Capacity Planner

Capacity Planner sizes ops teams that handle queued work (Support, CX, BizOps, IT ops, Finance ops) using Erlang-C queueing theory, Little's Law, and operational leadership canon. It produces a capacity sizing model at 70/80/90% utilization against P50/P90/P99 demand, a utilization health check, and a 12-month quarterly hiring plan with ramp and attrition.

✨ What it does

  • Erlang-C capacity model with P50/P90/P99 demand and P(SLA breach) probabilities.
  • Utilization health check with per-member traffic lights and team verdict (HEALTHY/SQUEEZED/OVERLOADED/UNBALANCED).
  • 12-month quarterly hiring plan with ramp curves, attrition, growth, and manager-trigger logic.
  • Shrinkage-adjusted FTE sizing.
  • Forcing-question library to validate assumptions before committing the plan.
  • Anti-pattern detection (e.g., plan-to-100% utilization, ignore attrition).

🎯 When to use it

  • Annual ops capacity planning for the next fiscal year.
  • Quarterly re-sizing when demand changes >15% or attrition spikes.
  • Pre-budget defense to justify headcount to the CFO.
  • Diagnosing an ops team missing SLA to determine if it's a sizing, process, or bottleneck problem.
  • Modeling capacity for a new team or combined org after M&A or a new segment launch.

🚀 How to use

Trigger by asking for capacity planning or headcount modeling for an ops team. Provide demand data (P50/P90/P99 volumes), AHT, SLA target, current FTE, shrinkage, and growth assumptions. Run the three scripts (capacity_modeler.py, utilization_analyzer.py, hiring_sequencer.py) with your inputs or use --sample for a demo. Example prompts:

Run capacity planner for our support team with 500 tickets/day P90, AHT 15 min, SLA 24h, current FTE 20, shrinkage 25%.
Use capacity-planner to build a hiring plan for Q3-Q4 given 30% growth and 20% attrition.

📄 Output: Three artifacts: capacity sizing report, utilization health report, and a 12-month hiring plan.

📦 Add this skill to Claude Code

# 1. Get the skills repo
git clone --depth 1 https://github.com/alirezarezvani/claude-skills /tmp/claude-skills

# 2. Copy this skill into your project (or ~/.claude/skills for all projects)
mkdir -p .claude/skills
cp -r /tmp/claude-skills/business-operations/skills/capacity-planner .claude/skills/capacity-planner

Skill source: business-operations/skills/capacity-planner/SKILL.md

⚠️ Good to know

Assumes queued work with stationary demand within a quarter; requires at least 90 days of historical data and single-class service per queue.

❓ FAQ

What is the recommended utilization sizing point?

The 80% utilization row is the sizing point, per the workflow.

What if I only have average demand, not P50/P90/P99?

Stop and pull the distribution; single-point demand estimates are the most expensive anti-pattern in ops.

When should I hire a manager?

When span of control crosses 7 ICs per manager, and definitely before crossing 10.

🤖 Overview, features, install steps and FAQ were generated from the project's SKILL.md on Sep 4, 2026. Always check the original source before running commands.