Multi-Framework Copy Laboratory
Write the same piece four ways, judge them against each other, then build a synthesis from the strongest parts.
Copywriting Frameworks
You are running a copy laboratory. Rather than picking a framework and hoping, you will write the same piece through four different structures, evaluate them adversarially, and synthesise the strongest result.
This works because frameworks are lenses. Each surfaces different material, and the best final piece usually borrows from several.
## Brief
- Product:
- Audience:
- Problem:
- Proof available:
- Desired action:
- Format:
- Voice:
---
## Phase 1 — Diagnose
Establish before writing:
- Awareness level of , with reasoning
- Consideration weight of the decision
- What evidence genuinely exists — be strict, and mark anything you do not actually have as [NEEDS DATA]
- Emotional register of : irritation, risk, or genuine distress
That last one is a constraint, not a note. If touches real distress — health, money worries, job security — agitation-based structures are off the table, and you should say so here.
---
## Phase 2 — Write four versions
Produce the full piece for four times, each committing fully to one structure. Do not hedge between them — a half-applied framework tests nothing.
**Version A — PAS.** Problem, agitate, solve. Find the sharpest real cost of .
**Version B — BAB.** Before, after, bridge. Describe both states with equal specificity.
**Version C — FAB.** Features, advantages, benefits. Push each chain until it reaches something the reader genuinely cares about.
**Version D — Star-Story-Solution.** Narrative. If no real story exists, write it with [STORY NEEDED] markers rather than inventing a customer.
Each version should be genuinely different in what it leads with, what it emphasises, and where it places the ask — not the same copy reordered.
---
## Phase 3 — Adversarial evaluation
Now judge each version as a hostile reader who does not want to buy. For each:
- **The first objection.** What is the very first thing a sceptic thinks reading the opening? Does the piece answer it?
- **The weakest claim.** Which assertion is least supported? Quote it.
- **The moment they leave.** Where exactly would a reader stop? Every piece has one — name the sentence.
- **The substitution test.** Could a competitor run this with their name swapped in? Mark every sentence where the answer is yes.
- **The hindsight test.** Having bought and used the product, would the reader think this copy was accurate? Anything that fails here is over-claiming.
Score each version 1–10 with reasoning. Be willing to score one low — a laboratory that finds every version good has learned nothing.
---
## Phase 4 — Synthesise
Now build the strongest possible piece by taking the best element from each:
- Which opening earned attention most honestly?
- Which section carried the most credible evidence?
- Which framing of the transformation was most believable?
- Which close felt proportionate rather than pushy?
Write the synthesis. It should not follow any single framework rigidly — it should follow the logic of the reader's questions, borrowing structure where structure helps.
State explicitly what you took from each version and why.
---
## Phase 5 — Final adversarial pass
Run the synthesis through the Phase 3 tests again. It should beat all four originals; if it does not, say so and explain what went wrong in the synthesis.
Then produce:
- The final copy
- The three claims most likely to be challenged, and what evidence each needs
- A test plan: the single highest-leverage variable to test, and what result would indicate the Phase 1 diagnosis was wrong
- An honest statement of the piece's remaining weakness
---
## Standing rules
- Never invent proof, customers, statistics or quotes. [NEEDS DATA] throughout.
- No manufactured urgency or scarcity. If there is no real reason to act now, do not fabricate one — sell the reason to act at all.
- If sits in the distress register, say plainly which techniques you are declining to use and why.
- If the honest conclusion is that no framework saves this because the offer or positioning is weak, say so.
## Avoid
- Opening with a definition, a rhetorical question, or "In today's fast-paced world".
- The words "delve", "tapestry", "landscape", "game-changer", "unlock", "elevate", "seamless", "robust", and the construction "it's not just X, it's Y".
- Padding: any sentence that restates the previous one in different words.
- Evenly-weighted sections. Real arguments have a centre of gravity.
- Hedging every claim into meaninglessness. Take a position.
## Final pass
Critique your own output against the quality bar, name its single weakest part, rewrite that part, and deliver the improved version. Show the critique in two or three sentences first.
From the Public Template Directory