What a website should cost, and why nobody will tell you

Almost no agency publishes pricing. The stated reason is that every project is different. The real reason is that quoting after a discovery call lets the number be shaped by what the buyer seems able to pay. We publish ours, which makes this article slightly awkward to write and considerably more useful to read.

The bands, honestly

Under five thousand dollars, you are buying a template implementation. That is not an insult, for a business that needs to exist online and convert a modest amount of traffic, a well-executed template is often the correct decision, and spending more would be waste.

Between five and fifteen thousand, you are buying custom design and custom front-end build against your actual content, with performance and accessibility treated as requirements rather than as a nice-to-have. This is where most small businesses with a real revenue dependency on their site should be.

Above fifteen thousand, you are usually buying integration rather than pages: a CMS your marketing team can operate without a developer, a design system, a booking or quoting flow wired into your CRM, or a catalogue that has to stay in sync with something else.

What actually drives the number

Not page count, which is what most quotes are built on. The cost drivers are the number of genuinely distinct templates, how many external systems have to be integrated, whether the content already exists, and whether anyone has decided what the site is for.

That last one is the most expensive and the least visible. A project where the client cannot agree internally on the primary action a visitor should take will overrun regardless of the technology, and no amount of budget fixes it. Deciding is cheaper than building twice.

Four questions that expose a bad quote

Ask who owns the code and the domain when the engagement ends. If the answer is anything other than 'you do, unconditionally', stop. A site you cannot take with you is a subscription with extra steps.

Ask what the performance target is, in numbers. A serious answer sounds like 'Largest Contentful Paint under 2.5 seconds on a mid-range Android on 4G'. An unserious answer is 'we build fast sites'. One of those can be verified on launch day and the other cannot.

Ask what happens when you want to change a page. If every text edit is a support ticket, the cheap build is not cheap, you will pay for it monthly, forever, in small increments that never get compared against the original saving.

Ask what is explicitly not included. Content writing, photography, and the migration of existing pages are the three that most often sit outside a quote and then arrive as change requests at the least convenient moment.

The number nobody quotes

The cost of the site is rarely the real cost. A five-thousand-dollar site that loads in six seconds and loses half its paid traffic before anything renders is far more expensive than a fifteen-thousand-dollar site that does not, the difference is just charged to the ad account instead, where nobody attributes it to the build.

That is the actual calculation. Not what the site costs, but what it costs you every month for as long as it is live.