Key Takeaways
- Customer support agent attrition averages 38 to 45 percent annually industry-wide in 2026, a rate roughly three to four times higher than the broader US workforce
- Early attrition during the first 90 days accounts for 20 to 30 percent of annual departures at most contact centers, making the onboarding window the highest-risk period
- Voluntary attrition makes up 70 to 80 percent of total departures; the most common cited reasons are burnout, pay, and lack of career growth
- Replacing a single agent costs between $10,000 and $20,000 in direct and indirect costs; a 100-seat center with 40 percent attrition carries an annual burden of $400,000 to $800,000
- Contact centers with remote or hybrid work, structured career pathing, and AI-assisted coaching consistently post attrition rates below 25 percent
Customer support consistently ranks among the highest-attrition functions in the modern business. Annual departure rates that would signal a management crisis in most departments are treated as baseline operating conditions in contact centers. The question most operations leaders are quietly asking is not whether attrition is a problem, but how bad their numbers are relative to everyone else's.
What attrition means in customer support and how it is measured
Attrition and turnover are used interchangeably in most industry reports, but there is a useful technical distinction. Attrition refers to the rate at which agents leave a team over a defined period, regardless of whether those positions are backfilled. Turnover refers specifically to the replacement cycle. In contact center operations the difference is often academic, since most vacancies are filled, but the distinction matters when headcount is shrinking by design.
The standard formula for annual attrition rate:
Annual attrition rate = (number of agent departures in 12 months / average headcount during the period) x 100
A 100-agent center that loses 40 agents over a year, replacing them as they go, posts a 40 percent attrition rate. If average headcount was 95 due to ongoing vacancies, the rate is calculated against that lower base, so the number shifts slightly.
Most benchmarking surveys use this formula or a close variant, which makes cross-company comparison reasonably reliable. Where comparisons break down is in how organizations classify involuntary departures, how they handle temporary or seasonal staff, and whether they count agents who leave before completing training.
2026 attrition rate benchmarks
Industry-wide, customer support agent attrition runs between 38 and 45 percent annually in 2025 and 2026. That range comes from ICMI's Contact Center Benchmark Report (2025), Deloitte's Global Contact Center Survey, and multiple workforce management studies covering US and global operations.
For context, the Bureau of Labor Statistics Employee Tenure Summary (January 2024) puts median tenure for service occupation workers at 2.7 years across the broader labor market. Customer support agents fall well below that, averaging 14 to 15 months before departure, based on ICMI workforce data.
The industry spread is wide. Operations at the low end post rates below 20 percent annually. Operations at the high end, typically under-resourced voice centers in US metro labor markets, run 60 percent or above. The average of 38 to 45 percent masks those extremes.
Attrition benchmarks by industry segment
Different support environments carry different stress profiles, and attrition follows stress closely.
| Industry Segment | Annual Attrition Range | Key Driver |
|---|---|---|
| Financial services and banking | 48 to 62% | Compliance pressure, high-conflict interactions |
| Healthcare payer and provider support | 47 to 58% | Emotionally charged calls, regulatory complexity |
| Insurance claims and billing | 42 to 52% | Claims disputes, performance pressure |
| Retail and e-commerce | 40 to 52% | Seasonal volume spikes, peak-period stress |
| Technology and SaaS | 32 to 42% | Higher pay partially offsets stress |
| Telecom and utilities | 38 to 48% | High call volume, billing escalations |
| Government and public sector | 24 to 34% | Job security, lower performance pressure |
| BPO and outsourced centers (all sectors) | 25 to 40% | Varies significantly by provider quality |
Sources: ICMI Contact Center Benchmark Report 2025, Deloitte Global Contact Center Survey 2023, Insignia Resources Contact Center Workforce Study 2025.
The financial services and healthcare segments sit at the top because a meaningful share of their call volume involves conversations that are acutely stressful for callers, and that stress transfers. Agents handling denied claims, billing disputes, or coverage questions navigate emotional intensity on almost every interaction. The cumulative load shows up in departure data faster than it does in lower-stakes environments.
