All skills
jamesrochabrun avatar

/design-brief-generator

@c877664

Generate comprehensive design briefs for design projects. Use this skill when designers ask to "create a design brief", "structure a design project", "define design requirements", or need help planning design work.

Use this Skill: https://skilld.dev/gh/jamesrochabrun/skills/design-brief-generator

This session only. Nothing lands on disk.

referencesuser_research_methods.md

≈4.3k tokens on demand. Your agent reads this file only when SKILL.md points to it.

User Research Methods

Comprehensive guide to user research methods for design projects.


Why User Research Matters

Reduces risk: Validate assumptions before building Saves time: Fix issues before development Builds empathy: Understand users deeply Drives decisions: Data-informed design choices Increases success: Solutions that actually work

Rule: The cost of fixing a problem increases 10x at each stage (design → development → post-launch)


When to Use Which Method

Discovery Phase (Understanding)

Goal: Learn about users, their context, and needs

Methods:

  • User interviews
  • Contextual inquiry
  • Diary studies
  • Surveys
  • Analytics review

Questions answered:

  • Who are our users?
  • What are their goals?
  • What problems do they face?
  • How do they currently solve this?

Exploration Phase (Ideation)

Goal: Generate and evaluate design directions

Methods:

  • Co-design workshops
  • Card sorting
  • Concept testing
  • Competitive analysis
  • Design critiques

Questions answered:

  • What solutions might work?
  • How should we organize information?
  • What do users expect?
  • What works elsewhere?

Validation Phase (Testing)

Goal: Evaluate specific designs before development

Methods:

  • Usability testing
  • Prototype testing
  • A/B testing
  • First-click tests
  • Tree testing

Questions answered:

  • Can users complete tasks?
  • Where do they struggle?
  • What works well?
  • Which version performs better?

Measurement Phase (Optimization)

Goal: Understand performance and iterate

Methods:

  • Analytics analysis
  • Heatmaps
  • Session recordings
  • Surveys (post-use)
  • A/B testing

Questions answered:

  • How are users actually using it?
  • Where do they drop off?
  • What's working vs. not?
  • How can we improve?

Research Methods Guide

1. User Interviews

What: One-on-one conversations with users

When to use:

  • Discovery phase
  • Understanding motivations
  • Complex topics needing depth
  • Building empathy

Duration: 30-60 minutes per interview

Participants: 5-8 users per user segment

Format:

  • Semi-structured (planned questions with flexibility)
  • Open-ended questions
  • Follow-up probes

Sample questions:

  • "Tell me about a time when..."
  • "What were you trying to accomplish?"
  • "How did that make you feel?"
  • "What would make this better?"

Best practices:

  • Start broad, get specific
  • Ask "why" 5 times
  • Listen more than talk
  • Avoid leading questions
  • Record (with permission)

Deliverables:

  • Interview transcripts
  • Key findings summary
  • Quotes and themes
  • User journey maps

2. Contextual Inquiry

What: Observing users in their environment

When to use:

  • Understanding workflows
  • Physical context matters
  • Complex processes
  • B2B/enterprise products

Duration: 1-3 hours per session

Participants: 3-6 users

Format:

  • Observe users in natural environment
  • Ask questions as they work
  • Minimal interruption
  • Take photos/notes

Four principles:

  1. Context: Go to the user's environment
  2. Partnership: User and researcher collaborate
  3. Interpretation: Develop shared understanding
  4. Focus: Guide based on research questions

Best practices:

  • Master-apprentice model
  • Think-aloud protocol
  • Photograph artifacts
  • Note workarounds
  • Ask about exceptions

Deliverables:

  • Observation notes
  • Photos of workspace/tools
  • Workflow diagrams
  • Pain points and opportunities

3. Surveys

What: Quantitative data from many users

When to use:

  • Large sample sizes needed
  • Measuring satisfaction
  • Prioritizing features
  • Demographic data

Duration: 5-10 minutes (completion time)

Participants: 100+ for statistical significance

Types of questions:

  • Multiple choice
  • Likert scales (1-5 rating)
  • Ranking
  • Open-ended (limit to 1-2)

Best practices:

  • Keep it short (< 10 questions)
  • One question per page
  • Avoid bias in wording
  • Test before sending
  • Incentivize completion

Common formats:

  • NPS (Net Promoter Score)
  • CSAT (Customer Satisfaction)
  • SUS (System Usability Scale)
  • Custom satisfaction surveys

Deliverables:

  • Response data
  • Statistical analysis
  • Charts and visualizations
  • Key findings report

4. Usability Testing

What: Users attempt tasks with your design

When to use:

  • Validating designs
  • Finding usability issues
  • Comparing design options
  • Pre-launch validation

Duration: 45-60 minutes per session

Participants: 5 users per user type

  • Nielsen Norman: 5 users find 85% of issues

Format:

  1. Introduction
  2. Background questions
  3. Task scenarios (3-5 tasks)
  4. Post-task questions
  5. Debrief

Task example: "You need to update your credit card information. Show me how you would do that."

