---
name: brand
description: "Use when planning a campaign, reviewing content, or settling who owns a metric between brand and revenue."
---

## Brand gate

Every brand deliverable passes a brand review before approval. Verdicts: **APPROVE | REQUEST_CHANGES | REFUTED**. Voice/positioning defines; it does not gate. Verify audit never skipped.

- Traffic + brand owned here. Price × conversion × close owned by revenue.

## Pattern → evidence reference (from the CMO Classify table)

| Pattern                     | Evidence expected                            |
| --------------------------- | -------------------------------------------- |
| New brand / campaign        | voice/positioning → content plan → review    |
| Content                     | brief → creation → clarity pre-gate → review |
| Email / paid / SEO / social | channel plan → review                        |
| Performance                 | finding → per-finding execution → review     |

## Taxonomy contract with revenue (from marketing-analyst matrix)

- Ownership: brand owns taxonomy (MQL/SQL, CAC/LTV defs), attribution windows, benchmarks, methodology. Revenue consumes traffic-to-revenue attribution read-only.
- Briefs declare `Type: A | B | C + attribution window + cohort keys`. No Type = return for clarification.
  - Type A (marketing performance) → brand-side; strategist gate ONLY if positioning/brand claim.
  - Type B (traffic-to-revenue) → revenue review MANDATORY.
  - Type C (joint: budget/CAC-LTV/forecast) → both gates; user synthesizes.
- Taxonomy changes arrive via user only (`NEEDS-TAXONOMY-CHANGE`); every insight carries data + window + cohort keys; attribution documented, no double-counting.

## Playbooks (condensed)

- **Positioning:** RESEARCH (context, audience, landscape) → DEFINE (voice pillars, messaging hierarchy, positioning) → DOCUMENT (clear actionable guidelines) → REVIEW (audit content vs guidelines; findings only — verdict is the gate's call).
- **Content analysis:** COLLECT (brief data) → ANALYZE (patterns, anomalies) → HYPOTHESIZE (testable, with size/duration/measurement) → RECOMMEND (specific, prioritised, impact-stated).

See `references/playbooks.md` for the 7 condensed domain playbooks (copy, keywords, contenidos, analytics).