Technology and SaaS support roles benefit from better relative pay and often attract agents with stronger market alternatives, which means compensation has to compete. The attrition rates in that segment are lower partly because the pay structure is more competitive, not because the work is inherently less demanding.
Voluntary versus involuntary attrition
Not all departures are the same. How an agent leaves affects both what the number tells you and what you can do about it.
Voluntary attrition covers resignations, and it accounts for the vast majority of departures in customer support. ICMI's 2025 benchmark data puts voluntary attrition at 70 to 80 percent of total agent departures across the industry. Agents are leaving, not being let go.
Involuntary attrition covers performance-based terminations, attendance-related separations, and role eliminations. In most contact centers, this runs between 20 and 30 percent of total departures.
The distinction matters for diagnosis. High voluntary attrition that concentrates in the first 90 days suggests an onboarding or job preview problem. High voluntary attrition among agents with one to two years of tenure typically points to compensation, career path, or burnout issues. High involuntary attrition often signals a recruiting and selection problem, where organizations hire people who cannot meet performance standards. That is frequently a symptom of below-market wages driving down candidate quality.
What departing agents report as their reasons for leaving
Exit survey data compiled by workforce analytics firms and industry research consistently surfaces a short list of departure drivers. The top reasons agents give, in approximate order of frequency:
- Burnout and stress (cited by 60 to 67 percent of voluntary leavers, per Deloitte and CustomerThink workforce studies)
- Below-market or non-competitive compensation (cited by approximately 48 percent)
- No clear path for advancement or career growth (cited by approximately 42 percent)
- Poor relationship with direct supervisor or inadequate management support (cited by approximately 37 percent)
- Schedule inflexibility or lack of remote work options (cited by approximately 35 percent)
- Dissatisfaction with workload volume or intensity (cited by approximately 30 percent)
These percentages reflect self-reported exit data and should be read as directional rather than precise. Agents sometimes give socially acceptable reasons on exit surveys rather than their primary motivation. The pattern is consistent across multiple independent surveys over several years, though. Burnout and compensation show up at the top everywhere.
Early attrition: the 90-day problem
First-year attrition is the most expensive attrition profile because it captures the full cost of recruiting and training without recovering any meaningful productivity. The problem concentrates even earlier than the first year.
SHRM's 2024 workforce research found that 20 to 30 percent of annual agent departures occur within the first 90 days. ICMI's contact center data puts the first-year departure rate between 65 and 70 percent at a significant share of centers, meaning most new hires who will eventually leave do so before their first anniversary.
The first 30 days carry the highest single-month exit rate.
Job preview mismatch
Agents discover in week one or two that the actual role differs from what they understood during hiring. Call volume, emotional intensity, monitoring environment, performance expectations, all of it tends to be underrepresented in standard job descriptions. Recruiters who use sanitized postings to fill seats faster create an attrition timing problem rather than preventing one.
Onboarding structure gaps
Agents who complete a structured, coach-supported onboarding process depart at significantly lower rates than those placed on the floor with minimal guidance. SHRM's 2024 onboarding research found that agents who go through formal structured programs are 58 percent more likely to remain past three years. The difference between structured and informal onboarding is not just knowledge transfer. It is whether the new agent feels supported during the highest-stress stretch of their tenure.
Immediate performance pressure
Some centers move new agents onto live queues with full performance accountability too quickly. Agents who struggle before building confidence tend to leave before completing the learning curve rather than pushing through discomfort.
Early self-selection
Early attrition often reflects an accurate personal assessment. Agents who discover the culture, pace, or management style is not compatible with what they want leave quickly and voluntarily. That is not always a problem to solve. Getting the genuinely wrong-fit agents out early is less costly than retaining them long enough to transfer disengagement to colleagues or customers.
Attrition by operation type and geography
The operation type and location of a contact center significantly affect the baseline attrition rate, independent of management quality.
US-based operations
In-house US contact centers running primarily voice channels in metro labor markets post some of the highest attrition rates in the industry. Metro areas have more employment alternatives available to support agents. The BLS median annual wage for customer service representatives of $39,680 (May 2024 OEWS data) leaves US-based agents with real alternatives in warehouse, logistics, and gig economy roles, particularly as those employers have raised hourly rates.