Best practices:

  • Test with prototype or real product
  • Think-aloud protocol
  • Don't help or lead
  • Note time to completion
  • Record screen + audio
  • Use realistic scenarios

Metrics to track:

  • Task success rate
  • Time on task
  • Errors made
  • Satisfaction rating

Deliverables:

  • Usability issues (with severity)
  • Task success rates
  • Video highlights
  • Recommendations

5. A/B Testing

What: Compare two versions with real users

When to use:

  • Optimizing designs
  • Data-driven decisions
  • High-traffic pages
  • Incremental improvements

Duration: 1-4 weeks (until statistical significance)

Participants: Varies (need large enough sample)

Format:

  • Version A (control)
  • Version B (variant)
  • Randomly assign users
  • Measure conversion

Statistical requirements:

  • Sample size calculator
  • 95% confidence level
  • 80% statistical power
  • Account for baseline conversion

Best practices:

  • Test one variable at a time
  • Run until statistically significant
  • Consider external factors (holidays, etc.)
  • Don't stop test early
  • Account for novelty effect

Tools:

  • Optimizely
  • Google Optimize
  • VWO
  • Custom implementation

Deliverables:

  • Winning variant
  • Conversion lift
  • Statistical confidence
  • Implementation recommendation

6. Card Sorting

What: Users organize content into categories

When to use:

  • Information architecture
  • Navigation design
  • Taxonomy creation
  • Menu structure

Duration: 20-30 minutes per session

Participants: 15-30 for reliable patterns

Types:

Open card sort:

  • Users create their own categories
  • Discovers mental models
  • Use early in design

Closed card sort:

  • Users sort into predefined categories
  • Validates existing structure
  • Use later in design

Format:

  1. Give users cards with content items
  2. Ask them to group related items
  3. Ask them to name groups
  4. Analyze patterns

Tools:

  • Optimal Workshop
  • UserZoom
  • Miro (remote)
  • Physical cards (in-person)

Deliverables:

  • Dendrogram (similarity matrix)
  • Common groupings
  • Category names
  • IA recommendations

7. Tree Testing

What: Users find content in text-only hierarchy

When to use:

  • Testing information architecture
  • Before visual design
  • Validating navigation
  • After card sorting

Duration: 10-15 minutes

Participants: 50+ for statistical significance

Format:

  1. Present text-only site structure
  2. Give user a task
  3. User clicks through structure
  4. Measure success and path

Metrics:

  • Success rate
  • Directness (optimal path?)
  • Time taken
  • First click

Best practices:

  • Test before visual design
  • 5-10 tasks
  • Realistic task scenarios
  • Don't test too deep (3-4 levels)

Tools:

  • Optimal Workshop
  • Treejack
  • UserZoom

Deliverables:

  • Success rates per task
  • Problem areas
  • Recommended structure changes
  • Piecharts showing paths taken

8. Diary Studies

What: Users log experiences over time

When to use:

  • Long-term behavior
  • Infrequent events
  • Context switching
  • Habitual use

Duration: 3-14 days

Participants: 10-20 users

Format:

  • Users log entries daily
  • Photo + text entries
  • Structured prompts
  • Follow-up interview

Prompts example:

  • "When did you use [product] today?"
  • "What were you trying to do?"
  • "Photo of where you were"
  • "What worked/didn't work?"

Best practices:

  • Keep it lightweight
  • Daily reminders
  • Incentivize completion
  • Follow up with interview

Tools:

  • dscout
  • Indeemo
  • Custom forms
  • Mobile apps

Deliverables:

  • Usage patterns over time
  • Context of use
  • Pain points
  • Opportunity areas

9. Competitive Analysis

What: Evaluate competitor products

When to use:

  • Understanding landscape
  • Identifying opportunities
  • Benchmarking
  • Inspiration

Duration: 1-2 weeks

Competitors to analyze: 3-5 direct + 2-3 indirect

Framework:

  1. Identify competitors
  2. Define criteria
  3. Evaluate each
  4. Synthesize findings

Criteria examples:

  • Features offered
  • User experience quality
  • Pricing/model
  • Visual design
  • Performance
  • User reviews

Best practices:

