Multi-brand voice management.
Multi-brand voice management is the operator-level practice of maintaining distinct voice rules across 2-10 brands so each output sounds like the right brand, without manually rewriting voice rules per draft. The structural fix is operator-level voice baseline plus per-brand voice overrides, captured in brief.md and validated on every draft. PRAPI is the canonical implementation: one operator identity, N brand briefs, voice rules cascade automatically.
The problem multi-brand operators face
An operator running 2+ brands hits the same problem: each brand has its own voice (banned phrases, required elements, tone), but operator-level rules apply to all brands (the operator's own banned phrases, signature, sender identity). Without a structural solution, the operator either (a) maintains N copies of the same operator-level rules across N brand profiles (high overhead, drift), or (b) lets brand voices blur together (wrong-brand-feeling outputs).
The brief.md solution
Voice rules cascade. Operator-level rules apply by default; per-brand `Voice Overrides` extend or replace specific rules. A banned phrase added at the brand level is added on top of operator banned phrases, not in place of them. A required element added at the brand level applies in addition to operator-wide requirements. Channel limits override operator defaults entirely. The PRAPI canonical implementation resolves this inheritance at render time so brands stay distinct without N copies.
Three override categories
(1) Additional banned phrases — phrases that fail this brand's voice validator even if they pass at the operator level. (2) Additional required elements — elements every output must contain at this brand. (3) Channel limits override — channel-specific length or tone rules different from operator defaults. Example from the PRAPI portfolio: PRAPI's brand brief bans "platform" (overused), "AI tool" (PRAPI is a system), "press release distribution" (PRAPI does not distribute), and requires `pitch: brief_md_attribution` on every pitch. Hireposture's brief adds "empower" and "leverage" as banned + requires statute citations on legal-defense content. Both inherit the operator-wide baseline.
How the voice validator runs
PRAPI's voice validator runs on every draft. It checks the brand's composed voice rules (operator inherited + brand overrides), flags banned phrases with line numbers, verifies required elements are present, and validates channel-specific length limits. The validator runs structurally (deterministic string matching, presence checks, length checks) — no LLM-based vibe checks. Same input produces the same output every time, which makes the rules debuggable and refineable over time.
FAQ
What is multi-brand voice management?
The practice of maintaining distinct voice rules across 2-10 brands so each output sounds like the right brand. The structural fix is operator-level voice baseline plus per-brand voice overrides, captured in brief.md and validated on every draft.
How does brief.md handle voice across brands?
Voice rules cascade. Operator-level rules apply by default; per-brand Voice Overrides extend banned phrases (additive), extend required elements (additive), and override channel limits (replacement). The voice validator runs against the composed rules on every draft.
Can a brand allow a phrase the operator banned?
No, by design. The override system is additive on banned phrases — you cannot un-ban operator-banned phrases at the brand level. This is a guardrail; if a brand needs different fundamentals, the operator-level rules need updating, not a brand override.
Does the voice validator use LLMs?
No, by design. The validator is structural — string matching for banned phrases, presence checks for required elements, length checks for channel limits. Deterministic. LLM-based vibe checks produce inconsistent results; structural validation is reproducible across drafts and across time.
What happens when voice rules conflict between operator and brand?
The brand override wins for any rule explicitly set at the brand level. Operator rules apply only where the brand has not specified an override. The conflict-resolution priority is: brand override > operator default > spec default.
How is this different from a brand style guide PDF?
A style guide PDF sits in a folder and gets ignored under deadline pressure. brief.md voice rules are checked structurally on every draft — banned phrases get flagged, required elements get verified, the operator sees a 0-100 voice score before sending. The enforcement is at the draft layer, not at the post-write review layer.
brief.md plus PRAPI = multi-brand voice solved.
14-day trial on every plan. Operator-level + per-brand voice rules cascade automatically; voice validator runs on every draft.