Ask most teams when the design work starts and they will point at the moment someone opens a design tool. By then the product has already been designed. What is left is decoration.
The decisions that determine whether a product is any good are made earlier and more quietly: what it is for, who it is for, what it refuses to do, and what has to be true for anyone to bother switching. Those are design decisions. They just do not look like design decisions, so they get made in passing, by whoever happened to be in the meeting.
The not-doing list
Every engagement we run produces two documents in the first week. One is a scope. The other is a list of things we have explicitly decided not to build, each with a sentence saying why.
The second document is the one that gets read. A scope tells people what is coming; a not-doing list tells them the argument is over. It converts a hundred future disagreements into one uncomfortable afternoon.
Who has to be in the room
Everyone who can say yes, and nobody who cannot. That is a smaller group than most companies expect — usually three or four people. If the group is bigger than six, someone in it does not have authority, and their presence will turn decisions into consultations.
- The person who owns the number the product is meant to move.
- The person who will answer for it if it fails.
- The person who has spoken to the most customers this month.
- Whoever is going to build it.
What good looks like
You know the scope is right when someone in the room is slightly unhappy about something being cut, and cannot make an argument for keeping it. That is the feeling of a decision actually being made rather than deferred.
You know it is wrong when everybody leaves pleased. It means nothing was cut, which means nothing was decided, which means it will all get decided again in week five when it is expensive.