  • Sign up and use products
  • Complete key tasks
  • Read user reviews
  • Screenshot key flows
  • Note strengths/weaknesses

Deliverables:

  • Competitive matrix
  • Feature comparison
  • UX evaluation
  • Screenshots
  • Opportunities

10. Analytics Review

What: Analyze quantitative usage data

When to use:

  • Understanding current behavior
  • Identifying issues
  • Measuring impact
  • Continuous improvement

Data sources:

  • Google Analytics
  • Product analytics (Mixpanel, Amplitude)
  • Heatmaps (Hotjar, Crazy Egg)
  • Session recordings
  • Error logs

Key metrics:

  • Page views
  • Bounce rate
  • Time on page
  • Conversion rate
  • Drop-off points
  • Feature adoption

Best practices:

  • Define questions first
  • Look for anomalies
  • Segment users
  • Combine with qualitative
  • Track over time

Deliverables:

  • Usage dashboard
  • Key metrics report
  • Problem areas identified
  • Hypotheses for testing

Research Planning

Research Questions

Bad questions:

  • "Do users like this design?"
  • "Is this easy to use?"

Good questions:

  • "Can users find the checkout button within 10 seconds?"
  • "What prevents users from completing signup?"
  • "How do users currently manage their projects?"

Research Plan Template

## Research Plan: [Project Name]

### Background
[Why we're doing this research]

### Goals
- [Goal 1]
- [Goal 2]

### Research Questions
1. [Question 1]
2. [Question 2]
3. [Question 3]

### Method
[Which research method(s)]

### Participants
- **Who:** [User segment]
- **Number:** [How many]
- **Recruiting:** [Where/how]
- **Compensation:** [Incentive]

### Timeline
- Planning: [Dates]
- Recruiting: [Dates]
- Sessions: [Dates]
- Analysis: [Dates]
- Reporting: [Dates]

### Deliverables
- [Deliverable 1]
- [Deliverable 2]

### Team
- Researcher: [Name]
- Designer: [Name]
- Note-taker: [Name]

Recruiting Participants

Screening Criteria

Example screener questions:

1. How often do you [relevant behavior]?
   - Daily (CONTINUE)
   - Weekly (CONTINUE)
   - Monthly (TERMINATE)
   - Never (TERMINATE)

2. What tools do you currently use for [task]?
   - [Competitor A] (CONTINUE)
   - [Competitor B] (CONTINUE)
   - Other: _____ (CONTINUE)
   - None (TERMINATE)

3. What is your role?
   - [Target role] (CONTINUE)
   - Other (TERMINATE)

Recruiting Sources

Internal:

  • Customer database
  • Email campaigns
  • In-app recruitment
  • Support ticket users

External:

  • User Interviews (platform)
  • Respondent.io
  • Ethnio
  • Craigslist (with screening)
  • Social media

Incentives

Typical rates:

  • 30 min interview: $50-75
  • 60 min interview: $75-125
  • 90 min session: $100-200
  • B2B participants: $150-300
  • Executives: $300-500

Conducting Research

Interview Guide Template

## [Project] Interview Guide

**Duration:** 60 minutes

### Introduction (5 min)
- Thank you for joining
- Recording with permission
- No right/wrong answers
- Can stop anytime
- Questions before we start?

### Warm-up (5 min)
- Tell me about your role
- Typical day at work
- [Relevant context questions]

### Main Questions (40 min)

**Section 1: Current Behavior**
- Walk me through how you [task]
- What tools do you use?
- What works well?
- What's frustrating?

**Section 2: [Topic]**
- [Questions]

**Section 3: [Topic]**
- [Questions]

### Concept Test (if applicable) (10 min)
- Show prototype/mockup
- What do you see here?
- What would you do first?
- What questions do you have?

### Wrap-up (5 min)
- Anything we didn't cover?
- Final thoughts?
- Thank you + compensation

Facilitation Best Practices

DO:

  • ✅ Build rapport first
  • ✅ Ask open-ended questions
  • ✅ Probe deeper ("Tell me more...")
  • ✅ Embrace silence
  • ✅ Stay neutral
  • ✅ Take notes (or have note-taker)

DON'T:

  • ❌ Lead the witness
  • ❌ Explain your design
  • ❌ Defend your choices
  • ❌ Ask yes/no questions
  • ❌ Put words in their mouth
  • ❌ Multi-part questions

Analyzing Research

Affinity Mapping

Process:

  1. Write observations on sticky notes
  2. Group related notes
  3. Name themes
  4. Identify patterns
  5. Prioritize insights

Tools:

  • Miro
  • Mural
  • FigJam
  • Physical wall + stickies

Finding Themes

Look for:

  • Repeated phrases
  • Common pain points
  • Similar behaviors
  • Workarounds
  • Strong emotions

Quotes to capture:

  • Memorable phrases
  • Representative of theme
  • Emotionally resonant
  • Actionable insights

Prioritizing Insights

Framework:

  • Frequency: How many users mentioned it?
  • Severity: How big a problem is it?
  • Impact: How much would solving it help?

