The 15 essential contact centre KPIs fall into four pillars: accessibility (service level, abandonment rate), efficiency (FCR, AHT, repeat contact rate), quality and customer experience (CSAT, NPS, quality compliance), and employee engagement (turnover, absenteeism, EX score). The core principle: a maximum of 10 KPIs actively managed at operational level. The rest stay on monthly review.
You’re tracking 30 indicators. Your managers spend hours filling in spreadsheets. And yet you feel like you’re flying blind. The problem isn’t a lack of data, it’s too much of it. The best-performing contact centres don’t track more KPIs than everyone else. They track the right ones, with a clear methodology and concrete action plans.
Here are the 15 truly essential KPIs, each with a field definition, calculation formula, observed benchmarks, and immediate actions to put in place.
An indicator that doesn’t generate action is a useless indicator. Before adding a KPI to your dashboard, check it meets four criteria: it’s understood by frontline teams, reliable and measurable without guesswork, actionable with identified levers, and aligned with both customer experience and economic performance. If any one of these criteria is missing, the indicator has no place in your day-to-day management.
Service Level measures the percentage of calls answered within a target time frame. It’s the reference accessibility indicator in a contact centre, also known as the answer rate or SLA.
Formula:
Service Level = (Calls answered within target time / Inbound calls) × 100
Reference standard: 80% of calls answered in under 20 seconds (80/20)
A weak Service Level signals understaffing and generates customer frustration. But be careful: it’s an accessibility indicator, not a quality one. Slightly relaxing the SLA to give advisors more time can significantly improve FCR.
| Context | Benchmark |
|---|---|
| Market standard | 80/20 |
| Premium services | 90/15 |
| Technical support or back office | 70/30 |
Concrete actions: recalibrate your SLA by contact type (product information vs complex complaint), analyse abandonment after 30 seconds rather than overall SLA compliance, and implement dynamic WFM with intra-day adjustments.
The abandonment rate measures the percentage of callers who hang up before being handled. It’s an indicator of lost contacts and customer frustration. A high rate signals capacity disorganisation or an unmet customer promise.
Formula:
Abandonment rate = (Abandoned calls / Inbound calls) × 100
| Level | Interpretation |
|---|---|
| Below 5% | Very good level |
| 5 to 8% | Acceptable, to monitor |
| Above 10% | Immediate operational alert |
A minute-by-minute analysis of abandonment often reveals that most of it happens after 45 seconds, when the IVR message becomes too long or repetitive. Shortening these messages delivers fast, measurable results.
Concrete actions: map the exact moment of abandonment (not just the overall rate), limit your IVR messages to 30 seconds before queuing, and systematically offer an automatic callback after 2 minutes of waiting.
FCR measures the percentage of interactions resolved on the first contact, with no need for the customer to call back. It’s the KPI that reconciles quality and performance: a strong FCR means fewer repeat contacts, less customer frustration, and better overall efficiency. Resolving it right the first time costs less than handling it twice.
Formula:
FCR = (Contacts resolved on first contact / Total contacts) × 100
Recommended measurement method: no repeat contact from the same customer on the same issue within 7 days (via CRM), or a direct question at the end of the interaction: “Was your request resolved during this contact?”
| Level | FCR |
|---|---|
| Market average | 65 to 70% |
| High-performing centres | 75 to 85% |
Concrete actions: enrich your knowledge base on the 20 most frequent issues, give advisors more decision-making autonomy (commercial gestures, escalation levels), and identify the 10 issues with the highest repeat contact rate to address as a priority.
AHT measures the average time an advisor spends fully handling a contact, from pickup to closure, including after-call work. It’s used to size teams and build schedules. But it’s also the most misused indicator in contact centres: managed in isolation, it pushes advisors to handle calls quickly at the expense of resolution.
Formula:
AHT = Total handling time (conversation + after-call work) / Number of contacts
| Contact type | Indicative AHT |
|---|---|
| Standard customer service | 4 to 6 min |
| Technical support | 8 to 12 min |
| Insurance back office | 12 to 18 min |
Centres with the lowest AHT often have the highest repeat contact rate. Advisors close quickly, poorly, and customers call back. Never manage AHT globally: analyse it by contact reason. A 3-minute AHT for a billing query is normal. The same AHT for a complex complaint is a red flag.
Concrete actions: break down AHT by contact reason, analyse after-call work time separately (often 30 to 40% of the total), and automate low-value tasks such as sending documents or simple CRM updates.
Occupancy measures the percentage of time advisors spend in productive activity (conversation + after-call work) relative to total paid time. It indicates how efficiently resources are being used. An excessively high rate generates fatigue, errors, and turnover.
Formula:
Occupancy rate = (Productive time / Paid time) × 100
| Level | Interpretation |
|---|---|
| 75 to 85% | Optimal zone, balancing performance and wellbeing |
| Above 90% | Risk of burnout, errors, and turnover |
Slightly reducing occupancy often improves absenteeism and handling quality. 85% occupancy with strong FCR is always better than 95% with a high repeat contact rate.
