What is shrinkage?
Shrinkage is the gap between the agent hours you pay for and the agent hours actually available to handle customer contacts. Every contact centre has it, because people take breaks, go on leave, attend training and sometimes get sick.
The formula is shrinkage = paid hours not available for contacts / total paid hours x 100. If shrinkage is 30%, then for every ten paid hours only seven are spent available to customers.
Shrinkage is different from idle time. An agent who is logged in and waiting for the next call is not shrinkage; that is low agent occupancy. Shrinkage only counts time when the agent isn’t available to take contacts at all.
Planned and unplanned shrinkage
Shrinkage is usually split into two groups.
Planned shrinkage
- Scheduled breaks and meal breaks
- Annual leave and public holidays
- Training, coaching and team meetings
- Quality assurance feedback sessions
- Project work or time on non-contact tasks
Unplanned shrinkage
- Sick leave and other unplanned absence
- Lateness and early finishes
- System outages that stop agents working
- Unscheduled meetings or extended breaks
Planned shrinkage can be scheduled around forecast demand, for example by putting training into quiet periods. Unplanned shrinkage has to be estimated from history and covered with a buffer.
Why shrinkage matters
Shrinkage is a major input into workforce management. The staffing requirement produced by an Erlang C calculation tells you how many agents must be handling contacts in each interval. To get that many people on the phones, you need to roster more, using scheduled agents = required agents / (1 - shrinkage).
If shrinkage is underestimated, the roster looks right on paper but falls short on the day, and service level suffers. If it’s overestimated, you carry more staff than you need. Because shrinkage often sits around a third of paid time, small errors in the estimate translate into large differences in staffing and cost.
Managing shrinkage well
- Measure it properly. Track each category separately and by time of day, day of week and season, rather than using a single annual figure.
- Schedule planned activity into quiet periods. Put training, coaching and meetings where forecast demand is lowest.
- Look at the drivers of unplanned absence. Persistently high sick leave can point to fatigue from sustained high occupancy, rostering problems or low engagement.
- Reduce training time. Better tools, such as guided scripting and a good knowledge base, shorten onboarding and refresher training.
- Build in flexibility. Home-based agents, flexible shifts and shared resourcing make it easier to fill gaps when unplanned absence spikes.
Shrinkage can’t be eliminated, and it shouldn’t be: breaks, leave and training are part of running a healthy team. The goal is to plan for it accurately.