Call center turnover rate: Benchmarks, causes, and how to reduce it
A 500-seat contact center that loses 40% of its agents each year must hire and ramp 200 new people to keep its headcount steady. Those hires spend weeks in a classroom and months below full proficiency. Customers feel that stretch as longer calls and repeat contacts.
This guide is for enterprise contact-center CX and operations leaders who need to measure that problem and act on it. A call center turnover rate only supports a benchmark comparison once you know who was surveyed, when, and how you defined turnover. Attrition has a cost, agents leave for documented reasons, and only some reduction tactics have measured evidence behind them.
What is call center turnover rate?
Call center turnover rate is the share of agents who leave the organization during a defined period, expressed as a percentage of average headcount for that period. The standard turnover formula (opens in a new tab) is:
Turnover rate = (separations during the period ÷ average number of employees during the period) × 100
Contact-center teams (opens in a new tab) often call the same metric attrition: the rate at which staff voluntarily leave the workforce over a given period, also known as staff turnover. Some count only voluntary leavers and others count every separation. Counting every separation (opens in a new tab) includes agents who leave the workforce whether they exit internally or externally, so a promotion into another department adds to the turnover figure.
Check: before you compare your rate to any benchmark, confirm whether the source counted voluntary leavers only, all separations, or internal moves too.
Why agent turnover deserves CX leaders’ attention
Turnover lowers first contact resolution (FCR) (opens in a new tab), lengthens average handle time (AHT), drags down customer satisfaction (CSAT), and raises cost per contact. Each effect starts with the same cause: replacement agents who are still learning the job.
Lower first contact resolution
New agents resolve fewer issues on the first contact, and the gap lasts for months. A replacement agent is still learning the products, the systems, and the policy exceptions that experienced agents handle without escalating. Until new agents catch up, more customers call back about the same issue, and the remaining team absorbs that extra volume.
Longer handle times
New agents take longer on every contact because they search for answers experienced agents already know. Contact Center Pipeline notes that new hires bring longer call-handling times and higher error rates (opens in a new tab) while they adjust. Longer calls mean fewer contacts handled per hour, so queues grow even when headcount is back to plan.
Lower customer satisfaction
Customers feel the learning curve as slower answers and more transfers, and satisfaction drops with them. In a Metrigy study, 97% of contact center leaders (opens in a new tab) said agent tenure directly affects customer satisfaction, and 46% reported higher CSAT from longer-tenured agents. Constant turnover leaves too few tenured agents to hold those scores.
Higher cost per contact
Each of these effects raises cost per contact (opens in a new tab). Repeat calls add volume, longer calls need more paid agent hours for the same work, and every departure adds hiring and training costs on top. A center with high turnover pays more for each resolved issue, even before you factor in the cost of replacing agents.
Check: split your CSAT, FCR, and AHT by tenure band (0–90 days, 3–12 months, over 12 months) to see how much of your performance gap sits with recent hires.
How to calculate turnover and attrition rate
Calculate turnover with one formula, one period, and one written rule for who counts as a separation; then apply it the same way every month. The worked example below uses a 500-seat center in a single month:
Agents on payroll at the start of March: 505
Agents on payroll at the end of March: 495
Average headcount: (505 + 495) ÷ 2 = 500
Agents who left in March: 15 (12 resigned, 3 were dismissed)
Monthly turnover is 15 ÷ 500 × 100 = 3.0%. Voluntary turnover is 12 ÷ 500 × 100 = 2.4%, and involuntary turnover is 3 ÷ 500 × 100 = 0.6%.
Compare annual and monthly rates on the same time scale, and separate voluntary from involuntary exits using a consistent definition.
Calculating annual vs monthly rates
A 3% monthly rate is not a 3% annual rate. To produce an annual figure comparable to contact-center benchmarks, divide the year’s total separations by average headcount for the year. If the same center loses 180 agents across 12 months on an average headcount of 500, annual turnover is 36%. A Bureau of Labor Statistics (BLS) release notes that an annual rate built this way is approximately equal to the sum (opens in a new tab) of the 12 monthly rates.
