# Playbook: RMBC — Research · Mechanism · Brief · Copy
> 60% research, a unique named mechanism, a one-page brief, THEN the draft.
## Origin
**Stefan Georgi**, the modern high-ticket/health/finance direct-response controls school. RMBC formalizes a discipline that Ogilvy and Hopkins practiced informally (research before writing) and adds a step neither of them named explicitly: the **mechanism must be given a proper, ownable name** before a word of sales copy is drafted. This directly operationalizes Schwartz's Sophistication axis (§3.1) — at Stage 3 and above, the market has heard the plain claim and the amplified claim; only a differentiated *how* still moves them.
## When to use it
- **Sophistication Stage ≥3** markets (§3.1, §5.2) — where "the how" is the only thing left that can differentiate.
- High-ticket, considered offers where the buyer will investigate before purchasing — the research investment pays for itself in proof density downstream.
- Any brief where funnel-psychology has handed over a **Solution-Aware** or higher prospect and named-mechanism weight is required (§5.1's sophistication → mechanism-weight rule).
## When NOT to use it
- **Low-ticket, Sophistication Stage 1-2 markets** — the plain claim still works; a 60%-research investment is wasted effort where Hopkins's simple reason-why would do.
- **Time-boxed, high-volume ad testing** (Caples' domain, §4) — RMBC's research phase is too heavy for a same-day 5-variant ad-set sprint; use Great-Leads-selection + AIDA instead and save RMBC for the page those ads point to.
- Do not skip straight to "Copy" without the Brief — this is the single most common failure mode of the school: writers borrow the *shape* of RMBC copy (mechanism-sounding language) without doing the research that would make the mechanism true, producing Bencivenga's "hype substituting for proof" failure (§8.7).
## Step-by-step build
1. **Research (≈60% of total build time).** Collect: competitor claims at this sophistication stage, the prospect's own language if VoC material exists (Wiebe, §6.6), the actual mechanism of the product/offer (how it technically works, not how it's currently described), and existing proof assets (data, testimonials, demos).
2. **Mechanism.** Name it. A proper noun or ownable phrase the prospect can repeat back — not a category description. Test it against Reeves' USP bar (§4): can a competitor claim this exact string? If yes, it isn't a mechanism yet, it's a feature.
3. **Brief (one page).** Promise (the single core claim) → Mechanism (the named string) → Proof (which of the research assets back the promise, pre-tagged) → Offer (price/stack/guarantee inputs, received from funnel-psychology, not invented here). This brief is the gate before drafting — if any of the four boxes is empty, stop and go back to Research.
4. **Copy.** Only now does prose get written, using the Brief as the spine and the component order from §3.2 (headline → lede → mechanism section → proof block → offer → objections → CTA → PS).
## Worked mini-example
**Research finding:** most builder-passport competitors sell "portfolio sites" — static, unranked, no buyer-facing readiness signal.
**Mechanism named:** "the 8-Level Readiness Ladder" — an ownable, specific string, not "our platform."
**Brief:** Promise — "prove your business is buyer-ready before a buyer ever asks." Mechanism — 8-Level Readiness Ladder. Proof — 3 case builders who moved up the ladder with before/after readiness scores (tagged `[data]`), 1 buyer testimonial citing the ladder by name (tagged `[testimonial]`). Offer — Builder Passport, price/guarantee per funnel-psychology brief.
**Copy (headline drawn from the brief):** "The 8-Level Readiness Ladder: How builders prove they're buyer-ready before the first call."
## How the verify gate applies
The mechanism-section requirement in §3.2.3 and the §8.1 rubric's "named mechanism string present" check is RMBC's own step 2 made machine-checkable — a sales page with no named mechanism string fails this rubric outright regardless of how good the prose is. Because RMBC front-loads research, it should have the *easiest* time passing the **proof-coverage gate** (§8.6, every numeric claim tagged) — if it doesn't, the research phase was skipped, not just the writing. The Brief itself is worth keeping as an artifact even though it isn't the shipped deliverable: it's the fastest way to audit, before drafting, whether Promise/Mechanism/Proof/Offer are actually distinct and non-circular.