"How long will this take?" is the second question every client asks, right after price, and it is the one most proposals answer with a single confident number that turns out to be wrong — not because anyone lied, but because a website project runs on two clocks at once: the production clock (discovery, design, build, QA, launch) and the approval clock (how fast anyone actually reviews and signs off on what gets sent to them). The first one is genuinely predictable. The second is where almost every real delay comes from, and it rarely gets named as the cause. Click through each phase below for what happens in it, what usually causes delay, and the milestones that mark real progress.
← All resources
— Timeline
How long a website project actually takes, week by week.
The phases alone add up to six to fourteen weeks for a multi-page business site — but the calendar rarely matches that, and the gap is almost never the build. Here is the real sequence, and where projects actually lose time.
From kickoff to launch, phase by phase
Week 0 is the kickoff call. Select a phase to see what happens in it, what usually causes delay, and the milestones that fall inside it.
Weeks 0–2
Discovery & scoping
Nothing gets built yet. This is where scope, content ownership and the platform decision get fixed in writing — and it is the only phase where a vague decision is nearly free to fix. Once design starts, the same fix costs days instead of minutes.
What happens
- A kickoff call fixes what the site needs to do, who owns which piece of content, and which pages exist before anyone opens a design tool.
- Sitemap and page-level scope get written down and signed off — this one document is what every later estimate, and every later "is this in scope" argument, refers back to.
- The platform decision gets made here, not discovered mid-build. Changing it later does not extend the timeline — it restarts a chunk of it.
- You get asked for content — copy, images, brand assets — in week one, not week six. Most projects that later feel "behind" were on schedule until this exact request sat unanswered.
What causes delay
- Vague scope. "A modern, clean design" is not a scope, and an agency that does not push back on it here is deferring the disagreement to the design round, where it costs more.
- Content ownership left unassigned. If nobody is named as responsible for copy, the default owner becomes whoever eventually gets annoyed enough to write it — which is not a schedule.
- A platform decision made by committee after work has already started elsewhere.
Milestones
- W0Kickoff call, contract and scope signed
- W1Sitemap and page list approved
- W2Content requirements sent to client
Phase-duration ranges are drawn from published web-agency estimates (Designveloper, Ramotion); the weekly structure and milestones themselves are our own, from client projects — treat the weeks as a realistic band for a multi-page business site, not a formula. A one-page site can clear all four phases in two to three weeks; a custom platform with heavy integrations routinely runs six months or more. Approval-delay figures are from a 2025 survey of 500 marketing and creative professionals.
Why the phases add up to less than the calendar shows
Add the phase ranges together — discovery, design, development, QA and launch — and you get somewhere between six and fourteen weeks of actual work for a multi-page business site. That number is genuinely accurate, and it is also not the number most projects finish on, for a reason that has nothing to do with how fast anyone works.
A website project runs on two clocks at once. One measures the work: designing a page, building a template, testing a form. The other measures the gaps between the work — the days a design sits in someone’s inbox before it gets opened, the week a content deadline passes quietly, the "let’s regroup on this" that takes longer to schedule than the meeting itself. The first clock is what a proposal quotes. The second is what actually determines the finish date.
This is not a uniquely web-industry problem. A 2025 survey of 500 marketing and creative professionals found that more than 60% lose up to a full day of their work week specifically chasing approvals, and 74% said the approval process takes more effort than the creative work under review. That is the second clock, measured directly.
Where the two clocks diverge, phase by phase
Discovery is the one phase where the two clocks mostly agree — there is very little to approve yet, just decisions to make once. That is exactly why it is worth treating scope, content ownership and the platform decision as seriously as this guide does: a fix here costs a conversation. The same fix found in month three of development costs a rebuild.
Design is where the gap usually opens. Published estimates put the work itself at two to six weeks — but that range assumes feedback returns in days, not the following sprint. A Strategy Session to nail scope before design starts is cheap insurance against exactly this gap, compared with diving straight into visuals and hoping.
Development is where the gap is more often content than feedback. A build can be code-complete with zero copy written, and "code-complete" is not a milestone a client can see — which means it does not feel like progress even when it is. A Business Website engagement scopes content responsibility in writing at kickoff for exactly this reason.
Launch is where the gap is shortest in theory and longest in practice, because a "final review" has a way of becoming one more design round at the exact moment everyone is least prepared to absorb a delay. Protecting that boundary is worth more than any technical launch-day task.
What actually shortens a timeline, and what only feels like it does
Fixed scope and capped revisions shorten timelines. Not because they prevent disagreement, but because they move every disagreement to a moment when it is cheap to resolve — the scoping call — instead of a moment when it is expensive — the fourth unplanned design round.
A single named approver per phase shortens timelines more reliably than anything about the build itself. A Technical Audit before a redesign has the same effect on the development phase: it turns "we will figure out the integrations as we go" into a list decided before anyone starts billing hours against discovering it.
What does not help: adding people to review a decision "just to be safe." More reviewers produce more opinions, not more clarity, and reconciling three conflicting opinions takes measurably longer than accepting one person’s judgment call.
What also does not help: paying for a faster build. Compressing the coding time does nothing to the approval gaps, which is where the real time goes — it is the same mistake as expecting a bigger budget to speed up how fast Google indexes a page.
How we scope this in practice
A Business Website engagement starts with the scoping conversation this guide describes as phase one, in writing, before design work begins — because the fix is nearly free here and expensive everywhere after.
If scope is the open question rather than the design itself, a Strategy Session is the cheaper way to answer it: a short, paid engagement that produces the sitemap, the platform decision and the content plan before a design tool gets opened.
After launch, the stabilization window is where a site either gets left alone until something breaks, or gets a proper first month of attention — content updates, monitoring, the small fixes real traffic surfaces that staging never did. That is what a Monthly Maintenance retainer is scoped for, rather than sold as an afterthought.
Sources
The phase-duration ranges and approval-delay figures above come from these, checked August 2026. The weekly structure and milestones are our own, from client projects.
- Designveloper — Web Development Timeline: How Long Does It Take to Build a Website? ↗
Phase-by-phase estimate: planning 1–2 weeks, UI/UX design 2–4 weeks, development 4–12 weeks, testing/QA 1–3 weeks, deployment 1 week — and identifies client feedback speed as one of the biggest timeline drivers.
- Ramotion — Website Development Timeline: Stages, Timeline & Breakdown ↗
A second, independent phase breakdown (planning 1–2 weeks, design 3–6 weeks, build 4–8 weeks, testing/launch 2–4 weeks) plus typical total duration by project type, from a one-page site (1–2 weeks) to a custom platform (6+ months).
- StreamWork — 2025 Approval Research Study ↗
Survey of 500 U.S. marketing and creative professionals conducted by Pollfish, October 7–22, 2025: over 60% lose up to a full workday per week chasing approvals, 74% say the approval process takes more effort than the creative work itself, and over three-quarters said approval delays directly wasted budget.
— FAQ
Frequently asked questions
Not sure why your project has stalled?
Tell us which phase you are in and what is outstanding, and we will tell you honestly whether it is a build problem or an approval one.