manage-a-team-backlog

# manage-a-team-backlog Treat the backlog as a decision surface, not storage. Every retained item should have enough evidence to compare and a reason to exist. ## Procedure 1. Define intake channels, ownership, item types, workflow states, service expectations, and the team's capacity boundary. 2. Reconcile requests from tickets, documents, messages, commitments, incidents, and recurring work into one controlled view. 3. Give each item an outcome, requester, evidence, scope, urgency reason, dependencies, effort range, risk, and owner. 4. Deduplicate related requests while preserving stakeholder-specific obligations. 5. Apply explicit gates for readiness, safety, compliance, strategic fit, and minimum evidence. 6. Order work by value, harm, deadline, dependency, risk, cost of delay, and capacity rather than requester volume. 7. Limit active work and expose blocked, aging, and unowned items. 8. Review changes with the appropriate decision owner and communicate accepted, deferred, rejected, and changed requests. 9. Archive stale items with rationale and periodically test whether the ordering model predicts real outcomes. ## Failure plan - Do not hide mandatory operational or compliance work behind feature ranking. - Never treat a loud requester or old timestamp as sufficient priority evidence. - If demand exceeds capacity, force a scope, date, or resource decision rather than increasing invisible work. - Protect confidential requests in restricted linked records. ## Done - Retained work has outcome, evidence, state, owner, and ordering rationale - Active-work limits, blocked items, and capacity conflicts are visible - Decisions and stakeholder notifications are recorded in the system of record