Attrition Rate

Attrition vs turnover

This is the question most people searching the term want answered, and the honest answer is that it depends on who is talking.

In traditional HR and finance language, employee turnover counts all departures, and the organization expects to refill those seats. Attrition counts departures that are absorbed: the role is eliminated, merged or left open, so the workforce gets smaller over time. A company in a hiring freeze shrinks by attrition. A retailer losing cashiers and hiring new ones has turnover.

In contact centers, BPOs and much of India and the Philippines' services industry, attrition simply means people leaving, and it is reported monthly. An operations director who says attrition is 6% means 6% of agents left last month, refilled or not.

Neither usage is wrong. The mistake is mixing them in the same report. Put the definition in a footnote and keep it fixed.

Monthly vs annualized attrition

Contact centers and staffing operations watch attrition monthly because classes start monthly and a bad month shows up in service level within weeks. Executives and benchmarks usually talk in annual figures.

Convert carefully. Adding up 12 monthly rates gives an annualized number that matches most industry reporting. Multiplying a single month by 12 works only when that month is normal, and the month after a big hiring wave rarely is.

Also watch the denominator. If you count trainees in headcount but only count production agent exits, your rate will look lower than it is. Many centers report three separate figures: training attrition (people who leave during classroom training), nesting attrition (during the supervised transition to live calls) and production attrition. The call center attrition page covers those stages in detail.

Typical attrition ranges

Ranges practitioners see on annualized, all-departures attrition:

• In-house US contact centers: often 30% to 45%.
• Outsourced and offshore contact centers: commonly 40% to 80%, with entry-level voice programs at the higher end.
• Retail and quick-service hourly staff: frequently above 60%.
• Corporate and salaried professional roles: typically under 15%.

First-90-day attrition is often half or more of the annual total in hourly roles, which is why new hire turnover deserves its own line on the dashboard.

Why attrition drives hiring volume

Attrition is the single biggest input into a high-volume hiring plan. A simple way to see it: required hires = planned headcount growth + (average headcount × annual attrition) ÷ (1 − training attrition).

Take a 500-agent program with 50% annual attrition and 20% training attrition. Replacement alone needs 250 agents on the floor, and because one in five trainees drops out, the recruiting team must start about 313 people in training. Bring attrition to 35% and that falls to roughly 219 trainees, nearly 100 fewer hires, plus the savings in training seats and trainer hours.

This is also why a recruiting team that hits its start numbers can still miss service level: if attrition outruns hiring, headcount drifts down month after month.

Common mistakes when measuring attrition

• Hiding training attrition. If people who quit in week two of training never appear in the rate, recruiting looks responsible for a problem that sits partly in training design.
• Comparing a BPO's monthly rate to a corporate annual rate. Convert to the same basis first.
• Treating all exits as equal. Losing a top-quartile agent costs far more than losing someone already on a performance plan. Track regrettable attrition separately.
• Excluding involuntary exits without saying so. Some teams report only voluntary attrition. That is fine if the label says so.
• Waiting for the quarter to close. Weekly exit counts by team leader let you act while the cause is still fresh.

How to lower attrition

Find the stage first. If most exits happen in training or the first 60 days, the cause is usually expectation mismatch: candidates did not understand the schedule, the call type or the metrics they would be held to. A job preview that includes a short call simulation and the real shift pattern removes many of those hires before they start.

If exits cluster under particular team leaders, look at span of control and coaching time. Team leaders running 20 or more agents with no protected coaching hours tend to lose more people.

If attrition tracks with tenure milestones, look at pay progression. Agents who see no raise path between month 3 and month 18 often leave for a competitor offering a small increase. Track all of this monthly in an attrition tracker so you can tie changes to results.