Persona, Angle, Format Creative Strategy

Based on Dara Denney's public method. Not affiliated with or endorsed by Dara Denney.

When to use

Use this skill to audit paid social ads, turn customer research into concepts, brief production, or organize a creative testing roadmap. It is especially useful when an account has many executions but little clarity about which customer needs or messages they test. The deliverable connects evidence, a strategic diagnosis, selected concepts, production briefs, and a learning plan. [S3] [S6]

Start with the situation that makes someone consider buying. Build a persuasive angle for that situation, identify what the buyer already knows, and then choose the creator and format. Keep the angle portable across executions. A collection of carousels, grids, and talking heads does not by itself establish a strategy. [S2] [S7]

For an audit, emphasize the evidence and coverage gap before recommending production. For concept planning, complete the selected briefs. For a roadmap, add priorities, cadence, and review decisions. Respect the requested scope: identifying a landing-page mismatch does not authorize rebuilding the funnel or launching campaigns. See references/method.md for the full decision rules and attribution boundaries.

Procedure

1. Define the assignment. Record the offer, campaign objective, audience scope, available research, production constraints, and requested deliverable. Identify missing inputs without inventing facts. Open templates/output-template.md and complete its assignment fields; use the evidence rules in references/method.md.
2. Inspect evidence manually. Read a representative mix of customer reviews, comments, support questions, and interviews before clustering. Include complaints alongside praise and retain source locators. AI may organize the material, but verify its conclusions against the originals. Follow references/method.md. [S3] [S4]
3. Map actual buyer situations. Extract triggers, problems, desired outcomes, objections, and relevant context. Rank situations separately by observed frequency and emotional intensity. Label competitor-derived personas as provisional. Use references/method.md. [S3]
4. Map current creative targets. Tag each ad's implied persona using its premise, creator, message, and imagery. Compare those tags with buyer evidence to find underserved needs or excessive concentration. Keep unknown audiences explicit. Follow references/method.md. [S2] [S3]
5. Write the diagnosis. Explain the specific mismatch and the next opportunity in one or two sentences. Distinguish observed gaps from interpretations and check whether current concepts only address existing demand. Use references/method.md and the contrasting drafts in examples/worked-examples.md. [S3] [S8]
6. Develop portable angles. For each priority persona, state one persuasive idea, the belief it addresses, its customer signal, its proof, and its likely objection. Apply the human desire, demographic context, strategic angle, and user journey lenses without forcing a fixed quota of concepts. Read references/method.md. [S2]
7. Choose the awareness stage. Match the entry point to the buyer's existing knowledge. Consider broader educational entries when current ads rely on purchase readiness, while preserving relevant product-aware tests. Use the awareness table in references/method.md. [S2] [S5]
8. Select execution and proof. Choose talent, imagery, setting, and format because they communicate the angle credibly. Treat reuse across formats as a new execution to assess. Consult references/method.md and examples/worked-examples.md. [S2] [S7]
9. Prioritize the roadmap. Keep a central backlog, explain evidence and confidence, and choose a feasible set with room for a justified experiment. Connect quarterly coverage, monthly selection, and weekly production where appropriate. Follow references/method.md. [S3] [S6]
10. Complete production briefs. Give each selected concept a persona, trigger, short idea, three messaging options, proof assets, creator, format, and learning question. Specify which options are hook variations and which introduce another angle. Fill templates/output-template.md. [S3]
11. Plan measurement and the next decision. Separate commercial outcomes from diagnostic signals. Record comparison conditions, unresolved attribution, and what evidence would justify expanding, revising, or stopping the concept. Use references/method.md and the result fields in templates/output-template.md. [S2]
12. Review, deliver, and update. Apply checklists/review-checklist.md, repair failures, and expose material unknowns. After results arrive, revisit the diagnosis and backlog. Return the evidence summary, buyer-versus-ad map, diagnosis, concept table, selected briefs, and review plan at the requested depth. Check attribution in references/sources.md.

Files in this skill

• SKILL.md: Entry point, use cases, and the agent's working loop.
• references/method.md: Detailed principles, decisions, warnings, and edge cases.
• references/sources.md: Fetched public sources and their specific contributions.
• examples/worked-examples.md: Invented weak drafts and improved strategies with rule explanations.
• templates/output-template.md: Fill-in strategy, concept, production, and learning fields.
• checklists/review-checklist.md: Pass/fail review and practical failure repairs.