From idea to execution: how projects actually come together
The honest middle of a project — the part agencies hide behind pitch decks and process diagrams.
Every studio website has a ‘Process’ page. Five tidy steps. Usually with icons. Usually called things like ‘Discover,’ ‘Define,’ ‘Deliver.’ None of it is wrong, exactly — it’s just not what actually happens.
What actually happens is messier and more interesting. Here’s the honest version of how a project moves from a first conversation to something shipped.
Week 0: before the project starts
Most of the useful work happens before we’ve signed anything. A good intake conversation is half therapy and half interrogation. We want to know what the founder actually wants — not the brief they’ve written, but the thing underneath. ‘We need a rebrand’ almost always means something more specific: ‘our sales calls are harder than they should be,’ or ‘a new investor is coming in and the current site is embarrassing,’ or ‘we’re pivoting and nothing fits anymore.’
Until that underneath is named out loud, the project will drift. So we ask the uncomfortable question as early as possible: ‘If we did this perfectly, what would change for you six months from now?’
Week 1: the argument
Before we design anything, we write. Not copy — an argument. One page. What the product is, who it’s for, what it wants to be instead of what it is. The founder reads it and either nods or argues. Both are useful.
Arguing at this stage is cheap. It costs an hour of conversation. Arguing at Week 4, over a completed design, costs two weeks of rework. So we push the arguments as early as we can.
“The cheapest hour in a project is the one where someone disagrees on the premise before a single pixel is placed.”
Weeks 2–3: the shape
Design at this stage is not about finish. It’s about posture. Is the product confident or inviting? Sharp or warm? Fast or measured? We sketch two or three postures and let the client feel them. The right one is usually obvious within a day.
This is also where we commit to the structural bones — what the homepage argues, how the work reads, where the weight lives. It’s deliberately rough. Rough work is easy to change. Polished work is not.
Weeks 4–6: the slow build
This is the stretch nobody writes about on process pages. It’s not glamorous. It’s a designer refining a grid for the fifth time. An engineer moving a component into a shared pattern. A writer rewriting the About page at 11pm because the second paragraph still feels false.
This middle is where quality is actually made. Fast projects hide it. Good projects respect it.
Weeks 7–8: the last 10%
The last 10% takes 30% of the time. Motion. Copy edits. The empty states nobody thought about. The transitions between pages. The loading rhythm. Little things that individually mean nothing and collectively make the product feel considered.
Most work gets shipped before this part. That’s usually why it feels almost-right but not right.
Launch: anti-climactic, on purpose
A good launch is boring. Nothing surprises anyone. The founder has seen everything. The team has tested the edges. We’ve rehearsed the story the product tells. Launch day is just the day we stop moving things.
The drama is a bad sign. If launch week is chaotic, it means the earlier weeks skipped something.
The part of the process we’re proudest of isn’t on any process page — it’s the willingness to sit in the quiet middle, where the work stops being novel and starts being real.