Concrete actions: build in 5-minute planned micro-breaks every 2 hours, adjust staffing intra-day rather than pushing occupancy up, and don’t confuse occupancy with productivity.
The quality compliance rate measures adherence to defined customer relationship standards: relational posture, accuracy of responses, process compliance, regulatory compliance. Without this KPI, there’s no risk control and no consistency of experience.
An effective quality scorecard covers four dimensions: relational posture (listening, empathy, personalisation), accuracy of responses (regulatory compliance, precision), process adherence (tools, traceability), and effectiveness (resolution, time management). Limit it to a maximum of 5 truly essential criteria.
| Score | Level |
|---|---|
| Below 85% | Insufficient, operational risk |
| 90% and above | Good level |
| 95% and above | Excellence |
Concrete actions: share call listening sessions collectively, not just one-on-one, and consistently link quality compliance to CSAT to identify correlations.
CSAT measures immediate customer satisfaction following an interaction with the contact centre. It’s a hot-take perception indicator that helps quickly identify irritants and friction points in the experience. To go deeper on calculation and improvement levers, see our complete guide to CSAT.
Formula:
CSAT = (Number of 4 and 5 out of 5 responses / Total number of responses) × 100
Sample question: “On a scale of 1 to 5, are you satisfied with your interaction?”
| CSAT score | Level |
|---|---|
| Above 90% | Excellent |
| 85 to 90% | Good level |
| Below 75% | Corrective action needed |
Consistent field observation: satisfaction is more closely tied to resolution than to waiting time. A customer who waits a few minutes and gets a complete solution is generally more satisfied than one who’s handled immediately but poorly resolved.
Concrete actions: analyse verbatims, not just the score (the words reveal the irritants), link CSAT to contact reasons to identify broken journeys, and build action plans by identified irritant rather than generic actions.
NPS measures how likely a customer is to recommend your brand. It assesses the overall relationship, where CSAT measures one-off satisfaction. A customer can be satisfied with an interaction and still not recommend the brand. To understand the differences between NPS, CSAT and CES and when to use each, see our comparative guide to CX metrics.
Formula:
NPS = % Promoters (scores 9 to 10) – % Detractors (scores 0 to 6)
Sample question: “On a scale of 0 to 10, how likely are you to recommend our brand to a friend or colleague?”
Use NPS by customer journey, not just as a global figure. Post-complaint NPS is often 20 to 30 points lower than overall NPS. These gaps point precisely to where to focus efforts. Prioritise the neutrals (scores 7 to 8): they’re the easiest to convert into promoters.
The repeat contact rate measures the percentage of customers who contact the centre again for the same issue within a given period, typically 7 days. It’s the mirror image of FCR: where FCR measures what’s working, the repeat contact rate highlights what’s structurally broken.
Formula:
Repeat contact rate = (Repeat contacts on the same issue / Total contacts) × 100
| Level | Interpretation |
|---|---|
| Below 10% | Good resolution level |
| Above 15% | Structural problem to address |
Key action: map the 3 issues with the highest repeat contact rate and address the root cause (process, training, tools), not the symptom.
Schedule adherence measures advisors’ compliance with planned hours: logging in, logging off, breaks. Without adherence, your WFM forecasts are worthless. An advisor missing from their post at scheduled times throws off the entire schedule and directly degrades customer service.
Formula:
Adherence = (Time actually worked / Planned time) × 100
| Score | Level |
|---|---|
| Above 95% | Very good level |
| Below 90% | Forecast reliability degraded |
Adherence is built through discipline and managerial example, not through sanctions. Real-time monitoring via your WFM tool is essential to catch deviations before they affect service.
Absenteeism measures the percentage of unplanned absence time (sickness, unauthorised absence) relative to planned time. It’s a social warning signal, not just an operational one. High absenteeism almost always reveals underlying problems: workload overload, a deteriorating social climate, weak management. It’s a symptom, not a cause.
Formula:
Absenteeism = (Unplanned absence hours / Planned hours) × 100
| Rate | Interpretation |
|---|---|
| Below 5% | Healthy level |
| 5 to 7% | Monitoring required |
| Above 8% | Managerial alert |
Absenteeism is a lagging indicator of operational overload. It systematically climbs in the weeks following a poorly managed activity spike or a prolonged period of strain. Address the causes (workload, management, recognition), not the symptoms.
Turnover measures the rate at which employees leave over 12 months. High turnover is costly: recruitment, training, lost expertise, declining quality. And it shows directly in the customer experience. An advisor who leaves after 3 months takes the entire training investment with them.
Formula:
Turnover = (Number of departures over 12 months / Average headcount) × 100
| Rate | Level |
|---|---|
| Below 15% | Excellent |
| 20 to 25% | Contact centre market standard |
| Above 30% | Major risk to performance |
The three levers that significantly reduce turnover: a visible continuous training path, real career progression (towards supervisor, trainer, expert roles), and management quality with regular feedback and recognition.