Federal data use a different convention that trips up many comparisons. The BLS Job Openings and Labor Turnover Survey (JOLTS) calculates its annual average rates (opens in a new tab) by dividing the sum of 12 monthly separation levels by the sum of 12 monthly employment levels, which keeps the result on a monthly scale. The 2025 annual average total separations rate for all US nonfarm jobs was 3.3 percent (opens in a new tab), with quits at 2.0 percent. Neither figure is comparable to an annual contact-center turnover rate of 30%.
Where turnover concentrates
A single turnover rate hides where agents actually leave. Breaking it down by segment, by exit type, and by tenure shows which centers, roles, and hiring classes drive the number, and whether the fix belongs in hiring, onboarding, or retention.
Turnover by segment
Turnover varies sharply by who runs the center, the agent’s role, and how the job is designed.
By ownership: Outsourced centers sit at the high end. ContactBabel’s US outsourcing research states that outsourcing attrition rates tend to be two or three times higher (opens in a new tab) than the US industry average. COPC’s 2022 outsourcing benchmark cites a median of 87% agent attrition (opens in a new tab), drawn from its certification programs.
By role: Mercer’s 2016 US Contact Center Survey, as reported by Contact Center Pipeline (opens in a new tab), found turnover of 27% for entry-level agents, 12% for senior agents, 11% for team leaders, and 6% for team or group managers.
By job design: The Global Call Center Report found typical turnover of 9% where jobs are high quality (opens in a new tab), meaning very high-quality jobs with high agent discretion and low monitoring. Turnover is 36% in centers with low-quality jobs, which give agents low discretion and high monitoring.
Check: compare an outsourced partner’s rate against outsourced benchmarks and your in-house sites against in-house benchmarks, never the two against each other.
Voluntary vs involuntary attrition
Tracking voluntary and involuntary exits separately shows whether you have a retention problem or a hiring problem. The JOLTS separation definitions (opens in a new tab) give the clearest official split, sorting leavers into three groups:
Quits: Employees who left voluntarily, excluding retirements and transfers to other locations.
Layoffs and discharges: Involuntary separations initiated by the employer, including firings for cause and terminations of seasonal or short-term employees.
Other separations: Retirement, death, disability, and transfers to other locations of the same firm.
In contact centers, the split runs close to even. ContactBabel’s US guide reported that the voluntary share of attrition (opens in a new tab) had a median of 52% and a mean of 47%, excluding agents who move into another role in the organization. A center that treats its whole rate as a retention problem may be misreading half of it. Resignations point to pay, workload, and supervision. Dismissals and other separations point to selection and performance management.
Check: split every rate you report into quits, dismissals, and other separations, using the same categories each month.
Benchmarks and what counts as a high rate
Published averages and medians for US contact centers put annual agent turnover between roughly 24% and 38%. Outsourced centers run well above that. Each figure below comes with its population, period, and definition as far as the source states them:
Source | Figure | Population | Data period |
|---|---|---|---|
SQM Group, via No Jitter (opens in a new tab) | 35% (2021), 38% (2022) | Over 500 North American call centers benchmarked annually | 2021 and 2022 |
ContactBabel 2024 US Decision-Makers’ Guide (opens in a new tab) | 31% mean, 24% median | 189 US contact center managers, in-house and outsourced pooled | Year-end 2023 |
ContactBabel analyst Steve Morrell, to No Jitter (opens in a new tab) | About 30% average | US contact centers; sample not stated | Interview, November 2025 |
Gartner, via CX Today (opens in a new tab) | 25% median | Contact center agents; sample not stated | March 2022 benchmark |
Up to 60%; nearly 20% at best-in-class companies | Most contact centers, based on McKinsey’s own experience | Not stated |
ContactBabel analyst Steve Morrell put US agent attrition at about 30% on average (opens in a new tab) in November 2025 and described that level as standard for the past seven or eight years, so recent US figures cluster around 30% and have held steady. Rates vary widely by sector and region.
Check: if your annual rate is above 30%, you sit above the ContactBabel 2024 median of 24%; about a third of that pooled in-house and outsourced sample reported rates above 30%.
Use segment-specific benchmarks and track voluntary, involuntary, and early-tenure exits separately to identify which departures need attention.
