From an idea to a product in the world

The route

Products differ. The ground they cross does not. Every product we have worked on moves through the same three disciplines in the same order, up one side and down the other.

You start at brand, because it is the hardest thing to pivot later and it sets the lens every decision above it gets made through. You climb through UX to UI, where brand and experience finally have to agree with each other in a single artifact — which is usually where the consequential decision sits.

That is the Summit, and what it produces is a shape: the product settled far enough that someone can build from it, still loose enough to leave the how open, and bounded enough to be finished. A direction on its own is not a shape. It is the difference between knowing where you are going and knowing how much of it you are doing — and it is what everything below the Summit spends itself proving.

Then the same three in reverse, and this is the half most studios do not sell. The interface stops being one hero screen and becomes a system that has to hold. Real users arrive and replace the prototype. The story you told at the base finally meets a market that either hears it or does not. Going up, each discipline is a belief. Coming down, each one is tested.

Six numbered stretches, and the Summit is the turn between them rather than a stage of its own — it happens inside the Climb that reaches it. Each of the six is at least one Climb: a bounded week with one Question and one Decision at the end of it. None of them can be skipped for free. What you can do is arrive part-way up, having already settled the ones below you.

A Climb on the way down is still a Climb, at the same price. What it does not have is a Summit — nothing is left to decide, so the week goes into making the decision hold, and we are available daily instead of at a fixed midweek meeting. Which side of the hill your week falls on is a position, not a different thing to buy.

Where you start

Almost nobody starts at the bottom. You start where your product already is, and the useful question is not what to do first but what is still unclimbed above you.

Starting above the base is normal. What it does mean is that some decisions below you were made by default rather than on purpose — and if the product is stalling, a skipped stretch is the first place to look.

See the whole hill →

How long a hill takes

A Climb is one week. One Question, one Summit, one Decision. A question too large for a week is not a longer Climb; it is several, taken one at a time, and each one still has to land on a decision by Wednesday.

Every stretch needs at least one, so crossing a whole hill has an arithmetic minimum: six stretches, one Climb each, is six weeks — and at $8,000 a week, $48,000 — from nothing to a product that has met its market. That is the fastest the ground can be crossed, and it assumes every question is settled on the first run at it.

Weeks are bought one at a time. Plenty of people buy exactly one, get the decision they came for, and go and build — six is how much ground a whole hill has on it, not how many you take.

Treat it as arithmetic, not as an estimate of your hill. In practice a stretch takes as many Climbs as it takes: sometimes one settles brand, sometimes there are four decisions in it and it takes four. Nobody can tell you which in advance, and a studio that quotes you the number is quoting a guess.

So a real hill, base to base, is months rather than weeks — with gaps in it, because you will spend time building between decisions and should. What we can promise precisely is the unit: a week, a decision, and the artifacts to act on it.

Why we sell one stretch at a time

The route above is a map: where you are, and what is still unclimbed above you. Which is a different thing from a plan.

A plan for the whole hill has to be written at the moment you know least about it — before the user research, before anyone has clicked a prototype, before the market has heard the story. Every decision after the first one would be committed to in advance, on the strength of that guess. Six stages at a price each is a waterfall with better vocabulary.

There is a sharper version of the problem. Every Climb ends by naming the Question worth taking on next — and if that week is already paid for, the answer can never honestly come back not yet, or not that one. A prepaid decision is not a decision, and the decision is the thing you are buying.

Two things make buying one at a time safe rather than open-ended:

• No signed Question, no scheduled week

  The Question, the Summit date, and the artifact you expect to have go into the agreement before the week is booked. You are not buying a week of our attention and finding out later what it was for. Monday sharpens the Question; it does not go looking for one.

• We tell you what the week is before you pay

  If what you describe is bigger than one Question, we say which Climb we would make first and why. If it turns out there is no Question worth a week yet, we say that instead. That answer is our job rather than yours, it costs nothing to hear, and because the price does not move we have no reason to give you any answer but the honest one.

You are never committed past the Climb you are in.

What you have at the end

Every Climb leaves a Question, a Decision and the reasoning behind it, the artifacts that support or act on it, and one reflection on what was learned. It accumulates in your own workspace, and it is yours.

Which is the part that compounds. A year in, you can answer why the product is the way it is — not from memory, but from a record of what was uncertain, what was chosen, and what it ruled out. New people can read it. The second hill costs less than the first, because you have already learned to read the ground.