User Persona Generator

Create detailed user personas from research data or product descriptions for product design and marketing.

Use this prompt — free See example output

The Prompt

(2 messages)

System

You are a UX researcher who creates actionable user personas. Build personas that product teams can actually use to make decisions — not fictional characters that sit in a slide deck.

Guidelines:
- Focus on behaviors, motivations, and pain points — not demographics for the sake of demographics
- Ground each trait in evidence or realistic scenarios
- Include what this persona is trying to accomplish (jobs to be done)
- Describe their current workflow and where it breaks down
- Make the persona specific enough to be useful but broad enough to represent a real segment
- Include anti-goals: what this persona does NOT care about

User

Create user personas for:

Product: {{product}}
Target market: {{market}}
Number of personas: {{num_personas}}
Research context: {{research}}

Variables

Fill in these inputs to customize your output:

{{product}}

Example: A prompt management platform for engineering teams

{{market}}

Example: B2B SaaS, mid-market to enterprise companies building AI features

{{num_personas}}

Example: 2

{{research}}

Example: Users range from ML engineers writing prompts daily to engineering managers evaluating tools. Key pain points: prompt versioning chaos, no way to test prompt changes safely, hard to track which prompt version is in production.

Example Output

Here's what this prompt generates with the sample inputs:

Sample outputclaude-sonnet-4-5

{
  "personas": [
    {
      "name": "Maya the ML Engineer",
      "role": "Senior ML Engineer",
      "company_context": "Works at a 200-person SaaS company. 3-person ML team embedded in a larger engineering org.",
      "jobs_to_be_done": [
        "Iterate on prompts quickly without breaking production features",
        "Compare prompt outputs across model versions before shipping",
        "Share prompt changes with the team for review"
      ],
      "current_workflow": "Writes prompts in a Python file. Tests manually by running scripts locally. Copies the final version into an environment variable or config file. Version history is git blame.",
      "pain_points": [
        "No easy way to A/B test prompt changes on real data",
        "Lost track of why a prompt was changed 3 months ago",
        "Testing locally doesn't catch edge cases that appear in production"
      ],
      "motivations": [
        "Ship prompt improvements faster without risk",
        "Look competent — prompt regressions are embarrassing",
        "Spend time on model work, not prompt infrastructure"
      ],
      "anti_goals": [
        "Does NOT want a heavy enterprise tool with long onboarding",
        "Does NOT want to change their development workflow dramatically",
        "Does NOT care about dashboards for management — wants tools for building"
      ],
      "quote": "I just want to change a prompt and know it won't break something in production."
    },
    {
      "name": "David the Engineering Manager",
      "role": "Engineering Manager, AI/ML Team",
      "company_context": "Manages 8 engineers at a Series B company. Responsible for the AI features roadmap.",
      "jobs_to_be_done": [
        "Understand what prompts are in production and who changed them",
        "Ensure prompt changes go through review before deployment",
        "Report on AI feature quality and cost to leadership"
      ],
      "current_workflow": "Gets Slack messages when something breaks. Reviews PRs that include prompt changes but can't easily see the actual output difference. Tracks costs via billing dashboards.",
      "pain_points": [
        "No visibility into prompt changes unless he reads code diffs",
        "Can't enforce a review process for prompt updates",
        "Leadership asks about AI costs and quality — hard to answer precisely"
      ],
      "motivations": [
        "Reduce prompt-related incidents (they disrupt sprint goals)",
        "Build team processes that scale as more engineers write prompts",
        "Have data to justify AI infrastructure investments"
      ],
      "anti_goals": [
        "Does NOT want to become the bottleneck for prompt approvals",
        "Does NOT want another dashboard he has to check daily",
        "Does NOT care about prompt syntax — delegates that to the team"
      ],
      "quote": "I need to know what changed, who changed it, and whether it was tested."
    }
  ]
}