Claude Code Plugins

Community-maintained marketplace

Feedback

user-persona-creation

@aj-geddes/useful-ai-prompts
4
0

Create detailed user personas based on research and data. Develop realistic representations of target users to guide product decisions and ensure user-centered design.

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name user-persona-creation
description Create detailed user personas based on research and data. Develop realistic representations of target users to guide product decisions and ensure user-centered design.

User Persona Creation

Overview

User personas synthesize research into realistic user profiles that guide design, development, and marketing decisions.

When to Use

  • Starting product design
  • Feature prioritization
  • Marketing messaging
  • User research synthesis
  • Team alignment on users
  • Journey mapping
  • Success metrics definition

Instructions

1. Research & Data Collection

# Gather data for persona development

class PersonaResearch:
    def conduct_interviews(self, target_sample_size=12):
        """Interview target users"""
        interview_guide = {
            'demographics': [
                'Age, gender, location',
                'Job title, industry, company size',
                'Experience level, education',
                'Salary range, purchasing power'
            ],
            'goals': [
                'What are you trying to achieve?',
                'What's most important to you?',
                'What does success look like?'
            ],
            'pain_points': [
                'What frustrates you about current solutions?',
                'What takes too long or is complicated?',
                'What prevents you from achieving goals?'
            ],
            'behaviors': [
                'How do you currently solve this problem?',
                'What tools do you use?',
                'How do you learn about new solutions?'
            ],
            'preferences': [
                'How do you prefer to communicate?',
                'What communication channels do you use?',
                'When are you most responsive?'
            ]
        }

        return {
            'sample_size': target_sample_size,
            'interview_guide': interview_guide,
            'output': 'Interview transcripts, notes, recordings'
        }

    def analyze_survey_data(self, survey_data):
        """Synthesize survey responses"""
        return {
            'demographics': self.segment_demographics(survey_data),
            'pain_points': self.extract_pain_points(survey_data),
            'goals': self.identify_goals(survey_data),
            'needs': self.map_needs(survey_data),
            'frequency_distribution': self.calculate_frequencies(survey_data)
        }

    def analyze_user_data(self):
        """Use product analytics data"""
        return {
            'feature_usage': 'Which features are most used',
            'user_segments': 'Behavioral groupings',
            'conversion_paths': 'How users achieve goals',
            'churn_patterns': 'Why users leave',
            'usage_frequency': 'Active vs inactive users'
        }

    def synthesize_data(self, interview_data, survey_data, usage_data):
        """Combine all data sources"""
        return {
            'primary_personas': self.identify_primary_personas(interview_data),
            'secondary_personas': self.identify_secondary_personas(survey_data),
            'persona_groups': self.cluster_similar_users(usage_data),
            'confidence_level': 'Based on data sources and sample size'
        }

2. Persona Template

User Persona: Premium SaaS Buyer

---

## Demographics

Name: Sarah Chen
Age: 34
Location: San Francisco, CA
Job Title: VP Product Management
Company: Series B SaaS startup (50 employees)
Experience: 8 years in product management
Education: MBA from Stanford, BS in Computer Science
Income: $180K salary + 0.5% equity

---

## Professional Context

Industry: B2B SaaS (Project Management)
Company Size: 50-200 employees
Budget Authority: Can approve purchases up to $50K
Buying Process: 60% solo decisions, 40% committee
Evaluation Time: 4-6 weeks average

---

## Goals & Motivations

Primary Goals:
  1. Improve team productivity by 25%
  2. Reduce project delivery time by 30%
  3. Increase visibility into project status
  4. Improve team collaboration across remote locations

Success Definition:
  - Team using tool daily
  - 20% reduction in status meetings
  - Faster decision-making
  - Higher team satisfaction

---

## Pain Points

Current Challenges:
  - Existing tool is slow and outdated
  - Poor mobile experience
  - Limited reporting capabilities
  - Difficult to customize for company needs
  - Vendor is unresponsive to feature requests

Frustrations:
  - Wasting time in status update meetings
  - Lack of real-time visibility into project health
  - Can't easily identify bottlenecks
  - Integration with other tools is difficult

---

## Behaviors & Preferences

Daily Tools:
  - Slack: Constant communication
  - Google Workspace: Document collaboration
  - Jira: Technical work tracking
  - Spreadsheets: Status reporting (workaround)

Work Patterns:
  - Typically works 8am-6pm Pacific
  - Checks email every 15 minutes
  - In meetings 50% of day
  - Works 20% of time outside office hours

Information Gathering:
  - Reads G2/Capterra reviews: High trust
  - Asks for peer recommendations: Very influential
  - Requests demos: Hands-on evaluation
  - Wants to see case studies: Similar companies

Decision Drivers:
  - ROI and measurable impact: 40%
  - User adoption potential: 30%
  - Ease of implementation: 20%
  - Price: 10%

---

## Technology Comfort

Tech Savviness: High (uses 15+ tools daily)
Mobile Usage: 40% of work on mobile
Prefers: Intuitive UI, minimal training
Adoption Speed: Fast (new tools in 1-2 weeks)
Integration Importance: Very high

---

## Customer Journey

Awareness: Product recommendations from peers
Consideration: Reviews, demos, talk to customers
Decision: Cost-benefit analysis, team input
Onboarding: Expects self-service + minimal support
Ongoing: Wants regular feature updates, responsive support

---

## Communication Preferences

Prefers: Email and Slack (avoid calls)
Response Time: 4-24 hours typical
Best Time: Tuesday-Thursday mornings
Frequency: Weekly updates during evaluation
Format: Data-driven, executive summaries preferred

---

## Key Quotes

"I need something that my team will actually use, not something
I have to force them to adopt."

