Workforce planning fundamentals

Shrinkage explained — what it is and how to calculate it

Shrinkage is the gap between the number of agents you roster and the number actually available to handle contacts at any given time. Understanding and managing shrinkage is the difference between hitting your service level target and missing it by 10 points.

What is shrinkage?

Shrinkage is the proportion of paid agent time that cannot be used for contact handling. It's the collection of everything that takes an agent away from being ready to answer a call, respond to a chat, or work a ticket, including planned activities (breaks, meetings, training) and unplanned ones (sickness, lateness, system downtime).

Shrinkage % = (hours not available for contacts ÷ total paid hours) × 100

If an agent is paid for 8 hours but spends 2.4 hours in breaks, meetings, and training, shrinkage = 2.4 ÷ 8 × 100 = 30%.

The shrinkage formula

The key formula converts seated agents (from Erlang C) to the scheduled headcount you actually need to roster:

Scheduled headcount = Seated requirement ÷ (1 − shrinkage rate)

Example: Erlang C says 20 seated agents are needed. Shrinkage = 30%. Scheduled headcount = 20 ÷ (1 − 0.30) = 20 ÷ 0.70 = 28.6 → 29 agents.

Common mistake: using multiplication instead of division

20 × 1.30 = 26 is wrong. Multiplying by (1 + shrinkage) understates your requirement. Always divide by (1 − shrinkage rate). At 30% shrinkage, the right answer is 29, not 26: a 3-agent gap.

What's included in shrinkage?

Planned — scheduled off-phone

Breaks (paid)10–15%2 × 15-min breaks in an 8-hour shift = 6.25%. Varies by local law and CBA.

Lunch / meal break0–8%Paid lunches count; unpaid lunches are usually excluded (agent is off the clock).

Team meetings & briefings2–4%Daily or weekly team meetings, shift handovers, briefing sessions.

Training & coaching3–6%Induction, product updates, compliance training, coaching sessions.

Admin & back-office tasks2–5%Case notes, follow-ups, and admin assigned during on-shift time.

One-to-ones & appraisals1–2%Scheduled 1:1s with team leaders, performance reviews.

Unplanned — not scheduled

Sickness and absence4–8%Unavoidable in any operation. Well-managed centres target below 5%.

Lateness / early departure1–2%Agents arriving after shift start or leaving before end.

System downtime0.5–2%CRM outages, telephony problems, IT issues preventing contact handling.

Personal time (unofficial)1–3%Unscheduled breaks, extended personal activities, and adherence losses.

Total typical shrinkage: 25–35%. Add each category that applies to your operation and use the shrinkage calculator to compute your scheduled headcount.

Planned vs. unplanned shrinkage

Planned shrinkage (breaks, training, meetings) can be accounted for in scheduling — you know roughly when it will occur and can build it into shift patterns or off-phone time blocks.

Unplanned shrinkage (sickness, lateness, system downtime) cannot be scheduled — it arrives randomly during the shift and reduces your effective seat count. The only mitigation is a higher total scheduled headcount (i.e. a higher overall shrinkage rate in your planning model) or overstaffing buffer.

Planned (manageable)

Schedule breaks and meetings at off-peak times. Use training time during low-volume periods. Rotating off-phone rosters smooth the impact.

Unplanned (build in as buffer)

Track your historical sickness and absence rate by week and season. Add this to your planned shrinkage when computing scheduled headcount.

Shrinkage vs. occupancy vs. utilisation

Shrinkage

Non-contact time ÷ total paid time

The fraction of paid time lost to non-contact activities. Applied during planning to inflate seated to scheduled.

Occupancy

Traffic intensity ÷ agents seated

The fraction of seated time actually spent handling contacts. Computed by Erlang C. Target: 70–85% for voice.

Utilisation

All productive time ÷ total rostered time

Broader than occupancy: it includes training, meetings, and back-office. Typically 85–92% in well-run operations.

Shrinkage — frequently asked questions

What is shrinkage in a contact centre?

Shrinkage is the proportion of paid agent time that is not available for handling contacts. It includes anything that prevents an agent from being ready to take a call or chat: breaks, meetings, training, sickness, off-phone admin work, system outages, and so on. Typical contact centre shrinkage runs between 20% and 35%. A shrinkage rate of 30% means that for every 10 hours of paid time, only 7 hours are available for contact handling.

What is the formula for shrinkage?

Scheduled headcount = Seated requirement ÷ (1 − shrinkage rate). If Erlang C says you need 20 seated agents and shrinkage is 30%, you need to schedule 20 ÷ 0.70 = 28.6 → 29 agents. The formula uses the reciprocal because shrinkage reduces the effective fraction of each scheduled agent's time.

What is included in shrinkage?

Shrinkage includes both planned and unplanned time off-phone. Planned: breaks (15–30 min paid), lunches (30–60 min), team meetings, training, coaching, one-to-ones, after-call work (ACW) if counted separately, administrative tasks. Unplanned: sickness and absence, lateness, system downtime, unplanned training, personal time. Some centres also track 'adherence shrinkage': agents logged in but handling non-contact activities.

What is a good shrinkage rate?

A typical well-managed contact centre has total shrinkage of 25–30%. Under 20% is unusual and may mean insufficient training time. Above 35% often signals absence management or adherence problems. The 'right' rate depends on the operation: centres with long induction periods, complex products, or high training requirements will have higher shrinkage than simple transactional operations.