Managing Backlog

Goal

Maintain a healthy, prioritized list of work items that are ready for development ("Definition of Ready"), preventing bottlenecks and "garbage in, garbage out."

When to Use

• •Before sprint planning.
• •When the backlog grows too large (> 50 items).
• •When stakeholders add new requests.

Instructions

1. Prioritization Frameworks

Don't just use "High/Medium/Low". Use a model:

• •RICE: Reach * Impact * Confidence / Effort.
• •MoSCoW: Must have, Should have, Could have, Won't have.
• •Cost of Delay: "If this waits a month, how much do we lose?"

2. Refinement (Grooming)

Make tasks actionable.

• •Clarify: Does everyone understand the "Why"?
• •Estimate: Is it a 1-day task or a 2-week task?
• •Split: If it's big, break it down.

3. Definition of Ready (DoR)

A task can only move to "Todo" if:

• •It has a clear user story.
• •It has acceptance criteria.
• •It has dependencies identified.
• •It has an estimate.

4. Backlog Hygiene

• •Delete: If it's been in the backlog for 6 months and nobody cares, delete it.
• •Archive: Don't let the "Icebox" become a graveyard.

Constraints

✅ Do

• •DO: Re-prioritize regularly. Priorities change.
• •DO: Say "No". A backlog is not a wishlist; it's a plan.
• •DO: Link backlog items to strategic goals (OKRs).

❌ Don't

• •DON'T: Let stakeholders push items to "In Progress" directly.
• •DON'T: Estimate in hours. Use points or T-shirt sizes (S, M, L) for relative complexity.
• •DON'T: Allow "Zombie tickets" (tasks with no clear owner or value) to survive.

Output Format

• •Updates to tasks.md (priority ordering, new tags).
• •backlog_refinement_report.md.

Dependencies

• •product/defining-user-stories/SKILL.md
• •planner/project-planning/SKILL.md