Get close to the real problem before sketching a single screen.
Every brief starts as a goal, a context and a quality bar rather than a feature list. Research is where those three get written down plainly enough that the rest of the work can be checked against them.
A brief, then the questions it cannot answer yet.
The intake fixes what the work has to achieve and what counts as done. Classification decides which specialists it needs. Only then does anyone start producing.
- 01
Brief
Goal, context and quality criteria, written before any solution is proposed.
- 02
Intake
Goal and acceptance criteria made explicit, so delivery has something to be measured against.
- 03
Classification
The relevant specialism and depth are chosen for this brief, not for briefs in general.
- 04
Open questions
What is assumed rather than known is listed as a question, not smoothed over in the copy.
Less repeated work
By reusing knowledge and build patterns that already exist rather than rediscovering them.
Assumptions in the open
A weak idea is easier to correct while it is still a paragraph.
Adapted from the studio's architecture overview, September 2026. The lines represent work and knowledge flows; review and publication are explicit steps. It does not imply fully automated feedback or measured speed gains.