Shrinkage

Shrinkage is the share of paid agent time that never touches a live customer contact. It counts every scheduled minute lost to breaks, coaching, meetings, absence, and system downtime. Every operation pays it. The question is whether you plan for it.

Two buckets sit inside the number. External shrinkage covers what happens away from the desk: sick days, unplanned leave, personal emergencies, jury duty.

Internal shrinkage covers paid time on the clock but off the queue. It absorbs coaching sessions, team huddles, breaks, admin work, system outages, and end-of-shift wrap-up.

Industry benchmarks move with channel and program mix. Time Doctor’s call-center shrinkage analysis puts working shrinkage between 30% and 35% for most inbound operations. Staff 100 agents at 33% and you are really fielding 67 live seats.

The number lives where three functions meet — HR policy, workforce planning, and finance. HR sets absence rules, planners forecast it into schedules, and finance costs it into the seat rate. When those three drift apart, service quality follows.

Key takeaways

• Shrinkage measures paid time agents spend NOT handling contacts, split into external (absence) and internal (on-shift, off-phone) buckets.
• Total shrinkage typically lands between 20% and 35% for mature inbound operations, and anything over 40% points to a planning or attendance problem.
• The formula is simple: (external hours + internal hours) divided by total scheduled hours, times 100.
• Ignore it and you guarantee missed service level agreements, longer wait times, and tired agents covering the gap.
• Forecasting, workforce management software, and disciplined schedule adherence are the standard controls.

How it works

Shrinkage calculation starts from the schedule. Take the total paid hours a team is meant to spend on live contacts, subtract every hour lost to a non-productive activity, divide that loss by the original schedule, and multiply by 100.

The two-bucket split matters because the controls differ. External shrinkage responds to attendance policy, absence management, and health benefits. Internal shrinkage responds to scheduling discipline and workforce management tooling.

Planners bake the number into every forecast. If a client needs 100 agents on the phones at peak hour and shrinkage runs 30%, the schedule has to book 143 seats to deliver those 100 live ones.

That arithmetic is why service level misses so often trace back to a planning assumption rather than to agent effort. A roster built on 20% shrinkage inside a 32% operation was short before anyone logged in.

Modern platforms compute the figure automatically from adherence logs and agent status codes. They flag outliers by day of week and half hour, so team leads can act before the next reporting window closes — not a fortnight later.

Two habits separate the teams that hold the number. They track shrinkage weekly at team level, not just monthly at program level. And they schedule coaching and admin into fixed blocks, well away from the peak hours the client actually pays for.

Examples

Shrinkage behaves differently across call center operations depending on program mix, agent tenure, and regulatory load. The same 30% headline can hide a training problem in one account and an attendance problem in another.

A 500-seat inbound customer service program in Manila reported 32% total shrinkage in 2024, split 12% external and 20% internal. The internal figure spiked during onboarding waves, when new-hire coaching absorbed four hours per agent per week.

Once tenure stabilised through 2025, internal shrinkage settled back to 16%. Policy never changed — the training load simply stopped landing on every roster at once.

A United States collections operation working under Telephone Consumer Protection Act (TCPA) rules watched shrinkage climb to 38% during a mandatory recertification month in 2025. Compliance training is not optional, so the hours had to come from somewhere.

The business process outsourcing provider padded that month’s schedules by 5% and held its service levels. Peer operators without the buffer took a four-point service level hit and paid contractual credits back to the client.

A 200-seat back-office staff leasing engagement reported far lower shrinkage in 2025, at 18% to 22%. Email and case work tolerate short interruptions that phone queues never do, so breaks cost less live capacity.

That split matters when you size a Philippine team. The Information Technology and Business Process Association of the Philippines (IBPAP) tracks sector headcount in its IT-BPM industry roadmap, and voice and non-voice programs sit on very different shrinkage curves.

A hybrid technical support account in Cebu went further for its 2026 planning cycle. It forecast two curves instead of one, roughly 34% on office days and 28% on home days, where commute-driven tardiness largely disappeared.

Blending them into a single average had been overstaffing Tuesdays and understaffing Mondays for a year.

Related terms

Shrinkage sits among the workforce planning metrics that decide whether a schedule holds. Each term below either feeds the shrinkage calculation, competes with it for the same staffed hours, or shows what happens downstream when the assumption is set too low.

• Workforce Management: the discipline that forecasts demand, builds schedules, and checks adherence against shrinkage assumptions.
• Service Level: the queue metric that shrinkage most directly protects or destroys.
• Idle Time: hours agents are logged in and available but not on a contact, a real-time signal that complements shrinkage.
• Occupancy Rate: the flip side of shrinkage, tracking the share of logged-on time spent on live contacts.
• Agent Turnover: a driver of external shrinkage, because churn creates absence gaps and fresh retraining cycles.
• Business Process Outsourcing: the delivery model where shrinkage assumptions get baked into commercial pricing.
• Staff Leasing: an engagement model in which the client sees agent-level shrinkage in monthly reports.

FAQ

What is a healthy shrinkage rate for a call center?

Most mature inbound operations run between 20% and 35%. Under 20% is rarely a win — it usually means a team is skipping breaks or shortchanging coaching. Over 40% points to absence problems, scheduling errors, or over-booked meetings.

How do you calculate shrinkage?

Add external hours (absence, paid leave, tardiness) to internal hours (coaching, meetings, breaks, downtime), divide by total scheduled hours, then multiply by 100. Track it weekly at team level and monthly at program level to catch drift early.

What is the difference between shrinkage and idle time?

Shrinkage counts scheduled hours that never touch a live contact, planned absences included. Idle time counts hours agents are logged in and available but waiting. Shrinkage is a planning metric; idle time is a real-time one.

Does BPO pricing include shrinkage?

Yes. Reputable providers price seats gross of shrinkage, so the hourly rate already assumes they will schedule extra agents to hit your contracted service level. For sizing and growth context, see Precedence Research’s business process outsourcing market report.

How can call-center managers reduce shrinkage?

Absence management, tighter schedule adherence, and workforce management software are the standard levers. Experienced call center managers also cluster coaching and admin work into fixed blocks so it never leaks into peak hours.

Where does shrinkage sit in the workforce management stack?

It sits at the top of every seat-count calculation, because forecasting software applies historical shrinkage by day and hour before it generates a single schedule.

Compare vetted providers who staff to your shrinkage assumptions on the Outsource Accelerator hubs page.