"Show me the data on time savings, not just promises."

"Our tool should work as hard as we do - seamlessly across
all our devices and workflows."

---

## Persona Importance

Primary Persona: YES (key decision maker)
Frequency in User Base: 35% of customers
Influence: High (recommends to peers)
Revenue Impact: $30K ARR average

---

## Marketing & Sales Strategy

Messaging:
  - Emphasize productivity gains and ROI
  - Highlight ease of adoption
  - Show mobile-first experience
  - Demonstrate integrations

Sales Approach:
  - Provide customer references (similar companies)
  - Offer flexible demo (self-service + guided)
  - Focus on time-to-value
  - Provide ROI calculator

Success Metrics:
  - 50% adoption within 2 months
  - Net Promoter Score >50
  - Upsell to higher tier within 6 months

3. Multiple Personas

// Create persona set for comprehensive coverage

class PersonaFramework {
  createPersonaSet(research_data) {
    return {
      primary_personas: [
        {
          name: 'Sarah (VP Product)',
          percentage: '35%',
          influence: 'High',
          role: 'Decision maker'
        },
        {
          name: 'Mike (Team Lead)',
          percentage: '40%',
          influence: 'High',
          role: 'Daily user, key influencer'
        },
        {
          name: 'Lisa (Admin)',
          percentage: '25%',
          influence: 'Medium',
          role: 'Setup and management'
        }
      ],
      secondary_personas: [
        {
          name: 'John (Executive)',
          percentage: '10%',
          influence: 'Medium',
          role: 'Budget approval'
        }
      ],
      anti_personas: [
        {
          name: 'Enterprise IT Director',
          reason: 'Not target market, different needs',
          avoid: 'Marketing to large enterprise buyers'
        }
      ]
    };
  }

  validatePersonas(personas) {
    return {
      coverage: personas.reduce((sum, p) => sum + p.percentage, 0),
      primary_count: personas.filter(p => p.influence === 'High').length,
      recommendations: [
        'Personas cover 100% of target market',
        'Focus on 2-3 primary personas',
        'Plan for secondary use cases',
        'Define clear anti-personas'
      ]
    };
  }

  createPersonaMap(personas) {
    return {
      influence_x_axis: 'Low → High',
      adoption_y_axis: 'Slow → Fast',
      sarah_vp: { influence: 'High', adoption: 'Fast' },
      mike_lead: { influence: 'Very High', adoption: 'Very Fast' },
      lisa_admin: { influence: 'Medium', adoption: 'Medium' },
      john_executive: { influence: 'Very High', adoption: 'Slow' },
      strategy: 'Focus on Mike (influencer), design for Sarah (buyer), support Lisa (user)'
    };
  }
}

4. Using Personas

Applying Personas to Product Decisions:

---

## Feature Prioritization

Feature: Offline Mobile Access
  Sarah's Need: Medium (works with wifi)
  Mike's Need: Very High (field work, poor connectivity)
  Lisa's Need: Low (office based)
  Decision: PRIORITIZE (high-value user needs it)

Feature: Advanced Reporting
  Sarah's Need: Very High (executive visibility)
  Mike's Need: Low (not his responsibility)
  Lisa's Need: Medium (setup reporting)
  Decision: PRIORITIZE (key buyer needs it)

Feature: Bulk Import
  Sarah's Need: Medium (initial setup)
  Mike's Need: Low (day-to-day use)
  Lisa's Need: Very High (admin task)
  Decision: PRIORITIZE (admin enablement)

---

## Journey Mapping

Sarah's Evaluation Journey:
  1. Becomes aware (peer recommendation) → Email request
  2. Reads reviews (G2, Capterra) → Schedule demo
  3. Watches demo → Reviews case studies
  4. Wants reference → Talks to 2 customers
  5. Creates RFP → Evaluates pricing
  6. Gets team input → Makes decision
  → Timeline: 6-8 weeks

Mike's Adoption Journey:
  1. Learns about tool → Demo from Sarah
  2. Gets access → Starts with 1 project
  3. Learns through hands-on → Gradually adopts
  4. Becomes power user → Recommends to others
  → Timeline: 4 weeks

---

## Marketing Message by Persona

For Sarah (VP Product):
  Headline: "Increase project delivery speed by 30%"
  Focus: ROI, team productivity, visibility
  Channel: LinkedIn, industry publications
  CTA: "See ROI calculator"

For Mike (Team Lead):
  Headline: "Work faster, stress less"
  Focus: Ease of use, mobile, collaboration
  Channel: Twitter, Slack communities
  CTA: "Try free 30-day trial"

For Lisa (Admin):
  Headline: "Setup in 1 day, not 1 month"
  Focus: Easy administration, integrations
  Channel: Admin webinars
  CTA: "Download admin guide"

Best Practices

✅ DO

  • Base personas on real research, not assumptions
  • Include 2-3 primary personas
  • Make personas specific and detailed
  • Include direct user quotes
  • Update personas based on new data
  • Share personas across organization
  • Use personas for all product decisions
  • Include both goals and pain points
  • Create personas for different user types
  • Document research sources

❌ DON'T

  • Create personas without research
  • Create too many personas (>4 primary)
  • Make personas too generic
  • Ignore data in favor of assumptions
  • Create personas, then forget them
  • Use personas only for design
  • Make personas unrealistically perfect
  • Ignore secondary users
  • Keep personas locked away
  • Never update personas

User Persona Tips

  • Use real quotes from interviews
  • Include both job and personal details
  • Show clear motivations and pain points
  • Make personas memorable and shareable
  • Print and post personas in team space
  • Reference personas in design discussions