ops-capacity-plannerlisted

# Capacity Planning — Sizing a Queue-Processing Team **A team at 100% utilization misses its SLA daily and burns out.** Judge by available time, demand distribution, and risk bands — not by headcount. ## Formula (baseline) - **Required FTE** = (P90 daily volume × AHT) ÷ (1 person's daily available minutes × (1−Shrinkage) × target utilization) - Shrinkage = vacation, training, meetings, sick leave (non-work attrition, typically 25–40%) - **Utilization bands**: <70 SAFE · 70–80 WATCH · **80–90 AT_RISK** · >90 CRITICAL (throughput collapse) - Utilization spread across the team above 30pp = UNBALANCED → fix distribution before adding headcount ## Judgment Rules - **Model demand as a distribution**: sizing to P50 breaches SLA on half the days; sizing to P99 overstaffs by 30–50%. **Size operations at P90**; treat P99 separately as a surge response plan. - **No ramp stacking (1–4–20)**: the proficiency curve — expect 40% utilization at month 1, 70% at month 4, 100% at month 20. Never treat a new hire as ramped immediately. - **Account for attrition**: at 30% annual turnover, a 20-person team needs budget to backfill ~6. "Net +5" is really "hire 11." - **Manager trigger**: more than 7 ICs per manager breaks 1:1s and feedback → hire a manager before passing 10. - **P99 surge response**: document one of — overflow tier, BPO reserve, or contracted on-call. "We'll figure it out then" is a fire visible to the CEO. ## Output ``` # Capacity Plan — {team} {quarter/year} ## Capacity sce