Priority Matrix:

High Impact, High Frequency → P0 (Must fix)
High Impact, Low Frequency → P1 (Should fix)
Low Impact, High Frequency → P2 (Nice to fix)
Low Impact, Low Frequency → P3 (Maybe later)

Research Deliverables

1. Research Report

Structure:

# [Project] Research Report

## Executive Summary
- Key findings (3-5 bullets)
- Recommendations (3-5 bullets)

## Background
- Research questions
- Methods
- Participants

## Findings
### Theme 1: [Name]
- Finding
- Evidence (quotes, data)
- Implications

### Theme 2: [Name]
[Repeat]

## Recommendations
1. [Action item with priority]
2. [Action item with priority]

## Next Steps
- [What comes next]

## Appendix
- Full transcripts
- Detailed data

2. User Personas

Components:

  • Name and photo
  • Demographics
  • Goals
  • Frustrations
  • Behaviors
  • Quote
  • Scenario

Keep it:

  • Based on research
  • Focused on behaviors
  • Referenced in decisions
  • Updated regularly

3. Journey Maps

Components:

  • Stages of experience
  • User actions
  • Thoughts and feelings
  • Pain points
  • Opportunities

Format:

  • Timeline across top
  • Emotional curve
  • Touchpoints
  • Behind-the-scenes actions

Research Ethics

Informed Consent

Must include:

  • Purpose of research
  • What you'll do with data
  • Recording disclosure
  • Right to stop anytime
  • How data will be stored
  • Contact information

Privacy & Data

Best practices:

  • De-identify data
  • Secure storage
  • Limited access
  • Retention policy
  • GDPR/privacy compliance
  • Don't share recordings externally

Participant Wellbeing

Considerations:

  • Don't cause distress
  • Allow breaks
  • Respect "I don't know"
  • Compensate fairly
  • Follow through on promises

Research Checklist for Design Briefs

  • Research questions defined
  • Methods selected and justified
  • Participants criteria specified
  • Sample size determined
  • Timeline planned
  • Recruiting strategy defined
  • Incentives budgeted
  • Interview guide created
  • Consent form prepared
  • Analysis plan defined
  • Deliverables specified
  • Stakeholder review scheduled

Resources

Books

  • "Just Enough Research" - Erika Hall
  • "The User Experience Team of One" - Leah Buley
  • "Interviewing Users" - Steve Portigal
  • "Observing the User Experience" - Goodman, Kuniavsky, Moed

Tools

  • Recruiting: User Interviews, Respondent
  • Remote testing: UserTesting, Lookback, Maze
  • Surveys: Typeform, Google Forms, Qualtrics
  • Analysis: Dovetail, Airtable, Notion
  • Card sorting: Optimal Workshop, UserZoom

Templates


Quick Reference

Research Method Selection

Need Method Participants Duration
Understand users Interviews 5-8 30-60 min
Observe context Contextual inquiry 3-6 1-3 hours
Large sample Survey 100+ 5-10 min
Test usability Usability test 5 45-60 min
Compare options A/B test 1000+ 1-4 weeks
Organize content Card sorting 15-30 20-30 min
Test IA Tree testing 50+ 10-15 min
Long-term use Diary study 10-20 3-14 days
Competitor insight Competitive analysis N/A 1-2 weeks
Usage patterns Analytics N/A Ongoing

Remember: Research is not a phase, it's a practice. Build it into every project.

Source: SKILL.md on GitHub

3 warnings16d5 checks · Risk MEDIUM
  • Gen Agent Trust Hub16d

    The skill includes a shell script for brief generation that is vulnerable to command injection due to unsafe variable assignment using 'eval'. Additionally, it presents an indirect prompt injection surface as it writes unsanitized user input into generated documents.

  • Socket16d

    1 alert: gptSecurity

  • Snyk16d

    Risk: LOW · No issues

  • Runlayer7mo

    4/7 files flagged

  • ZeroLeaks5mo

    Score: 93/100 · 2 sections analyzed

Signed by skilld at c877664. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 months ago.

Dormantupdated 11 months ago

README badge

README badge for jamesrochabrun/skills/design-brief-generator