How Long a Business Website Takes to Build, and What Slows It Down

"How long will it take?" is the second question on every call, right after the price, and the honest answer is shorter than people expect and longer than they'd like. The code is the short part. A five-page site is a couple of weeks of actual building. The project around it is six to ten, and the gap between those two numbers is what this article is about.

Everything below is our reading of how projects run in 2026, for businesses in the GTA and the ones we work with elsewhere in Canada and the US. Other shops will draw the lines a little differently. Nobody honest will draw them very differently.

Realistic timelines, by what you're building

A template site can go faster than two weeks if the words and photos exist on day one, and almost never does, because they almost never do. A custom site can land in five weeks with a decisive client and can drift to sixteen with an indecisive one, and the build hours are identical in both cases.

The redesign row is the one people underestimate. If your current site ranks for anything, the new one has to carry every old address to its new home, and that mapping is real work with real testing. It's also the work that decides whether the launch costs you your rankings, which is covered properly in how to redesign without losing them.

The phases, and who each one is waiting on

A website project has five phases, and in each one somebody is waiting on somebody else. Knowing who is the whole trick to keeping it moving.

Discovery takes about a week and waits on you. What the business does, who the customers are, what the site has to make them do, what exists already (brand files, copy, photos, the old site's analytics), and access to the accounts. A shop that skips this is guessing about your business, and the guesses come back as revision rounds later.

Design takes one to three weeks and waits on feedback. Wireframes first, so the argument about what goes where happens on boxes rather than on finished pages, then the real design in your brand. Each round needs a decision from your side within a few days. A week of silence here is a week added to the end, exactly.

Content runs alongside design and is the biggest single variable in the table above. If you're writing it, this phase is however long it takes you to write eight pages about your own business, which is longer than anyone thinks. If we're writing it, it's two to three weeks including your review, and it's usually the better outcome because copy written to fit a layout reads like it.

Build takes two to four weeks for a custom site and waits on the builder. This is the only phase where the timeline is properly ours. We show progress weekly on a live staging address so you're looking at real pages rather than a status update.

QA and launch takes about a week and waits on two things you might not expect: access to the domain's DNS, and a final sign-off from whoever has to give it. Both are covered below because both have a habit of appearing on the last day.

The delays that stretch every project

Almost nothing that runs a website project long is technical. In our experience it's one of these five, and usually two of them at once.

The content. "Client to supply all copy and images" is the most common sentence in a web quote and the most common reason a project stalls for months. Writing about your own business is genuinely hard. You're too close to it, every sentence feels either obvious or boastful, and it's the thing that gets pushed to next week because a customer called. Decide before signing who writes it. If it's you, block the time in the calendar the way you'd block a meeting.

The photography. A template with your real photos in it looks like your business. A template with stock photos looks like a template. Most businesses don't have usable images, find out at the design stage, and then spend three weeks booking a photographer. If you don't have them, say so at discovery and book the shoot in week one, so the photos arrive when the build does.

The approvals. One decision-maker and a project moves. Three decision-makers who each see the design at a different meeting and a project stops. The fix is to name one person who says yes, agree that they'll answer within three business days, and treat that as a deadline on your side as real as the launch date is on ours.

The domain and DNS access. This one is almost comic in how often it appears. The domain was registered in 2014 by an employee who left, or by the previous web company, or on a personal email nobody checks any more. Nobody knows the login. Recovering it takes anywhere from a day to several weeks of back and forth with a registrar's support desk, and none of it can start until somebody looks.

The third-party pieces. The booking system, the CRM, the payment provider, the review widget. Each one has its own account, its own approval process, and its own documentation quality. Payment providers in particular can take a week or two to verify a business before anything works, and the verification can't be rushed from our side.

Launch is one date. Ranking is a different one

The site going live is the end of the build. It isn't the end of the wait, and the second wait is the one nobody quotes.

A brand-new domain has no history with Google. Indexing usually happens within days once the sitemap is submitted, and ranking for anything competitive takes months, because a site earns its position by being crawled, linked and clicked over time, and none of that can happen before it exists. Three to six months before a new site is bringing in search traffic worth measuring is a fair expectation for most local businesses, and longer in a crowded category.

An existing domain with an established site is different, and the difference cuts both ways. The history carries across if the redirects are right, so a redesign that maps every old URL to its new home should hold its rankings after a two or three week wobble while Google recrawls. A redesign that doesn't map them loses history that took years to build, and that's a drop that doesn't recover on its own.

There's a third timeline underneath both of those now, which is when the AI assistants start reading you. Most AI crawlers don't run JavaScript, so a site whose content only appears after a script runs is invisible to them regardless of how long it's been live. Server-rendered pages are readable from the first crawl. Which platform you build on decides that, and the comparison lives in Next.js or WordPress rather than being repeated here.

So when somebody promises the site will "be ranking by launch", they're describing something that doesn't happen. Launch starts the clock. The build decides how fast it runs.

How to shorten it, honestly

Some of the timeline is compressible and some of it is a fiction to compress.

The real gains are all on the waiting side. Start the content before the design, so the layout is built around words that exist rather than placeholder text that gets replaced with something twice the length. Book the photographer in week one. Name the single person who approves. Find the domain login. Get the third-party accounts created and verified while the wireframes are still being argued over. Every one of those turns a serial delay into a parallel one, and together they're the difference between six weeks and twelve.

Launching smaller is the other honest lever. A site with eight finished pages beats a site with fifteen pages of which seven are half-written and holding the launch. Go live with the eight, add the rest as they're ready, and get the domain earning history while you write. The pages you add later are a small piece of work each rather than one large one.

What doesn't shorten it is a rush fee. Rush pricing adds 50% to 100% to a quote in the GTA, and what it buys is priority in the queue, which is real, rather than fewer hours of work, which it can't. If the delay was content or approvals, a rush fee buys nothing at all, because the shop was never the bottleneck.

Skipping QA doesn't shorten it either. It moves the week from before launch to the week after, when it's called "fixing the site" and happens in public. And skipping the redirect mapping on a redesign shortens the project by a week and costs the rankings it was supposed to protect, which is the worst trade on this page.

What our timelines look like

Bounded work runs as a sprint: four to six weeks, fixed scope and fixed price agreed before anything starts, built by the person who scoped it. A landing page for a campaign, a technical audit and its fixes, a speed pass. What doesn't fit a sprint is a full website, and we'd rather say that here than on the call.

Full sites, stores and apps are quoted individually, and the timeline comes back with the number, within 24 hours, with the phases above marked against real dates and a plain statement of what we need from you and when. The sites on the projects page are all hand-built in Next.js, which is the platform half of the ranking-after-launch question above.

If you're planning a build and want to know whether your dates are realistic before you've committed to anyone, that's a short conversation, and the answer is the same whether or not we end up building it. The full scope of what web design and development covers, including the redesign and migration work, is on the service page.