Remote US operations consistently post lower attrition than on-site equivalents. Insignia Resources found remote contact center setups post attrition rates 15 to 20 percentage points lower than comparable on-site operations. Opening up the geographic hiring pool lets employers reach labor markets where the competitive wage is more favorable relative to local alternatives.
Outsourced and BPO operations
| Operation Type | Typical Annual Attrition |
|---|---|
| In-house US on-site center | 40 to 55% |
| In-house US remote/hybrid center | 22 to 35% |
| Philippines large-scale BPO | 28 to 38% |
| India Tier-1 BPO | 25 to 35% |
| Nearshore Latin America BPO | 15 to 28% |
| Caribbean nearshore BPO | 15 to 25% |
Sources: ICMI, Insignia Resources, Working Solutions, BPO industry surveys.
Well-managed outsourcing providers consistently post attrition 25 to 35 percent lower than industry-average in-house operations at comparable scale. Established BPO providers have built retention infrastructure over years, including defined career tracks, management-to-agent ratios focused on retention outcomes, and workforce flexibility tools most in-house operations have not invested in.
Geographic concentration also drives the lower rates in nearshore markets. Support roles offer meaningfully better relative compensation in Caribbean and parts of Latin American labor markets than the alternatives available locally. An agent in San José, Costa Rica or Bridgetown, Barbados faces different outside options than an agent in Dallas or Phoenix, and attrition reflects that difference.
What attrition costs a contact center
Most contact center cost models understate attrition cost by capturing only visible direct expenses and missing the productivity drag.
Direct cost per departure
| Cost Category | Estimated Range | Source |
|---|---|---|
| Job posting and recruitment advertising | $500 to $1,500 | SHRM 2024 |
| Recruiter and interviewer time | $750 to $2,000 | Internal labor estimates |
| Background check and pre-employment screening | $200 to $500 | Industry benchmarks |
| Onboarding administration | $300 to $800 | SHRM 2024 |
| Training program delivery | $1,000 to $2,000 | ICMI / contact center surveys |
| Total direct cost range | $2,750 to $6,800 |
SHRM's 2024 cost-per-hire benchmark for customer service roles runs $2,250 to $4,683, consistent with the direct cost range above when training delivery is included.
Indirect and productivity costs per departure
New agents reach baseline productivity for complex queues in six to eight months, per ICMI and contact center workforce research. During the ramp period, average handle times are longer, first-contact resolution rates are lower, and escalation rates are higher. The productivity gap between a new agent and a fully experienced one costs the business even when the new agent's salary is lower.
The estimated productivity drag during ramp runs $3,000 to $8,000 per agent, depending on queue complexity.
Combined, total cost per agent departure runs $10,000 to $20,000 under conservative assumptions. Workforce analytics firms that include customer satisfaction impact and increased repeat-contact rates estimate costs reaching $46,000 per departure for complex-queue roles.
Annual cost for a 100-seat center
| Attrition Rate | Annual Departures | Cost at $12,000/departure | Cost at $20,000/departure |
|---|---|---|---|
| 25% (strong performer) | 25 | $300,000 | $500,000 |
| 40% (industry average) | 40 | $480,000 | $800,000 |
| 55% (high attrition) | 55 | $660,000 | $1,100,000 |
The gap between a 25 percent and 55 percent attrition rate in a 100-seat center is between $360,000 and $600,000 annually in replacement costs alone, before any customer satisfaction or revenue impact is counted.
Leading indicators of attrition risk
Most attrition is predictable before an agent gives notice. Workforce analytics platforms and manager observation data consistently identify the same early warning signals.
A previously consistent agent whose daily handled contacts, resolution rates, or quality scores begin declining without an obvious external cause is showing a classic pre-departure pattern. The decline typically precedes resignation by four to eight weeks.
Increased schedule adherence issues are another clean signal. Agents who begin arriving late, leaving early, or taking unexplained absences at higher frequency are signaling disengagement. Schedule adherence data is one of the most reliable leading indicators available in contact center environments because it is objective and tracked automatically.
