How long does it take to build a website? (2026 guide)

Most websites take somewhere between a few weeks and several months to build. In our experience a small brochure site of five to ten pages, built on an existing template with content you have already written, is usually live within a month, because the discovery and design phases described below compress into a few days or disappear altogether when the design already exists. A custom-designed marketing site for a mid-sized company tends to take a couple of months, and a standard e-commerce catalogue a little longer again. Once a site carries logins, payments, bookings or a customer portal it behaves far more like a web application than a website, and the honest planning horizon stretches to several months or longer.

Where your own project sits in that spread depends less on the technology than on how quickly decisions get made on your side. In our experience the most common cause of a website running well past its quoted date is not development, it is content and approvals.

Typical timelines by type of site

The table below is not industry data. It is the planning shape we use ourselves, based on the projects we have delivered, and we have deliberately kept it qualitative because a single week either way says nothing useful until we have seen your scope.

Type of site	What we would usually plan for	What drives the time
Template site, 5-10 pages	Weeks rather than months	Content readiness, one round of amends
Custom-designed brochure site, 10-25 pages	A couple of months	Design rounds, copywriting, photography
E-commerce, standard catalogue	A little longer than a brochure site	Product data, payment and shipping setup
E-commerce with custom logic or ERP links	Several months	Integrations, stock and pricing rules
Web portal or customer login area (web application territory)	Several months	Roles, permissions, data security, testing
Full web application	Most of a year is not unusual	Scope, integrations, compliance

The last two rows have stopped being websites in any meaningful sense, and we plan them as software projects. These are working timelines, not sales timelines. They assume a normal working week, a client who can review within a few days, and a scope that is agreed before build starts.

What actually happens week by week

The phase lengths below are the ones we plan with on our own projects. They are not benchmarks drawn from any published study, and they move about depending on scope. They describe a custom build: on a small template site, discovery and design shrink to a matter of days or drop out of the schedule entirely, which is how those sites reach launch inside a month.

Discovery and definition: 1 to 3 weeks

This is where you decide what the site has to do, who it is for, what a visitor should do next, and how success will be measured. On smaller sites it is a couple of workshops. On anything with functionality behind it, it produces a written specification. We have covered what that document should contain in our guide to software requirement specifications.

Skipping discovery does not save time. It moves the time to a worse place, usually to the point where something has been built and has to be unbuilt.

Design: 2 to 5 weeks

We produce wireframes first, then visual design for the key page templates. You do not need a design for every page: you need designs for every type of page. A twenty-page site might only need six templates.

Two rounds of amends is normal. A third is usually a sign that the brief was not clear enough, and it is worth stopping to fix that rather than iterating visually. If you want more detail on this stage, our UI and UX design service page explains how we structure it.

Build: 3 to 12 weeks

For a brochure site, the build is often the shortest phase. Templates get created, the content management system is configured so your team can edit pages without going back to a developer, and integrations such as forms, analytics and CRM are connected.

For anything with user accounts, the build phase expands considerably, because every screen now has states: logged in, logged out, permission denied, empty, loading, error. That is where the weeks go.

Content: runs alongside, and usually late

This phase covers copy, images, product descriptions, staff biographies and legal pages. It is the phase we find clients underestimate most often. If you have twenty-five pages of new copy to write and one marketing manager writing it around a day job, that is not a two-week task.

Our honest advice: start writing content in week one of discovery, not week one of build.

Testing and launch: 1 to 3 weeks

Cross-browser and cross-device checks, accessibility, form testing, performance, redirects from old URLs, analytics verification, and a soft launch window. Redirect mapping matters more than people expect. A migration that drops old URLs can lose search visibility that took years to build, and recovering it takes months.

Why two sites of the same size take different lengths of time

The number of decision-makers

One decision-maker with authority is fast. A committee of six with no agreed tiebreaker is slow, and no amount of developer capacity fixes it. Before you start, decide who signs off design and who signs off content, and give them a stated turnaround time.

Integrations

A site that stands alone is quick. A site that has to talk to your accounting system, your stock system, your CRM and a third-party booking engine is not. Each integration brings its own documentation, its own rate limits, its own sandbox and its own failure modes. We budget time for every meaningful integration as a piece of work in its own right, and considerably more of it if the other system is old or poorly documented.

