<Surface> — <what this round changes>

Date: YYYY-MM-DD Status: discovery | aligning | scoped | validated Scope: one sentence. Say what is out of scope too. Prototype: path, if any

1. What the business actually models

Not what the surface shows — what the business has. For the concepts this surface touches:

Which concepts, states and roles exist
What each state obliges someone to do (a state nobody must act on is not a queue; a state a human is blocking is)
What each action does to the business — the event it produces, in domain language
Which concepts the business does not have. Absences are findings, and they are the expensive ones

2. User view model

Circumstance and job
Lookup objects and proof attached to each
First-scan questions and secondary dimensions
Evidence for each claim: observed | reported | inferred
Confidence and validation gaps

3. Diagnosis

Name the structural error(s) — something wrong regardless of taste. Not "it looks dated".

If you cannot name one, you do not have a diagnosis. You have an opinion.

4. Principles

Only the principles this diagnosis actually forced. Each with a rejected alternative attached — a principle with no rejected alternative is decoration.

5. Information architecture

Block by block. For each:

what lands there, and why that and not something else
the business concept behind it
what it looks like empty, and at 100×

6. Scope — what does the business already support?

Bucket	Capability	Meaning
✅ Already modeled, already exposed		Rearranging what is there
⚠️ Already modeled, never exposed		Lower domain risk; validate implementation cost
❌ Not a concept in the business yet		Domain expansion; assess cost and risk explicitly

Recommended coherent slice: explain how it completes the user's job. Does not, and why: every excluded item by name, with likely cost drivers.

7. Red lines

Concepts the business does not have, which therefore cannot be designed around — only proposed as domain changes. Bold them. These are not preferences (P-06).

8. Reality-check log

The T of SCLPT. One row per assumption that met the business model. Schema: trace-schema.md.

"What is true" must be written in domain language. If you cannot state the fact without naming a table, it is probably not a product finding.

Assumption	What is true	Model	Verdict	Pattern

Coverage: which sources, roles, permissions, states and lifecycle paths were checked? What evidence was unavailable? Which claims remain low-confidence? If no new gap was found, state only that none was found in this inspected scope.

Pattern candidates: list candidates or “none”. Check each against canon.md; append only after review and authorization.

9. Open decisions

Only the ones that change the shape of the answer. Each with a recommendation and its reasoning. Not a list of everything unresolved.