Shrinkage

Shrinkage

Shrinkage is a workforce management metric that quantifies the percentage of paid time during which agents are not available to handle customer contacts. It represents the gap between the number of agents on payroll and the number actually available to answer calls, chats, or other interactions at any given moment. Shrinkage is the single largest correction factor between the base staffing requirement calculated by Erlang C and the number of agents that must actually be scheduled.

In a typical contact center, shrinkage ranges from 25% to 38% of paid time, meaning that for every 100 hours of agent time purchased, only 62-75 hours are available for contact handling. Understanding, measuring, and managing shrinkage is essential to accurate workforce management planning.

Definition and Formula

Shrinkage is calculated as:

$\text{Shrinkage}=\frac{\text{Non-productive Time}}{\text{Total Paid Time}}\times 100\mathrm{\%}$

Or equivalently:

$\text{Shrinkage}=1-\frac{\text{Available Time (handling + idle)}}{\text{Total Paid Time}}$

The formula is applied to the total across all agents in a planning period (typically a day, week, or month). Shrinkage is used in staffing calculations as:

Failed to parse (syntax error): {\displaystyle \text{Scheduled Staff} = \frac{\text{Base Staff (from Erlang C)}}{1 - \text{Shrinkage \}}}

Components of Shrinkage

Shrinkage is composed of two broad categories: planned and unplanned activities.

Planned Shrinkage

Planned shrinkage includes scheduled activities that remove agents from contact-handling availability:

Component	Typical Range	Notes
Breaks	5-8%	Required by labor law; timing impacts interval-level staffing
Lunch	4-6%	Often 30-60 minutes; scheduling affects peak-hour coverage
Training	3-6%	New hire ramp, ongoing product training, compliance
Team meetings	1-3%	Huddles, one-on-ones, all-hands
Coaching	1-3%	Quality feedback sessions, side-by-side monitoring
Paid time off	4-8%	Vacation, personal days, holidays
System downtime	0-2%	Scheduled maintenance, platform outages

Planned shrinkage is largely controllable and should be factored into schedules proactively. WFM teams should build breaks, training, and coaching into the schedule rather than treating them as unexpected losses.

Unplanned Shrinkage

Unplanned shrinkage represents unpredictable absences and productivity losses:

Component	Typical Range	Notes
Absenteeism	3-8%	Sick days, no-call/no-show, family emergencies
Tardiness	0.5-2%	Late arrivals and early departures
Extended breaks	0.5-1.5%	Agents exceeding scheduled break time
System issues	0.5-2%	Unscheduled outages, slow systems, login problems
Unproductive aux time	1-3%	Excessive bathroom breaks, personal time, untracked activities

Unplanned shrinkage is harder to manage but must still be estimated and included in staffing plans. Historical patterns and seasonal trends inform the estimate.

Shrinkage in the Staffing Process

Shrinkage connects two critical steps in the WFM staffing process:

1. Forecast contact volume and AHT per interval
2. Calculate base staff required to meet the service level target using Erlang C
3. Add shrinkage to convert base staff to scheduled staff
4. Build schedules that deploy the scheduled headcount

Without shrinkage adjustment, WFM teams will systematically under-schedule, because Erlang C calculates how many agents must be available (in a ready state) — not how many must be scheduled (which includes those on break, in training, or absent).

Example

A contact center needs 50 agents available per interval to meet an 80/20 service level. If total shrinkage is 30%:

$\text{Scheduled Staff}=\frac{50}{1-0.30}=\frac{50}{0.70}=71.4\approx 72\text{ agents}$

The 22 additional agents cover the reality that at any given moment, roughly 30% of scheduled agents will be unavailable for contact handling.