Whether the requirements hold still

Scope changes are legitimate. Businesses learn things mid-project. What causes damage is unacknowledged scope change: features added without adjusting the date or the budget, until the plan quietly stops being true. We prefer to reprice and re-plan openly, which is part of how we work.

Whether you are replacing something

Rebuilds carry hidden work: exporting old content, mapping URLs, migrating user accounts, matching functionality nobody documented. A replacement site is almost always slower than the equivalent new build.

Can you build a website faster?

Yes, within limits. These are the levers that genuinely work.

• Reduce page count for launch. Ship the twelve pages that earn business and add the rest afterwards. A live site earning enquiries beats a perfect one in staging.
• Have content ready before design sign-off. Designing around real copy is faster than designing around placeholder text and then redesigning.
• Use an existing design system or theme. You lose some distinctiveness, and in return you take a whole design phase largely out of the schedule.
• Assign one empowered reviewer. In our experience, consolidated feedback within 48 hours removes more calendar time than most technical decisions.
• Fix the scope in writing. Then handle additions as a second phase rather than an expansion of the first.

The lever that does not work is adding developers to a late project. On a small website, two developers are not twice as fast as one, and the coordination cost can make things worse.

What about AI website builders?

AI tools can produce a working-looking site in an afternoon, and for a simple landing page that may be all you need. The difficulty appears when the site has to handle real data, real logins or real payments, because the generated code often has no tests, no access control and no clear structure to extend. We now see a steady stream of businesses who moved quickly and then needed the result made production-ready, which is why we offer vibe code rescue as a service. Speed at the prototype stage is real. Speed to a dependable, maintainable site is a different question.

How long before a new website performs?

Building it is not the same as it working. Allow a good few months after launch for search visibility to settle, particularly on a rebuild, and plan a round of changes based on real behaviour once you have data. A website is better treated as something you keep improving than a project that finishes.

Frequently Asked Questions

How long does it take to build a simple website?

In our experience a simple brochure website of five to ten pages is usually live within a month of kick-off, provided you are using an existing template or design system and your copy and images are ready before the build begins. If content still has to be written and photographed, expect that month to stretch, sometimes considerably. The build itself is rarely the constraint on a small site; content and approval turnaround usually determine the date.

How long does it take to build a website from scratch?

A fully custom website built from scratch, with original design, bespoke page templates and a content management system your team can use, generally takes two to four months in our experience. Discovery is usually one of the shorter phases, design and build take the bulk of the calendar between them, and testing and launch close the project out, with some of those phases overlapping rather than running strictly end to end. That shape is how we plan our own projects rather than a published industry standard. Complex integrations, e-commerce or user accounts push it well beyond that.

Why do website projects overrun?

The causes we see most often are content arriving late, feedback being slow or contradictory, and scope growing without the timeline being adjusted. We find technical problems cause fewer overruns than clients expect. You can protect against all three by writing content before design sign-off, naming one person who consolidates and approves feedback within 48 hours, and agreeing in writing that new requirements are handled as a second phase.

Is a website faster to build than a web application?

A website is considerably faster to build. A brochure website mainly presents information, so most pages behave the same way. A web application has to handle user accounts, permissions, data validation, error states and security, and each of those multiplies the work of testing. A typical marketing website takes weeks to a couple of months, whilst a comparable web application takes many months. Our guide to how long a typical web app takes covers that path in more detail.

How much does a bespoke web build cost in the UK?

Cost tracks scope more closely than it tracks page count. UK agency rates typically run between £75 and £150 per hour. For bespoke work, a simple application sits in the £10,000 to £50,000 range, a mid-range build between £50,000 and £150,000, and a complex platform from £150,000 to £1,000,000 or more. A straightforward brochure website, which involves no bespoke application work, costs a small fraction of even the lowest of those figures, because it is a far smaller piece of design and build work. Our cost estimator will give you a range for your own scope.

What should I prepare before a project starts?

Have three things ready: a clear statement of what the site must achieve commercially, a list of the systems it needs to connect to, and as much final copy and imagery as you can gather. Also decide who approves design and who approves content, and agree a turnaround time with them. In our experience, projects that begin with those settled tend to run far more smoothly than projects that do not.

Getting a realistic date for your project

If you want a timeline you can actually plan around rather than a best-case figure, the fastest route is a short conversation about scope. We will tell you where the risk sits, including the parts that depend on you rather than on us.

Have a look at our web development and web application development services, try the cost estimator for a starting range, or get in touch and we will give you an honest estimate.