Quality monitoring scores that trend downward for an agent who previously performed adequately often reflect declining motivation rather than skill regression.
In environments with active team channels, coaching participation, or voluntary training, agents who withdraw from optional engagement before formally resigning are often nearing an exit decision.
If your center has access to voluntary quit data by tenure cohort, mapping departures by months-of-service reveals when agents are most vulnerable. Many centers see pronounced attrition peaks at 6 months (often tied to initial job satisfaction reassessment) and 18 to 24 months (tied to compensation and career path reassessment).
A manager who acts on these signals before a resignation, through direct conversation, scheduling adjustment, or pay review, has a window to retain the agent. That window closes once the departure decision is made.
Factors that separate low-attrition from high-attrition operations
Operations that consistently post below-25-percent attrition share identifiable characteristics.
Remote and hybrid work
The single retention lever with the clearest payoff in the data. Remote arrangements reduce attrition by 15 to 20 percentage points, per Insignia Resources, and preference surveys find 81 percent of support agents prefer remote work. Centers that built remote infrastructure before it was forced on them now have a structural retention advantage over competitors who are still assembling on-site.
Structured onboarding with active supervisor engagement
Agents who go through formal 90-day onboarding programs with dedicated check-ins and performance coaching are 58 percent more likely to remain past three years (SHRM 2024). The difference between structured and informal onboarding is not just knowledge transfer. It is whether the new agent feels supported during the highest-stress period of their tenure.
Clear career advancement tracks
Deloitte's workforce research found organizations with active career pathing and skills development investment post 15 percent lower attrition than comparable operations without those programs. Agents who see a defined path, agent to specialist to team lead, or into QA, training, or workforce management, treat the role as a beginning rather than a ceiling.
Competitive compensation benchmarked regularly
Pay is not the primary lever, but it is a prerequisite. Agents whose compensation falls behind market alternatives notice over time. The frustration from being underpaid accelerates decisions that burnout alone might not produce. Regular market benchmarking and structured pay progression keep compensation from becoming an active departure driver.
AI-assisted coaching tools
Pilot deployments of real-time AI coaching tools for agents have shown attrition reductions of approximately 12 percent over three-month windows in Metrigy's 2025 contact center AI workforce study. The mechanism is reducing the feeling of isolation and inadequate support during difficult interactions. Agents with real-time guidance on complex calls report higher confidence and lower end-of-shift stress. One documented deployment at a mid-sized contact center attributed $1.5 million in annual attrition cost savings to AI coaching tool adoption.
Workload management and intensity controls
Contact centers that actively manage daily contact volume per agent, enforce mandatory break schedules, and use routing logic to distribute emotionally intense calls across the team rather than concentrating them on the same agents post lower burnout-driven attrition than operations that maximize utilization at the expense of agent experience.
2026 attrition benchmarks summary
| Metric | High Attrition (Bottom Quartile) | Industry Average | Strong Performer (Top Quartile) |
|---|---|---|---|
| Annual attrition rate | Above 55% | 38 to 45% | Below 25% |
| First-year attrition | Above 70% | 60 to 70% | Below 40% |
| 90-day attrition rate | Above 30% | 20 to 30% | Below 12% |
| Average agent tenure | Under 10 months | 14 to 15 months | 24+ months |
| Voluntary attrition share | 75 to 85% | 70 to 80% | 60 to 70% |
| Annual attrition cost per 100 seats | $600,000 to $1.1M+ | $400,000 to $800,000 | Under $300,000 |
Implications for customer support planning
If your attrition rate sits in the average or high range, the data on departure drivers and leading indicators points toward where the pressure is concentrated and which interventions have the strongest track record.
Early attrition is the highest-yield target. Departures in the first 90 days carry the full cost of recruiting and training with essentially no productivity return. Improving onboarding structure and job preview accuracy pays back faster than almost any other retention investment because it attacks the highest-cost attrition segment.
