ICMI is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.


Call Center Turnover: A Tough Measure

Call center turnover, presumably, measures management’s ability to retain a stable and experienced work force to sell or service customers. It is often used by executives comparing in house or outsourced call center operations or when shopping for an outsourced call center.

However in 24 years working in call centers as a line manager and as a call center consultant, I rarely see the turnover computation done correctly. Reading only the next paragraph, see if you can calculate the call center turnover rate.

You run a call center with 100 agents on January 1. During the month, 1 agent is promoted to supervisor, 1 takes a job in the mail room, 1 goes out on medical leave, 1 leaves the firm, and 7 new agents are hired (2 end the month in training while 5 are taking calls) What is your call center turnover rate?

The most common answers I see:

  • 7% — there are 7 new agents out of 100
  • 5% — there are 7 new agents out of 100 but we only count the ones taking calls
  • 4% — 4 former agents out of 100 no longer take calls
  • 3% — 4 former agents no longer take calls but the new supervisor still works in the call center and does not count towards turnover
  • 2% — 4 former agents no longer take calls, but the new supervisor and new mail room worker still work for the company and do not count towards turnover
  • 1% — 4 former agents no longer answer phones but only the one that left the firm counts toward turnover

The correct answer is 36%. Out of 100 agents 3 are no longer taking calls and were replaced. It does not matter if the replacements end the month in training or on the phones. Every agent that departs the call center floor counts; it does not matter if promoted in the mail room or out of the firm. They left the target population, which is always assumed to be agents. The employee on medical leave did not exit the target population. Note also that full- or part-time status does not matter; call center turnover is based on headcount, not hours worked.

The result for January, 3 out of 100 turned over (left and replaced) out of 100 is then annualized to 36%. You probably got a different result reflecting the lack of any real computational standard for call center turnover rates.

In fact, most firms don’t track the detailed information needed to compute call center turnover. When I sense that is the case, I automatically assign the rate provided little to no weight. Of course, there are special types of rates each firms compute for special purposes. For example, if one desired to judge their compensation relative to their marketplace, one might look exclusively at voluntary turnover rather than total turnover. Planning for future hiring would necessitate adjustments to the above computation as well.

Because the underlying data is often unavailable and because there is little agreement on a single standard, CSG has developed an alternative to the agent turnover rate: average tenure. This is simply the average number of days the call center agents have worked there. Most HR departments have the data to perform this calculation, which is a decent stand-in for call center turnover rate.

Call centers certainly perform only as well as its personnel perform, and their experience matters. It is unfortunate that measurements on this important characteristic are so often misleading.