Claude-optimized prompt structure shaped for chatgpt.
# Task
Create a buyer persona. Separate what you know from what you're guessing.
# Inputs- Product: {product}- Audience segment: {audience_segment}- Market: {market}- Use case: {use_case}- Known insights: {known_insights}# Pre-flight
If the inputs are too thin to build a useful persona (e.g., no known insights, vague segment), ask up to 3 clarifying questions AND build a preliminary persona with stated assumptions.
# Persona structure1. **Persona snapshot**: Name, role, seniority, company size, industry (1-2 sentences).
2. **Jobs to be Done**: 3-5 jobs in this format:
- When [situation], I want to [motivation], so I can [expected outcome].
3. **Pain points**: 3-5 specific frustrations tied to the use case (not generic).
4. **Buying triggers**: What events or changes make them start looking for a solution?
5. **Decision criteria**: What matters most when evaluating options? Rank the top 5.
6. **Objections**: 3-5 likely objections and what drives each one.
7. **Messaging angles**: 2-3 value propositions that would resonate, tied to specific pains.
8. **Channels**: Where this persona researches, gets recommendations, and makes decisions.
# Evidence vs. assumption
Tag every insight:
- [EVIDENCE] = supported by known insights
- [ASSUMPTION] = inferred, needs validation
# Output
Return the full persona using the structure above, with evidence/assumption tags throughout.
claudethinking_mediummulti_perspectiveinteractive
claude variant
Claude-optimized prompt structure shaped for claude.
<context><product>{product}</product><audience_segment>{audience_segment}</audience_segment><market>{market}</market><use_case>{use_case}</use_case><known_insights>{known_insights}</known_insights></context><task>Create a buyer persona. Separate evidence from assumptions.</task><instructions>
If inputs are too thin, ask up to 3 questions AND build a preliminary persona with stated assumptions.
Structure: snapshot, Jobs to be Done (When/I want/So I can), pain points, buying triggers, decision criteria (ranked), objections, messaging angles, channels.
Tag every insight [EVIDENCE] or [ASSUMPTION].
</instructions><output>
Return the full persona with evidence/assumption tags throughout.
</output>
geminithinking_mediummulti_perspectiveinteractive
gemini variant
Claude-optimized prompt structure shaped for gemini.
Product: {product}
Audience segment: {audience_segment}
Market: {market}
Use case: {use_case}
Known insights: {known_insights}
If inputs are too thin, ask up to 3 questions AND build a preliminary persona.
Structure: snapshot, Jobs to be Done (When/I want/So I can format), pain points, buying triggers, decision criteria (ranked top 5), objections, messaging angles, channels.
Tag every insight [EVIDENCE] or [ASSUMPTION].
Based on the entire content above, return the full buyer persona with tags throughout.
Related templates
More templates in the same working category.
LinkedIn Post
Write a LinkedIn post with a scroll-stopping hook, structured insight, proof, and engagement-optimized formatting.