How long a website really takes, and what makes it late

Most website projects have a date attached to something real. A trade show stand that will carry the URL. A tender submission where a stranger will type your company name into Google before deciding whether to open your document. A rebrand announcement, a new financial year, a plant opening.

The date is usually achievable. What goes wrong is that everyone assumes the risk to it sits with the studio, and it almost never does.

How long a website takes, honestly

Most of our projects run six to ten weeks from kickoff to launch. A refresh of a site that already works, where the copy and photography exist and one person signs off, can be quicker. An industrial or technical build where certifications, project records and specifications have to be assembled from scratch takes longer, sometimes considerably.

That range is not a hedge. It is the honest shape of the work, and the variable inside it is almost never the design. It is how fast material and decisions come back from your side.

We publish it that way on our pricing page rather than quoting a fixed number, for the same reason we publish price as bands. A fixed timeline either pads every straightforward project or quietly breaks on every complicated one.

Where the time actually goes

Every project runs through four phases. The shape of them matters more than the total, because each one has a different owner.

Discussion. Discovery, brief alignment, scope confirmation. This ends with a written scope and a page structure that both sides have signed off. Elapsed time here is short if you know what the site has to do and long if the answer is still being argued internally. Most of the work in this phase is yours, not ours.

Design and build. Structure first, then design, then the build itself. This is the phase people picture when they think about a website project, and it is the most predictable one, because it is entirely ours and it is the part we have done a hundred times. It is also, for that reason, the phase least likely to be the reason you are late.

Feedback and amendments. Where projects live or die. Feedback is collected once per phase and returned together. Changes inside an open phase cost nothing and are expected. Changes to a phase that was already approved reopen it.

Handover. Files, logins, hosting and domain access transferred to you. On a rebuild this is also where redirects get mapped, old URLs pointed at new ones, page titles and descriptions checked and the sitemap submitted. It is not a button press. Google’s own guidance on site moves is to keep redirects for at least 1 year so signals transfer, which means the redirect map is a deliverable, not an afterthought.

After handover there are 30 days of support covering bug fixes, minor content edits, image swaps and questions, extended to 45 days on larger builds. New pages and new features are outside it. That boundary exists so the support window does not turn into an unpriced second project, which is a real way for a studio to quietly go out of business.

The payment structure follows the same shape on larger work: 25% on signing, 35% on design sign-off, 40% on launch. Smaller projects run 50% to start and 50% at launch. The middle milestone is deliberately tied to design sign-off, because that is the moment the schedule becomes real.

Why is my website late?

Three reasons, in order of how often they occur.

The content did not arrive

This is the big one and it is not close.

Design can start on placeholder text. It cannot finish on it. And the content a B2B site actually needs is rarely sitting in a folder waiting to be pasted in. It is certification scope, current contract details, project records with dates and values, equipment lists, technical method, staff names and qualifications. That material lives in a certificate folder, an old tender submission, and one person’s memory.

Look at what a prequalification-led site has to carry. On Hong Hock Global, a water infrastructure contractor, the site states BizSafe Star, BCA-certified engineers and managers, and current contract scope early and plainly, because that is the order the buyer reads in. None of that exists as web copy anywhere in a contracting business before someone sits down and writes it.

So the honest question at kickoff is not “when can you launch” but “who is writing this, and when is their week free”. If the answer is “our operations manager, sometime”, the date is already at risk and nobody has drawn a single screen yet.

Feedback came in fragments

The second killer is quieter. Feedback arrives as five separate emails over nine days, each one revisiting something the previous one settled, sent by different people who have not read each other’s.

Each fragment is small. Each one costs a context switch, a re-open, a re-check of whether it contradicts the last. Together they cost more than the change itself and they make the phase impossible to close, which means the next phase cannot start.

Consolidating feedback into one set per phase is the single most useful thing a client can do for their own launch date, and it costs nothing. Collect it internally, resolve your own disagreements before sending, then send once.

An approver appeared at the end

The most expensive failure, and the most avoidable.

A director, a group marketing head, an overseas parent company, a founder who was travelling. Someone with real authority who was not in the discussion phase, did not see the structure, did not see the design, and now sees the finished site two days before launch with opinions about the navigation.

That is not their fault. Nobody put them in the room. But the cost lands on the schedule, and it lands late, when the work is most expensive to undo.

Name every approver at kickoff. If someone can veto the homepage, they need to see the wireframe. If they cannot make time for the wireframe, they need to delegate the veto in writing. One of those two things has to be true before design starts.

What unlimited revisions actually costs

We offer unlimited revision rounds, and we mean it. There is no counter, no “three rounds included, then hourly”. Design is iterative and a studio that charges for iteration is optimising for the wrong thing.

But that offer comes with one condition stated plainly, before work starts rather than after: revisions requested after a phase has been approved will extend the timeline accordingly.

Not “may”. Will. Reopening an approved design phase costs roughly what the design phase cost the first time, because the build sits on top of it. Moving a section on a wireframe is free. Moving the same section after the page is built is a week.

This is not a penalty and there is no extra invoice attached to it. It is arithmetic, and the only dishonest thing a studio can do here is pretend otherwise, agree cheerfully to a late change, and then quietly miss the launch date without ever having said the two were connected.

So the trade is: unlimited revisions, timeline-aware. Consolidate feedback per phase and you get iteration for free. Send it after sign-off and you are choosing a better site over an earlier one, which is often the right choice, as long as it is a choice.

What buys the time back

None of this requires a change to how we work. It requires four decisions on your side, and all of them are free.

Name the approvers before kickoff. Everyone who can say no, listed, in the first meeting. Anyone not on that list is a reviewer, not an approver.

Assign a content owner with actual hours. One person, named, with time blocked in their calendar. Not a committee and not whoever is least busy, because that person changes weekly.

Do the inventory early. Certifications with scope and expiry, project records with dates, equipment and capacity, staff qualifications, existing photography that is worth keeping. Most companies find half of what they need is missing or out of date, and finding that out in week one is cheap.

Batch the feedback. One set per phase, internally reconciled, sent once.

Do those and the six to ten weeks holds. Skip them and the project does not get worse, it just gets longer, and the launch slides for reasons nobody attributes correctly afterwards.

The part nobody says out loud

Studios rarely tell clients that the schedule is mostly theirs, because it sounds like advance excuse-making. So instead a date gets agreed, everyone nods, and eight weeks later there is a conversation about slippage where both sides are quietly annoyed.

We would rather have that conversation at the start. Your date is usually fine. It survives if the content has an owner, the feedback comes in one set per phase, and every person with a veto has seen the structure before anything is designed.

If you want to see how this shapes up against a real scope and a real budget, the bands and what moves them are published rather than quoted on request. Bring the date you are working towards to the first call. It is a scoping input, not a pressure tactic, and it is far more useful to us in week one than in week seven.