Skip to content

Goal Kit Quick Reference Card

🎯 Core Concept

Goal-Driven Development = Focus on outcomes (what users achieve) instead of specifications (what to build)

📋 When to Use Each Command

Situation Command Output File
🆕 Starting new project /goalkit.vision .goalkit/vision.md
🎯 New user outcome to achieve /goalkit.goal goals/###-name/goal.md
🔀 Choosing implementation approach /goalkit.strategies goals/###-name/strategies.md
📊 Planning measurable checkpoints /goalkit.milestones goals/###-name/milestones.md
🛠️ Building the feature /goalkit.execute goals/###-name/execution.md
📈 Tracking progress /goalkit.report goals/###-name/report.md
🔍 Retrospective analysis /goalkit.review goals/###-name/review.md
📚 Capturing insights /goalkit.learnings goals/###-name/learnings.md
✅ Breaking into tasks /goalkit.tasks goals/###-name/tasks.md

🔄 Basic Workflow

1. Vision → 2. Goal → 3. Strategies → 4. Milestones → 5. Execute → 6. Measure → 7. Review
     ↑                      ↑              ↑              ↑                           |
     └──────────────────────┴──────────────┴──────────────┴───── Feedback Loops ─────┘

Key principle: Be willing to loop back based on what you learn!

✅ Success Criteria Checklist

For each success criterion, ensure it's:

  • Measurable: Specific number/percentage (not "good" or "fast")
  • Time-bound: Clear deadline (by when?)
  • User-focused: User outcome, not feature (what users achieve)
  • Baseline defined: Current state documented
  • Achievable: Realistic given resources

Example: - ❌ Bad: "Users are happy with the feature" - ✅ Good: "NPS score ≥ 8/10 within 3 months (baseline: 6/10)"

🎯 Writing Good Goals

The Formula

[User/Business Outcome] + [Measurable Success Criteria] + [Timeframe]

Examples

❌ Bad (Specification):

Build a REST API with 5 endpoints for user management

✅ Good (Outcome):

Enable mobile app to sync user data reliably - 95% sync success rate - <2s sync latency (p95) - 90% user satisfaction - Within 6 weeks

🔀 Strategy Exploration Template

Always explore 2-3+ approaches before choosing:

Option 1: [Approach Name]
  Pros: [Benefits]
  Cons: [Tradeoffs]
  Cost: [Time/money]
  Risk: [High/Medium/Low]

Option 2: [Approach Name]
  [Same structure]

Option 3: [Approach Name]
  [Same structure]

Selected: Option [X]
Rationale: [Why chosen, why others rejected]

📊 Milestone Planning Checklist

Each milestone should:

  • Deliver standalone value: Can be demonstrated independently
  • Validate assumptions: Tests a hypothesis or risk
  • Be measurable: Clear success criteria
  • Build progressively: Each milestone builds on previous
  • Front-load risk: Riskiest assumptions validated first

Example: - M1: Prototype with 5 users (validates UX) - M2: Backend with real data (validates performance) - M3: Beta with 50 users (validates adoption)

🚦 Metric Decision Thresholds

Define what to do based on metric values:

Zone Threshold Action
🟢 Green ≥ Target Continue, scale, celebrate
🟡 Yellow Near target Investigate, adjust tactics
🔴 Red < Threshold Pivot strategy or goal

Pivot Trigger Example:

If metric stays in red zone for 2 weeks → Explore new strategy

🔄 When to Pivot

Check metrics:
├─ All GREEN → Continue current approach
├─ Some YELLOW → Investigate and adjust tactics
└─ RED for [X weeks] → Pivot
   ├─ Tactical pivot: Adjust execution
   ├─ Strategic pivot: Try different strategy
   └─ Goal pivot: Reframe the goal

📚 Common Patterns

Pattern 1: New Feature

Goal (user outcome) → Strategies (3+ options) → 
Milestones (3-5 checkpoints) → Execute → Measure → Review

Pattern 2: Performance Optimization

Goal (performance target) → Strategies (profiling) → 
Milestones (incremental improvements) → Execute ↔ Measure (tight loop)

Pattern 3: Exploratory Innovation

Goal (learning objective) → Strategies (experiments) → 
Milestones (hypothesis tests) → Execute → Measure → Pivot (rapid)

❌ Common Mistakes to Avoid

  1. Goals that are specs: Focus on outcomes, not implementation
  2. Unmeasurable criteria: Use specific numbers, not "good" or "fast"
  3. Skipping strategy exploration: Always explore 2-3+ options
  4. No baseline metrics: Measure current state before starting
  5. Milestones without value: Each should be demonstrable
  6. Not measuring until end: Measure after each milestone
  7. Refusing to pivot: Respond to data, don't ignore red metrics
  8. No learnings capture: Document insights for future goals

🎓 Decision Trees

Should I create a new goal?

New user outcome? 
  YES → Different metrics than existing goals?
    YES → Create new goal
    NO → Update existing goal
  NO → Implementation detail?
    YES → This is a strategy
    NO → This is a task

Should I explore new strategies?

Metrics meeting targets?
  YES → Continue execution
  NO → Why failing?
    Tactical issues → Adjust execution
    Strategy flawed → Explore new strategies
    Goal unrealistic → Reframe goal

📖 Key Principles

  1. Goals Over Specs: Focus on outcomes, not implementation
  2. Multiple Strategies: Always explore alternatives
  3. Measurable Success: Define clear metrics upfront
  4. Adaptive Execution: Be willing to pivot based on data
  5. Learning Integration: Treat execution as hypothesis testing

🚀 Getting Started (5 Minutes)

# 1. Install
uv tool install --from git+https://github.com/Nom-nom-hub/goal-kit.git goalkit

# 2. Initialize
goalkit init my-project
cd my-project

# 3. Start with vision
/goalkit.vision

# 4. Define first goal
/goalkit.goal

# 5. Explore strategies
/goalkit.strategies

📊 Metric Types

Ensure balanced measurement across:

  • User Behavior: How users actually use the feature
  • Business Impact: Revenue, cost, efficiency gains
  • Technical Quality: Performance, reliability, scalability
  • Learning: What we discovered, capabilities built

🔍 Quality Checks

Before starting execution:

  • Vision documented and team-aligned
  • Goal defines user outcome (not feature)
  • Success criteria are measurable
  • Baseline metrics captured
  • 2-3+ strategies explored
  • Strategy selected with rationale
  • Milestones deliver standalone value
  • Instrumentation plan ready

During execution:

  • Measuring metrics continuously
  • Documenting learnings as you go
  • Willing to pivot if needed
  • Regular team reviews

After completion:

  • All success criteria evaluated
  • Learnings documented
  • Insights applied to next goal

💡 Pro Tips

  1. Start small: Begin with one simple goal to learn the methodology
  2. Measure early: Don't wait until the end to collect metrics
  3. Document learnings: Use /goalkit.learnings after every goal
  4. Front-load risk: Validate scariest assumptions in M1
  5. Plan onboarding: Great features need great user education
  6. Set pivot triggers: Decide upfront when to change course
  7. Review this card: Check before each phase to avoid mistakes

📚 Learn More

🆘 Need Help?

Print this card and keep it handy while using Goal Kit!

Version: Phase 1 (2024) | Part of Goal Kit Methodology