Capacity Planning

Definition

Capacity planning determines production or service capacity needed to meet demand, then aligns resources to match. For manufacturing: machine hours, labor shifts, facility space. For SaaS: infrastructure, engineering headcount, sales capacity. Efficient capacity planning improves profitability by up to 10%. [src3]

Key Properties

• Three strategies: Lead (add before demand), Lag (add after), Match (add incrementally)
• SaaS dimensions: Infrastructure, engineering velocity, sales capacity, support capacity
• Manufacturing dimensions: Machine hours, labor, facility space, tooling, material throughput
• Target utilization: Manufacturing 75-85%, SaaS 50-70%, professional services 70-80%
• Planning horizons: Strategic (1-5yr), tactical (3-12mo), operational (daily-weekly)

Constraints

• Only as good as demand forecasting [src3]
• Over-capacity wastes capital; under-capacity loses revenue [src1]
• SaaS scales in minutes; manufacturing equipment takes 6-18 months [src5]
• Above 85% utilization (mfg) or 70% (SaaS) = insufficient buffer [src3]
• Only 15% of manufacturers have SaaS-based capacity management [src4]

Framework Selection Decision Tree

START — Business needs to plan capacity
├── Business type?
│   ├── SaaS → Infrastructure + engineering + sales + support
│   ├── Manufacturing → Machines + labor + facility + materials
│   ├── Professional services → Headcount + skills + utilization
│   └── Hybrid → Separate digital and physical plans
├── Planning horizon?
│   ├── < 3 months → Scheduling and shift optimization
│   ├── 3-12 months → Hiring, equipment, scaling
│   └── 1-5 years → Facility expansion, architecture
├── Demand pattern?
│   ├── Stable → Lag strategy (add after confirmed)
│   ├── Growing rapidly → Lead strategy ← YOU ARE HERE
│   └── Uncertain → Match strategy (incremental)
└── Cost of being wrong?
    ├── Under-capacity costly → Buffer higher
    └── Over-capacity costly → Buffer lower

Application Checklist

Step 1: Quantify current capacity

• Inputs: Resources, hours, productivity rates
• Output: Capacity in output units per period
• Constraint: Measure effective, not theoretical capacity [src3]

Step 2: Forecast demand

• Inputs: Historical data, growth targets, pipeline
• Output: Demand forecast with confidence intervals
• Constraint: Include scenario planning (best/base/worst) [src2]

Step 3: Identify capacity gaps

• Inputs: Current capacity + demand forecast
• Output: Gap analysis per scenario
• Constraint: Apply utilization ceiling (85% mfg, 70% SaaS) [src1]

Step 4: Design adjustment plan

• Inputs: Gap analysis, lead times, budget
• Output: Phased plan with trigger points
• Constraint: Account for lead times — hire 3 months before gap, equipment 6-18 months [src5]

Step 5: Monitor and adjust

• Inputs: Actual vs forecast, utilization rates
• Output: Monthly dashboard with variance tracking
• Constraint: Reforecast quarterly — 6-month-old assumptions are stale [src3]

Anti-Patterns

Wrong: Planning for peak demand

Sizing for absolute peak creates massive idle capacity during normal periods. [src1]

Correct: Plan for P90 with elastic overflow

Base capacity at 90th percentile; use auto-scaling, temp labor, or overtime for peaks. [src3]

Wrong: Using headcount as sole SaaS capacity measure

"20 engineers" doesn't describe capacity — output depends on team structure, tech debt, architecture. [src2]

Correct: Measure in output units

Track velocity, cycle time, and throughput to forecast delivery capacity. [src2]

Wrong: Annual planning with no mid-year adjustment

By Q3, the plan is usually obsolete. [src5]

Correct: Quarterly reforecast with trigger-based adjustments

Set utilization thresholds and pipeline milestones that activate capacity changes. [src3]

Common Misconceptions

Misconception: Higher utilization is always better.
Reality: Running near maximum creates fragility — one spike causes failures. 75-85% is optimal for manufacturing. [src3]

Misconception: SaaS doesn't need capacity planning because cloud scales automatically.
Reality: Infrastructure scales, but engineering, sales, and support capacity do not auto-scale. [src2]

Misconception: Capacity planning is only for large enterprises.
Reality: Every business with limited resources benefits. A 10-person agency over-committing burns out staff just as predictably as a factory exceeding machine capacity. [src3]

Comparison with Similar Concepts

When This Matters

Fetch this when a business asks about sizing infrastructure, manufacturing capacity, headcount planning, utilization targets, or aligning resources with demand forecasts.