User Research Skill

Atomic Skill: Conduct comprehensive user research to inform design decisions

Purpose

This skill provides structured methodologies for understanding users through qualitative and quantitative research methods.

Skill Invocation

Skill("custom-plugin-ux-design:user-research")

Parameter Schema

Input Parameters

interface UserResearchParams {
  // Required
  method: "interview" | "survey" | "persona" | "journey_map" | "usability_test";
  objective: string;

  // Optional
  participants?: {
    count: number;
    criteria: string[];
    recruitment_source?: string;
  };
  timeline?: string;
  deliverables?: string[];
}

Validation Rules

method:
  type: enum
  required: true
  values: [interview, survey, persona, journey_map, usability_test]

objective:
  type: string
  required: true
  min_length: 10
  max_length: 500

participants.count:
  type: number
  min: 1
  max: 100
  default: 5

Execution Flow

USER RESEARCH EXECUTION
────────────────────────────────────────────

Step 1: VALIDATE INPUTS
├── Check method validity
├── Verify objective clarity
└── Validate participant criteria

Step 2: PLAN RESEARCH
├── Select appropriate method
├── Design research protocol
└── Prepare materials

Step 3: EXECUTE
├── Conduct research sessions
├── Collect data
└── Document observations

Step 4: ANALYZE
├── Synthesize findings
├── Identify patterns
└── Generate insights

Step 5: DELIVER
├── Create deliverables
├── Present findings
└── Recommend actions

────────────────────────────────────────────

Retry Logic

retry_config:
  max_attempts: 3
  backoff_type: exponential
  initial_delay_ms: 1000
  max_delay_ms: 10000
  retryable_errors:
    - PARTICIPANT_NO_SHOW
    - SESSION_INTERRUPTED
    - DATA_INCOMPLETE

Logging Hooks

// Execution logging
interface ResearchLog {
  timestamp: string;
  event: "start" | "step_complete" | "error" | "complete";
  method: string;
  participants_completed: number;
  insights_generated: number;
  duration_ms: number;
}

// Log events
onStart: log({ event: "start", method, objective });
onStepComplete: log({ event: "step_complete", step, data });
onError: log({ event: "error", error, context });
onComplete: log({ event: "complete", summary });

Learning Modules

Module 1: Interview Techniques

INTERVIEW FUNDAMENTALS
├── Semi-structured interviews
├── Open-ended questions
├── Active listening
├── Probing techniques
└── Note-taking methods

ADVANCED TECHNIQUES
├── Contextual inquiry
├── Think-aloud protocol
├── Critical incident technique
└── Laddering interviews

Module 2: Survey Design

SURVEY BEST PRACTICES
├── Question types (Likert, multiple choice, open)
├── Survey flow and logic
├── Bias prevention
├── Response rate optimization
└── Statistical validity

Module 3: Persona Creation

PERSONA FRAMEWORK
├── Research-based personas
├── Behavioral archetypes
├── Goals and motivations
├── Pain points and frustrations
├── Jobs to be done
└── Validation methods

Module 4: Journey Mapping

JOURNEY MAP COMPONENTS
├── Stages and touchpoints
├── Actions and behaviors
├── Thoughts and feelings
├── Pain points and opportunities
├── Moments of truth
└── Service blueprints

Module 5: Usability Testing

USABILITY TEST PROTOCOL
├── Task design
├── Success metrics
├── Moderation techniques
├── Observation methods
├── Analysis frameworks
└── Reporting templates

Error Handling

Error Code	Description	Recovery
UR-001	Invalid method	Suggest valid methods
UR-002	Insufficient participants	Recommend recruitment strategies
UR-003	Unclear objective	Prompt for clarification
UR-004	Data quality issue	Flag and document
UR-005	Analysis blocked	Provide partial insights

Troubleshooting

Problem: Low participant recruitment

Diagnosis:
├── Check: Screening criteria too narrow?
├── Check: Incentive sufficient?
├── Check: Recruitment channels appropriate?
└── Solution: Expand criteria or channels

Steps:
1. Review screening criteria
2. Adjust incentive structure
3. Add recruitment sources
4. Extend timeline if needed

Problem: Biased findings

Diagnosis:
├── Check: Leading questions?
├── Check: Sample representative?
├── Check: Moderator influence?
└── Solution: Audit methodology

Steps:
1. Review question wording
2. Check sample demographics
3. Use neutral facilitation
4. Triangulate with other data

Unit Test Templates

describe("UserResearchSkill", () => {
  describe("parameter validation", () => {
    it("should reject invalid method", () => {
      expect(() => invoke({ method: "invalid" }))
        .toThrow("UR-001");
    });

    it("should require objective", () => {
      expect(() => invoke({ method: "interview" }))
        .toThrow("objective is required");
    });
  });

  describe("interview execution", () => {
    it("should generate interview script", async () => {
      const result = await invoke({
        method: "interview",
        objective: "Understand onboarding experience"
      });
      expect(result.deliverables).toContain("interview_script");
    });
  });

  describe("persona creation", () => {
    it("should create research-based persona", async () => {
      const result = await invoke({
        method: "persona",
        objective: "Define primary user archetype"
      });
      expect(result.persona).toHaveProperty("goals");
      expect(result.persona).toHaveProperty("pain_points");
    });
  });
});

Quality Metrics

Metric	Target	Measurement
Sample validity	> 80%	Representative coverage
Insight actionability	> 90%	Stakeholder adoption
Finding confidence	> 85%	Triangulation score
Participant satisfaction	> 4/5	Post-session rating

Version History

Version	Date	Changes
1.0.0	2025-12-30	Production-grade upgrade