How long does a website take to build?

The question is almost always phrased the same way: when will it be ready? And it almost always means two different things at once. One is how much time passes between the first email and the site going live. The other is how much of that is actual work. The two are rarely the same number, and the gap is not accidental: the calendar includes the waiting.

Both are worth knowing, because they answer different questions. The calendar tells you when you can plan around it. The workload tells you how to judge the price — that comes from the work, not from how many weeks went by.

What the calendar shows

A brochure site with the content ready — description, images, contact details — typically goes live in eight to ten days. That runs from the first email to publication, and it includes every pause: waiting for copy, waiting for a decision, waiting for feedback.

That number misleads in both directions. If there is always something to do, eight days is close to eight working days. If the project is waiting on material, the same eight days might hold three days of work and five of waiting. Same calendar, half the work — and the price follows the second one.

What the working days show

This breakdown is for a brochure site. The numbers come from my own projects, and they give a range rather than a point — the upper end is where there are more pages or more complex content.

• Scoping and questions — half a day to one day. The shortest stage, and the one that determines every other. It is not about which colour you like; it is about who will come to the site, what they are looking for, and the one thing they need to do there.
• Structure and visual direction — one to two days. Which pages exist, what goes where, how it looks. No code yet, only decisions — and these are cheaper to make now than to rewrite later.
• Building the interface — two to three days. This is the part that shows on the finished site — roughly a third of the total.
• Putting the content in — one day. Text, images, cropping, sizing, compression. It takes this long even when the material is ready: the images still have to be prepared.
• What nobody sees — one to two days. Testing across browsers and screen sizes, load speed, keyboard and screen reader use, the layer written for search engines, backups, and checking that the contact form actually arrives.
• Launch and the day after — half a day to one day. Domain, certificate, redirects — then the problems that only appear in production. This stage is always needed, and it is always the one missing from estimates.

Added up: six to ten working days for a brochure site.

More than a third of the work is invisible

Split that list by what can be pointed at on the finished site: scoping, testing and launch show nothing. There is no screenshot of them, nothing to point to and say “this is what came of it”. Together they account for more than a third of the effort.

None of it is optional. Skip the scoping and the wrong structure has to be rebuilt later. Skip the testing and the client finds the fault — on their own phone, in front of a customer.

A BROCHURE SITE — EIGHT WORKING DAYS

Scoping ¾ day · Structure and direction 1½ days · Building the interface 2½ days · Putting the content in 1 day · Testing 1½ days · Launch ¾ day

What can be shown

The highlighted bars are what can be shown on the finished site. The three fainter ones together are more than a third of the effort — and none of them produce a screenshot.

The calendar

The calendar runs longer than this: it includes the waiting.

Where it takes longer

With an editing interface: four to six days more. If you will be maintaining the content yourself — a price list, staff, opening hours — it is not enough to type it in once: there has to be a surface where it can be kept up to date. That is separate work, and the visitor never sees it. You do.

With connected systems: eight to fifteen days more. Payment, access, invoicing, notification. Half the development goes into making the chain hold when one link fails — which is the part with nothing to show, and the part without which the system is merely functional, not dependable.

What stretches the calendar

Three things, in order of how often they come up.

The content. Most common by far, and almost always the copy. Images can arrive later; text cannot, because it determines the structure. When the material is there on day one, calendar time and working time nearly coincide.

The decision. If several people weigh in, decisions take longer — which is fine, it just has to be planned for. One round of feedback costs two days; three rounds cost a week.

The third party. Photographer, accountant, previous developer, hosting provider. They do not work to my schedule, and waiting for access is measured in days.

Why an exact number is not possible upfront

Because scoping is itself discovery. The first conversation is where it emerges that there is an existing system to connect to; that the services follow a logic which will not fit into three pages; or the other way round, that half of what was asked for is not actually needed.

So my estimate gives a range and names what moves it. That is worth more than a precise number that will not hold: a single figure tells you nothing about what you can do to land at the shorter end.

Questions on this topic

← Back to Writing