Retention investment consistently beats replacement cost on the numbers. Spending $500 to $1,000 per agent annually on retention infrastructure, including remote work support, career development programs, and coaching tools, recovers two to five times that amount in avoided replacement costs under conservative estimates. The challenge is that retention investment is a diffuse ongoing cost while replacement spending hits as discrete events. Operations leaders who cut retention programs in lean periods typically see the replacement costs materialize six to twelve months later.
The outsourcing attrition premium is real. For organizations running in-house support at 40 to 55 percent attrition, moving to a well-managed outsourced model at 25 to 35 percent attrition produces a cost reduction that compounds annually. For a 100-seat operation, closing 20 percentage points of attrition gap at $15,000 per departure is $300,000 per year in avoided replacement costs, before accounting for quality improvements from lower ramp-period volume.
For deeper context on how attrition rates connect to support team economics, see our customer support agent productivity statistics 2026 and customer support agent turnover statistics 2026 research pages. For organizations considering outsourced models as a response to in-house attrition, our virtual assistant services page covers how distributed support staffing works in practice, including the workforce management structures that drive the retention advantage.
Data sources and methodology
The statistics in this article draw from publicly available workforce research, industry surveys, and labor market data published between 2023 and 2026.
Primary sources:
- Bureau of Labor Statistics, Occupational Employment and Wage Statistics (OEWS), May 2024 release
- Bureau of Labor Statistics, Employee Tenure Summary, January 2024
- ICMI, Contact Center Benchmark Report, 2025
- Deloitte, Global Contact Center Survey, 2023 (data referenced through 2025 in follow-up research)
- SHRM, Cost-Per-Hire Benchmarking Report, 2024
- SHRM, Employee Onboarding Research, 2024
- Insignia Resources, Contact Center Attrition and Remote Work Study, 2025
- Metrigy, Contact Center AI and Workforce Impact Study, 2025
- Gartner, Customer Service and Support Workforce Predictions, April 2026
- McKinsey and Company, The State of Customer Care, 2024
- CustomerThink, Agent Experience and Burnout Survey, 2024
- Five9 and ICMI, Contact Center Workforce Management Survey, 2024
- Calabrio, State of the Contact Center Report, 2025
- NICE inContact, Global Customer Experience Benchmarking Report, 2025
- Working Solutions, Contact Center Cost Benchmarks, 2026
- HDI (Help Desk Institute), Technical Support Workforce Survey, 2025
- CallMiner, The CallMiner Churn Index, 2024
- Workforce Institute at Kronos (UKG), Employee Retention and Attrition Report, 2024
Frequently Asked Questions
What is a good customer support agent attrition rate?
A strong-performing contact center in 2026 posts annual attrition below 25 percent. The industry average runs 38 to 45 percent. Operations that consistently land below 25 percent typically combine remote or hybrid work options, structured onboarding programs, and clear career advancement tracks. Attrition below 20 percent is achievable but requires deliberate investment in agent experience rather than simply competitive compensation.
What causes high attrition in customer support?
The three primary drivers of voluntary attrition in customer support are burnout and job stress, below-market compensation, and lack of career growth. Exit survey data places burnout at the top of the list, cited by 60 to 67 percent of voluntary leavers. The second tier includes inadequate management support and scheduling inflexibility. Operations that address burnout mechanisms first, workload intensity, remote work access, coaching support, see greater retention improvement than operations that lead with compensation increases alone.
How do outsourced support teams compare on attrition?
Well-managed outsourced and BPO operations consistently post attrition 25 to 35 percent lower than in-house US contact centers at comparable scale. The advantage is structural: established providers have built retention infrastructure, defined career tracks, and management-to-agent ratios focused on retention outcomes. Nearshore BPO operations in Latin America and the Caribbean often post attrition rates of 15 to 28 percent, partly because support roles offer stronger relative compensation in those labor markets.
What is the cost of customer support agent attrition?
Each agent departure costs an estimated $10,000 to $20,000 in direct and indirect costs, including recruiting, onboarding, training, and the productivity drag during the new hire ramp period. For complex roles in high-volume environments, total cost per departure can reach $46,000 when customer satisfaction impact and increased repeat-contact rates are included. A 100-seat contact center at 40 percent annual attrition carries an annual replacement burden of $400,000 to $800,000.