Common Benchmarks

Environment	Typical Total Shrinkage	Key Drivers
Large contact center (500+ agents)	30-35%	Higher training investment, more structured coaching programs
Mid-size center (100-500)	28-33%	Balanced; moderate training and meeting overhead
Small center (<100)	25-30%	Less formal training; but individual absences have outsized impact
BPO / outsourced	30-38%	Higher attrition drives more training; contractual breaks vary
Remote / work-from-home	27-32%	Reduced commute tardiness; but potentially higher unplanned aux

These ranges are approximate and vary significantly by industry, geography, labor laws, and organizational culture.

Shrinkage Forecasting

Effective WFM requires forecasting shrinkage, not just measuring it after the fact:

• Historical trending: Calculate rolling 4-8 week shrinkage averages by component, day of week, and month
• Seasonal patterns: Absenteeism spikes around holidays; training increases during product launches
• Event-based adjustments: Planned system upgrades, all-hands meetings, and training campaigns increase shrinkage for specific periods
• Day-of-week variation: Monday and Friday typically show higher absenteeism than mid-week

Separate tracking of planned vs. unplanned shrinkage improves forecast accuracy because the two categories have different drivers and patterns.

Shrinkage Reduction Strategies

Planned Shrinkage

Planned shrinkage is not inherently bad — breaks, training, and coaching are necessary investments. The goal is to schedule them efficiently:

• Intraday scheduling: Move training and coaching to low-volume intervals rather than peak hours
• Micro-learning: Replace 60-minute training blocks with 10-15 minute modules that fit between contacts
• Automated coaching delivery: AI-powered coaching delivered at the desktop during natural breaks in contact flow

Unplanned Shrinkage

Reducing unplanned shrinkage focuses on attendance management and early detection:

• Attendance incentives: Reward consistent attendance without penalizing legitimate illness
• Real-time monitoring: Flag extended breaks and unproductive aux time as they occur, not after the fact
• Predictive absence modeling: Identify patterns (e.g., certain days, weather events, payroll cycles) that predict higher absenteeism
• Flexible scheduling: Agents who have schedule input show lower unplanned absence rates

Shrinkage vs. Occupancy

Shrinkage and occupancy are often confused but measure different things:

	Shrinkage	Occupancy
What it measures	Time not available for contacts (out of total paid time)	Time handling contacts (out of available time only)
Includes breaks/training?	Yes — that's what shrinkage counts	No — occupancy only measures what happens when available
Relationship	Higher shrinkage means fewer available hours	Occupancy operates within the remaining available hours
Control	Partially controllable through scheduling and attendance management	Mathematical consequence of volume, AHT, and staffing

Both must be accounted for in staffing: Erlang C determines how many agents need to be available (occupancy is embedded in this calculation), and shrinkage determines how many must be scheduled to achieve that availability.

Maturity Model Position

Shrinkage understanding and management evolves across maturity levels:

• Level 1 (Reactive): Shrinkage estimated as a single flat percentage (e.g., "we use 30%"). No component-level tracking.
• Level 2 (Foundational): Shrinkage tracked by component (breaks, training, absenteeism). Planned vs. unplanned separated.
• Level 3 (Integrated): Shrinkage forecasted by day of week and component. Intraday scheduling optimizes planned shrinkage placement.
• Level 4 (Optimized): Predictive absence models inform dynamic shrinkage forecasts. Automation handles coaching and training delivery during optimal windows.
• Level 5 (Adaptive): Shrinkage concept evolves for blended human-AI workforces. AI agents have zero shrinkage; human shrinkage management focuses on well-being and sustainable workload.

See Also

• Workforce Management — Overview of the WFM discipline
• Service Level — Target metric that shrinkage planning supports
• Occupancy — Complementary metric operating within available time
• Average Handle Time — Input to base staffing that precedes shrinkage
• Erlang C — Staffing model that produces the base staff number
• Scheduling Methods — Where shrinkage is operationalized in schedules
• Real-Time Schedule Adjustment — Intraday management of planned shrinkage
• Adherence and Conformance — Measuring whether agents follow schedules
• Coaching and Agent Development — A major planned shrinkage component
• Workforce Cost Modeling — Shrinkage's impact on cost per FTE