What agent turnover costs a company
Gartner estimates that each agent who leaves costs an average of $14,113 to replace (opens in a new tab). That figure covers three cost components:
Hiring: Recruiting and selecting a replacement.
Onboarding: Training the new hire before they take calls independently.
Lost productivity: The months a new agent works below full proficiency.
At 40% annual turnover, a 500-seat center replaces 200 agents a year, about $2.8 million at Gartner’s average (200 × $14,113), a figure to rebuild with your own cost components, period, and measure before you rely on it.
Why call center agents leave
Agents most often leave because of low pay, limited career growth, heavy workloads and burnout, and poor supervision. Each reason below carries one data point.
Low pay: In a 2023 Call Centre Helper LinkedIn poll of 433 industry professionals, 46% chose low wages (opens in a new tab) as the top attrition reason.
Limited career growth: In the same poll, 39% chose limited career growth.
Heavy workload and burnout (opens in a new tab): CX Today reports BenchmarkPortal research naming increasing workload as the most significant attrition driver (opens in a new tab), with agents handling 7.07 calls per hour in 2021 compared with 6.07 two years earlier.
Poor supervision: McKinsey’s contact-center engagement study found dissatisfaction with the immediate supervisor (opens in a new tab), including the coaching process, was the number one driver of attrition.
These reasons overlap, so exit-interview data alone rarely shows the whole picture.
Check: compare your exit-interview reasons with a short stay survey of current agents.
How to reduce agent turnover
The tips below work on who you hire, the workload and pay agents get, their room to grow, and the support they get from supervisors. Each one ends with a signal that tells you whether it is working.
1. Test candidates for job fit before you hire
Add a structured job-fit test to your screening process, and score every candidate against the same criteria before making an offer. Choosing for fit up front keeps more of the people you hire, because a mismatch often ends in an early exit.
A team of researchers found that introducing job tests improves worker retention (opens in a new tab) across 15 firms employing low-skill workers. The result gives a direction without a figure you can promise to your own center. To know it is working, compare 90-day attrition of tested hiring classes with earlier classes, and check that the difference holds across several classes before you credit the test.
2. Show candidates the real job before they accept
Give candidates a realistic preview of the work, and pair it with a short step that lowers unrealistic expectations, such as a candid talk about difficult calls and shift patterns. Candidates who expect a different job have more reason to leave once reality arrives.
In a telemarketer field experiment (opens in a new tab), the preview combined with an expectation-lowering procedure produced the most days worked, and the preview alone did not differ from the control group, so use both together. Track days worked and 90-day attrition by hiring class, and review how often exit reasons cite the job not being as expected.
3. Pay new agents at or above the local market
Check your starting pay against current local job postings, and move it to the market rate or above. Pay is among the reasons managers cite most, and a low offer gives new agents a quick reason to leave.
ContactBabel found that centers with under 10% three-month attrition pay new agents an average of 9% more (opens in a new tab) than centers above 10%. This is an association, so it does not show that higher pay alone lowers attrition. Repeat the posting check each quarter and watch three-month attrition alongside it; the tip works when your starting pay holds steady against the market and early exits fall.
4. Put supervisors on the floor to coach
Free supervisors from desk work so they spend most of their time on the floor, and build floor coaching (opens in a new tab) into the weekly routine. In-the-moment coaching gives each agent regular attention from the person who shapes their day. Track supervisor floor time against team-level attrition, and see whether teams with more floor time keep more agents.
5. Build career paths agents can see
Define visible career steps, such as senior agent and team lead, and publish the criteria for moving up each one. Limited career growth is one of the most cited exit reasons, and agents who can see the next step have a reason to stay long enough to reach it.
Review the steps with agents and supervisors once a year to keep the criteria realistic. To see whether it’s working, track the internal promotion rate and review stay-survey answers on growth, which should show agents who can describe their next step.
6. Offer remote and hybrid schedules agents can plan around
Once agents are onboarded, let them work from home or on a predictable hybrid pattern, and keep initial training and nesting in person so new hires connect with their team. Predictable schedules let agents plan their lives around work, which gives them one less reason to leave. In a randomized experiment, call center agents who worked from home had 17% attrition over nine months (opens in a new tab), compared with 35% for office workers.
Check: track remote and on-site hires as separate cohorts and compare their attrition and productivity.
