Estimation & Planning
Definition
Estimation Techniques
Breaking Down Work
// EPIC → FEATURE → STORY → TASK
// EPIC: Large initiative (weeks to months)
// "Implement user authentication system"
// FEATURE: Deliverable piece of functionality (days to weeks)
// "Password-based login"
// "OAuth integration"
// "Two-factor authentication"
// USER STORY: Small user-facing value (hours to days)
// "As a user, I can log in with email and password"
// "As a user, I can reset my forgotten password"
// "As a user, I can enable 2FA for my account"
// TASK: Technical work items (hours)
// - Create login API endpoint
// - Implement password hashing
// - Build login form UI
// - Write unit tests
// - Add rate limiting
// Good story characteristics (INVEST):
// Independent - Can be delivered alone
// Negotiable - Details can be discussed
// Valuable - Delivers user value
// Estimable - Can be sized
// Small - Fits in a sprint
// Testable - Has clear acceptance criteria
Planning Approaches
Handling Uncertainty
// Strategies for dealing with unknowns
// 1. SPIKE: Time-boxed research
// "Spend 2 days investigating OAuth providers"
// - Has clear question to answer
// - Fixed duration
// - Output: findings/recommendation, not code
// 2. BUFFER: Add contingency
// Don't plan at 100% capacity
// 70-80% planned, 20-30% buffer
Sprint capacity: 40 story points
Planned work: 30 story points (75%)
Buffer: 10 story points (25%)
// 3. RANGES instead of point estimates
"This will take 2-4 weeks"
// Not: "This will take 3 weeks"
// 4. CONFIDENCE LEVELS
"70% confident we'll finish by end of sprint"
"90% confident we'll finish within 2 sprints"
// 5. PROGRESSIVE ELABORATION
// Early: "Q2 - Auth system (L)"
// Later: "March - Password auth (M), OAuth (M)"
// Now: "Sprint 7 - Login endpoint (3), Reset password (5)"
// 6. COMMUNICATE ASSUMPTIONS
// "Estimate assumes:
// - API docs are accurate
// - No major scope changes
// - One developer full-time"
Common Estimation Pitfalls
Tips & Tricks
