Four months is not a build estimate. It is the sum of a two-week build and fourteen weeks of waiting on decisions nobody owns. Shortening the project means shortening the decision path, not typing faster.
We have run this shape on every marketing site we have shipped in the last two years, across teams of four people and teams of four hundred. The schedule holds for a boring reason: each week ends with something a stakeholder can say yes or no to.
Where the four months actually go
Ask a team why their last rebuild slipped and you will hear about scope. Look at the calendar and you will find something else: three rounds of design review with a different set of reviewers each time, a copy pass that waited on a product launch, and a legal read that nobody booked until week nine.
The build was never the bottleneck. The bottleneck was that no single meeting had the authority to end an argument.
The four-week shape
Each week has one deliverable and one decision. Nothing moves to the next week until the decision is made, which is the part that feels uncomfortable and the part that saves the month.
- Week 1
Message and map
We interview sales, read your won and lost deals, and write the page map with the headline for each page. The decision at the end of the week: is this what we sell?
- Week 2
Design the three pages that matter
Home, the primary service page, and the contact flow. Everything else inherits from these. The decision: does this look like us?
- Week 3
Build and wire the tracking
Static build, real copy, analytics and CRM plumbing in place from day one rather than bolted on after launch. The decision: is anything missing before we point the domain?
- Week 4
Launch, then watch
We ship on a Tuesday so there are three working days to catch anything odd. The decision: what do we change first, based on real traffic?
Three habits that keep it on schedule
Write the copy before the design
A layout built around real sentences never needs a second pass to fit them. A layout built around placeholder text always does.
Review live, not in a thread
Thirty minutes on a call with the two approvers beats four days of comments. Everything that comes out of that call is a decision, not an opinion.
Cut the page nobody can describe
If a stakeholder cannot say who reads a page and what it should make them do, it is not a page. It is a placeholder for an unfinished argument, and it belongs in a later phase.
What you give up
Four weeks buys a site that sells one thing clearly. It does not buy a twelve-page resource library, a localisation layer or a design system. Those are real projects with their own months, and pretending otherwise is how a four-week plan becomes a four-month one. What it does buy:
- One service, described the way sales describes it
- Proof a buyer recognises, not a logo wall
- One path to a booked call, instrumented end to end
Start there. The library can come in month three, once you know which questions buyers are actually asking. You can see how this played out on two builds in our case studies.