7. Use AI agents to take repetitive calls off agents’ queues
Route routine first-layer calls, such as order status checks or appointment changes after verified identity (opens in a new tab), to AI agents, and make sure every escalation hands the human agent the full transcript and context so nobody restarts the conversation. This removes the most scripted, repetitive work from the queue, though no study shows it reduces turnover.
The International Customer Management Institute reports that as AI takes easier calls, agents are reporting higher stress (opens in a new tab) and attrition is rising, so the remaining human workload is harder and needs support. Before and after rollout, track human-agent average handle time, escalation volume per agent, and tenure-segmented attrition. Judge the AI on resolution, first contact resolution, and containment.
8. Avoid these common retention mistakes
Four measurement mistakes can hide whether your retention efforts work:
Treating containment as the only metric: Track resolution, first contact resolution, customer satisfaction, and human workload alongside containment rate (opens in a new tab).
Ignoring early-tenure attrition: Report 90-day attrition by hiring class, including people who leave during training and nesting.
Conflating voluntary and involuntary turnover: Split the rate by separation type, because dismissals point to selection and resignations point to pay and supervision.
Comparing outsourced and in-house rates directly: Benchmark each against its own segment.
At Gartner’s estimate of $14,113 per replacement, each prevented exit saves real money. Pilot each tip with a comparison group, and use the measures in the next section to see which ones work.
How to measure agent turnover
Measuring comes before fixing, because without a baseline you cannot tell which tactics work.
Annual and monthly turnover rate: Separations ÷ average headcount × 100, with a written rule on training dropouts, contractors, and internal transfers. Check monthly and roll the result up annually.
Early attrition: The share of each hiring class that leaves within its first 90 days, including training and nesting dropouts. Check for every hiring class.
Replacement cost per agent: Departures multiplied by Gartner’s $14,113 average, or your own figure. Check quarterly.
Exit and stay-interview reasons: The categorized reasons leavers and current agents give for leaving or staying. Check quarterly.
Ramp time to full productivity: The weeks until a new agent matches tenured agents on resolution and handle time. Check for every hiring class.
Review the five measures together each month so a change in one shows up against the others.
Make retention a measured part of contact center operations
Turnover costs show up in resolution, handle time, and budget. Retention therefore needs the same measurement and clear ownership as any other key performance indicator. Teams that review it monthly can act before costs compound.
Parloa (opens in a new tab) is an AI agent management platform that enables AI agents to take routine call volume off human agents’ queues and hand off escalations with the full transcript and context, so human agents don’t restart conversations. Parloa Lens monitors service quality across live conversations.
Lighter queues leave human agents more time for complex calls, where their skills count most. Book a demo (opens in a new tab) to see how AI agents can take repetitive calls off your human agents’ queues.
FAQ
How do you calculate call center turnover rate?
Divide the number of agents who left during a period by the average headcount for that period, then multiply by 100. A center that loses 180 agents in a year on an average headcount of 500 has an annual turnover rate of 36%. State whether training dropouts, contractors, and internal transfers count.
What is a high call center turnover rate?
For US contact centers, an annual turnover rate above 30% is high, since ContactBabel’s 2024 US Decision-Makers’ Guide puts the median at 24% (opens in a new tab) and the mean at 31%. Outsourced centers run considerably higher, so judge them only against outsourced benchmarks.
How much does it cost to replace a call center agent?
Gartner estimates that replacing a contact center agent costs $14,113 on average (opens in a new tab). The figure covers hiring, onboarding, and the lost productivity while the new agent ramps up. At that rate, a 500-seat center losing 200 agents a year faces about $2.8 million, so rebuild the estimate with your own wage, training, and ramp data.
What are the main causes of agent turnover?
Managers and industry polls most often cite low pay and limited career growth. Agent-level studies add stress, intensive monitoring, scripted and repetitive work, and dissatisfaction with supervisors. Rising workload and absenteeism also appear as drivers.
Why track voluntary and involuntary turnover separately?
Voluntary and involuntary exits have different causes, so they need different fixes. Resignations usually point to pay, workload, and supervision. Dismissals point to hiring and performance management, and a single blended rate hides which problem you have.
:format(webp))
:format(webp))