Starting point: You have a feature brief or decision artifact, and you need to write the current-state blueprint for the system or feature it describes.
Extraction process:
- Establish narrative intent:
- Why does this feature/system exist? What problem does it solve?
- What constraints or decisions shaped its current design?
- What trade-offs did the decision encode? (What alternatives were rejected and why?)
- Build the "why" section first:
- Open with the decision context: "We decided to build X because of Y constraint."
- Link to the decision artifact (brief, initiative, goal).
- Name the trade-offs: "This eliminated alternative Z because of constraint A."
- Describe current state:
- "Here's what actually shipped and how it works."
- Focus on user-facing behavior and internal architecture as it relates to the decision.
- Call out conventions or patterns specific to this feature.
- Capture consequences:
- How does this decision affect adjacent systems?
- What constraints does it create for future choices?
- What becomes possible now that this feature exists?
Concrete example: Blueprint for "User Onboarding"
## User Onboarding
We built a guided onboarding flow to reduce time-to-first-value
for new users. Prior to this ([goal: Improve Activation Rate](#ref:...)),
users hit a blank slate and had no guidance on core features,
resulting in a 40% drop-off in week one.
The onboarding flow walks users through three core actions:
(1) creating a brief, (2) writing a goal, (3) creating an initiative.
This sequence matches our core mental model and lets users experience
leverage quickly.
We opted for in-app guided tours over email drip campaigns because
we needed real-time feedback to optimize the sequence; email would
hide drop-off points. This trade-off means we can't reach users
between sessions, but gives us 24-hour iteration cycles instead of
weekly email windows.
Onboarding is gated: it only appears for users created in the last
7 days. We chose 7 days (not 1 day) to catch returning users who
signed up but didn't complete day-one onboarding.
Narrative hooks to include:
- Causality: "We did X because Y constraint meant Z was impossible."
- Constraints: "We chose this design to optimize for A while accepting limitation B."
- Alternatives considered: "We considered X but rejected it because Y."
- Consequences: "This decision means that systems depending on us need to handle Z."
Anti-pattern: Writing the blueprint as a todo list. "Build onboarding. Add tour. Track events." That would be a spec, not a narrative of why onboarding exists and how it works.