is wide. One US agency's 2026 pricing guide puts a typical small business site at $15,000 to $40,000, and quotes either side of that are normal.

So the price on its own tells you very little. The scope tells you everything. What are we actually trying to deliver? What state is the current site in? What problem are we solving? A store that needs three pages fixed and a store that needs rebuilding from the stock system up are different jobs. Price them as the same and someone ends up disappointed.

The things that move the number are usually the same handful. How many different page types you need. The platform. Anything custom, like a product configurator or booking that takes payment. Whether content and the build are included. And how senior the people doing the thinking are.

Common sense does most of the work. It's also why we don't publish project prices. Any number we put on this page would be wrong for most of the people reading it.

Pay cheap, pay twice

A cheap website isn't always a mistake. A small site with little traffic? A good template is a perfectly sensible spend.

It becomes a mistake when the site has a job to do. Traffic. Ads. A product launch. That's when cheap stops being cheap.

Usually the problem isn't how the cheap site looks. It's how it's put together. What people see first. Where the price sits. How many steps it takes to buy. A prettier version of the wrong structure fails in exactly the same way.

Take client X. No names, no shame. They came to us in a low point. They'd gone live with a lower-cost build a month earlier and hadn't made a single sale.

Not one.

By the time they found us they were struggling, and fair enough. It wasn't just the month of no sales. It was the money already paid to the first provider. The weeks of effort getting it built. The work that now had to be undone. Add it up and they'd lost two, maybe three months of selling a brand-new product.

Pay cheap, pay twice. And you pay on both ends.

The happy ending: we rebuilt the site around how their customers actually buy, relaunched, and the results have been great since. But they'd have saved themselves a quarter of a year by doing it that way first.

What a bad website costs you every day

Here's the sum nobody puts on the quote.

Take a store with 20,000 visits a month and an $80 average order. At a 1.5% conversion rate, that's $24,000 a month. At 2%, it's $32,000. That half a percent is worth $8,000 a month, or $96,000 a year. (Example figures.)

Now run it backwards. If your site converts at 1.5% when it could do 2%, it's costing you around $265 a day. Every day.

Selling services instead? Same sum, different words. Say 3,000 visits a month, 2% of visitors send an inquiry, and one in five inquiries becomes a $5,000 job. That's $60,000 a month. Get inquiries to 3% and it's $90,000. (Example figures.)

Now look at the quotes again. A $30,000 site that adds $8,000 a month pays for itself in under four months. A $5,000 site that adds nothing never pays for itself. One of those is the expensive option, and it isn't the one with the bigger number.

Sometimes the honest answer is that you don't need a new site at all. If the site is solid and one or two pages are where people give up, fix those pages. That's what conversion rate optimization is for. If the whole site is holding sales back, a website redesign is the spend that pays.

What to do next

• Work out what your site earns now: visits, conversion rate and the value of a sale.

• Work out what half a percent more would be worth to you every month.

• Judge every quote against that number, not against each other.