Message Autopsy
A close examination of where an argument loses recognition, credibility or momentum—and what the failure reveals upstream.
The Copywriting Laboratory
A record of language under pressure.
The laboratory studies what language does in the presence of markets, institutions and human resistance. It documents the movement from research and diagnosis to intervention, response and revision.
Rigour should be visible in the work.
The laboratory is not a theatre of random headline competition. It preserves the context that gives an experiment meaning: audience, offer, mechanism, proof, objection, channel and stage of belief.
Each record should make it possible to understand why a change was attempted, what evidence returned and what the result can—and cannot—justify.
Working modules
Offer Reconstruction
A record of how value, packaging, sequence, risk and proof are reorganised before the final language is written.
Headline Experiments
Competing compressions of the same strategic argument, evaluated for truth, relevance, distinctiveness and belief—not cleverness alone.
Proof Architecture
The placement and hierarchy of demonstrations, documents, testimony, mechanisms and limits beside the claims they can support.
Audience Language
A structured record of how real audiences name problems, alternatives, fears and desired futures, with context preserved.
Before and After
Changes shown with their diagnostic reasoning, so revision becomes intellectual evidence rather than cosmetic theatre.
AI-Assisted Rhetorical Testing
Models used to surface interpretations, objections and variants—under human judgement, evidence standards and explicit limits.
Project record
What the laboratory preserves.
These fields create an institutional memory of the argument, the assumptions behind it and the evidence that changed it.
01 · Product or institution02 · Market03 · Audience04 · Dominant desire05 · Dominant fear06 · Alternatives07 · Offer08 · Mechanism09 · Proof10 · Positioning11 · Objections12 · Message hierarchy13 · Final copy14 · Performance15 · Audience response16 · Lessons and revisions
AI in the laboratory
An instrument for testing—not a substitute for judgement.
Models can surface interpretations, generate counterarguments, compare message structures and help expose ambiguity. Their output remains a proposal, not market evidence.
Human responsibility governs the claim, the evidence, the ethical boundary and the final decision. The laboratory records where AI contributed so assistance does not become invisible authority.
Begin with the problem
The strongest revision explains what the first version failed to see.
A laboratory record can become part of a strategic engagement when the organisation wants learning—not merely deliverables—to survive the project.