sprint-planning

Core Philosophy

Sprint planning is not a ritualistic exercise in estimating imaginary story points or negotiating how many Jira tickets can be crammed into a two-week container. Sprint planning is the operational execution of the Iron Triangle of Project Management (Scope, Time, Resources). Fixed time (2 weeks) and fixed resources (team capacity) require variable, ruthlessly prioritized scope. High-performing engineering teams protect sprint commitments by budgeting for capacity realities, allocating explicit buffers for bugs and on-call, and making trade-offs explicit before work begins.

4-Step Rigorous Sprint Planning Framework

Step 1: Capacity Planning & The 60-20-20 Rule

1. Never Plan for 100% Feature Capacity:
   • Software teams operating at 100% capacity inevitably fail sprint commitments due to unexpected production outages, sick leave, and code reviews.
2. The 60-20-20 Capacity Budget:
   • 60% Feature Epics: Strategic roadmap deliverables that move company OKRs.
   • 20% Technical Debt & Bug Remediation: Refactoring fragile code, updating dependencies, improving tests.
   • 20% Operational Buffer & On-Call: Production alerts, code reviews, architectural discussions, meetings.
3. Calculating True Developer Availability: $$text{Sprint Hours} = (text{Working Days} - text{Holidays/PTO}) imes 6text{ Productive Hours/Day}$$

Step 2: Backlog Refinement & Sizing Discipline

1. Story Pointing (Fibonacci Sequence: 1, 2, 3, 5, 8):
   • Sizing measures complexity, uncertainty, and effort, not raw hours.
   • The 8-Point Ceiling: Any user story sized at 8 or higher is too large and ambiguous for a single sprint. It must be broken down into smaller, testable sub-stories before entering the sprint.
2. Definition of Ready (DoR):
   • A ticket cannot enter a sprint unless it has: Clear user story, technical architecture notes, unambiguous acceptance criteria (BDD/Gherkin), and designs attached.

Step 3: Sprint Goal & Scope Commitment

1. The Single Unifying Sprint Goal:
   • Every sprint must have one overarching business objective that unifies the team:
     • Example: "Enable enterprise users to successfully authenticate via Okta SAML SSO in staging."
   • If a sprint has no goal, it is simply a random collection of disconnected chores.

Step 4: Handling Mid-Sprint Interruptions

1. The Scope Trade-Off Law:
   • When leadership or sales requests an emergency mid-sprint feature addition:
     • Scope is never simply added on top of existing commitments.
     • The Swap Principle: "We can pull in Feature X (5 points), provided we drop Feature Y (5 points) back into the backlog. Which is the higher business priority?"

Deliverable Format: Sprint Commitment Plan (SPRINT-PLAN.md)

# Sprint Commitment Plan: Sprint [Sprint #]
*Duration: [YYYY-MM-DD to YYYY-MM-DD] (2 Weeks) | Sprint Lead: [Name]*

## 1. Sprint Goal
**Deliver production-ready OAuth2 refresh token rotation and session revocation across web and mobile.**

## 2. Capacity & Allocation Budget
- **Total Team Available Days**: 45 Dev Days (9 engineers x 5 days)
- **Committed Story Points**: **42 Points** (Historical rolling velocity: 44 points)
- **Allocation Breakdown**:
  - Feature Work (60%): 26 Points
  - Technical Debt & Maintenance (20%): 8 Points
  - Bug Fixes & Buffer (20%): 8 Points

## 3. Sprint Backlog & Priority Epics
| Issue ID | Summary | Story Points | Assignee | Priority |
|---|---|---|---|---|
| ENG-412 | Implement Redis token family revocation | 5 | @Dave | P0 (Goal) |
| ENG-415 | Wire client-side silent refresh interceptor | 3 | @Sarah | P0 (Goal) |
| ENG-420 | Fix PostgreSQL deadlock on customer export | 5 | @Elena | P1 (Tech Debt) |
| ENG-428 | Update Docker container base image to Alpine 3.20 | 2 | @Marcus | P2 (Chore) |

## 4. Definition of Done (DoD) Gate
- [ ] Automated tests pass in CI (`npm run test:all`).
- [ ] Code reviewed and approved by at least 1 peer engineer.
- [ ] Deployed and smoke-tested in staging environment.
- [ ] Documentation updated in `docs/api/`.

Worked Example: Preventing Sprint Over-Commitment Failure

• Context: Team committed 65 story points in Sprint 12; completed only 38 points, leaving morale shattered.
• Remediation: In Sprint 13, applied the 60-20-20 capacity rule; calculated rolling 3-sprint velocity (40 points); capped backlog commitment at 38 points.
• Outcome: Team completed 100% of committed stories on time; delivered the core sprint goal with zero weekend overtime.

Verification Checklist

• Clear, unifying Sprint Goal articulated and agreed upon by team.
• Team capacity accounts for holidays, PTO, and 20% operational buffer.
• No user story exceeds the 8-point complexity ceiling.
• Backlog stories satisfy the Definition of Ready (DoR).
• Mid-sprint scope additions enforce the reciprocal swap principle.

Anti-Patterns

• 100% Feature Planning: Planning zero capacity for bugs, code reviews, or production emergencies.
• Sprint Goals with 12 Disjointed Bullet Points: Having no cohesive focus for the two-week cycle.
• Mid-Sprint Scope Ballooning: Letting stakeholders dump 5 new urgent features into the active sprint without dropping existing tickets.