Cost per contact measures the average cost of handling a customer interaction, all expenses included. It’s the reference economic indicator for weighing channels against each other and assessing the ROI of improvement initiatives.
Formula:
Cost per contact = Total costs (salaries, overheads, tools, premises, management, training) / Number of contacts handled
| Channel | Indicative cost per contact |
|---|---|
| Phone | €3 to €6 |
| €4 to €8 | |
| Chat | €2 to €4 |
| Self-service | €0.10 to €0.50 |
Analyse cost by channel and by contact type. Weigh self-service (low cost, low relational value) against human contact (higher cost, high value on complex situations). The right balance depends on your sector and your customer positioning.
The digital tool adoption rate measures the proportion of visitors who actually use your self-service tools (FAQ, chatbot, customer portal) relative to the total number of visitors on your digital channels. You’ve invested in these tools: if no one uses them, it’s an investment with no return.
Formula:
Adoption rate = (Sessions on digital tool / Site or app visits) × 100
| Rate | Interpretation |
|---|---|
| Above 30% | Good adoption |
| Below 15% | Visibility or usability problem |
Concrete action: test your tools’ accessibility by putting yourself in the customer’s shoes. Is your FAQ reachable in fewer than 3 clicks from the homepage? If not, rethink the navigation before investing in content.
The employee experience score measures the level of engagement, wellbeing, and satisfaction of advisors in their work environment. Your advisors are your first customer interface. A disengaged employee delivers a degraded experience, regardless of how good your tools are. EX isn’t a “nice to have”: it’s a direct business lever.
It’s measured through a quarterly survey covering engagement, recognition, working conditions, management, and career progression. The overall score is expressed out of 100 or as eNPS (employee Net Promoter Score).
The correlation between employee experience and customer satisfaction is consistently observed in the field. Engaged teams who feel good in their role deliver a better customer experience, and show lower absenteeism and turnover.
Concrete actions: survey your teams every quarter, not once a year, share results transparently, and act on the top 3 irritants identified within the following 30 days.
You can’t actively track everything. The approach that works: structure your KPIs into 4 pillars, with a maximum of 10 indicators managed in daily operations. The rest stay on monthly review.
| Pillar | Key question | Associated KPIs |
|---|---|---|
| Accessibility | Can the customer reach you? | Service Level, abandonment rate |
| Efficiency | Are you solving the problem? | FCR, AHT by reason, repeat contact rate |
| Quality and experience | Is the customer satisfied? | CSAT, NPS, quality compliance |
| Employee engagement | Are your teams doing well? | Turnover, absenteeism, EX score |
These four pillars influence one another. Struggling teams (pillar 4) degrade resolution quality (pillar 2), which weighs on customer satisfaction (pillar 3), before the abandonment rate climbs (pillar 1). Structured management lets you see these chains forming before they become crises. To go further on building a measurement framework in an outsourcing context, see our guide on managing an outsourced contact centre.
Managing a high-performing contact centre isn’t about stacking up indicators. It’s about choosing the right KPIs, understanding them in depth, and turning them into concrete decisions. These 15 indicators cover every dimension of a contact centre: accessibility, efficiency, quality, and team engagement. Used together, structured into 4 pillars, they give a complete view without drowning managers in numbers.
If you’d like to assess your current management framework or build a dashboard tailored to your challenges, the Armatis teams can support you through our customer relationship audit and consulting services. To go further on customer satisfaction indicators, also see our guides on CSAT and on the NPS, CSAT, CES trio.
How many KPIs should a contact centre track?
A maximum of 10 KPIs actively managed on a daily or weekly basis. The rest can stay on monthly review. Beyond 10 active indicators, teams lose sight of where to focus their attention and action plans become scattered. The goal isn’t to be exhaustive, it’s to be decisive.
What’s the most important KPI in a contact centre?
FCR (First Contact Resolution) is often cited as the most strategic KPI, because it simultaneously impacts customer satisfaction, operational costs, and contact volume. Improving FCR mechanically reduces the repeat contact rate, lightens team workload, and improves CSAT. It’s the only KPI that reconciles quality and economic performance.
How can you quickly improve a contact centre’s CSAT?
Three levers have the greatest short-term impact: identify the 3 contact reasons with the lowest CSAT and address their root cause, analyse customer verbatims to understand the real irritants, and train teams on individual results rather than generic criteria. CSAT is more closely tied to resolution quality than to waiting time.
How do you build these KPIs into an outsourcing contract?
The most common KPIs in BPO SLAs are Service Level, FCR, and CSAT. Each indicator should come with a minimum monthly threshold, a measurement method shared between both parties, and a formal review frequency. Combining CSAT and FCR avoids perverse effects where a provider improves its AHT at the expense of resolution quality.
Should you measure employee satisfaction in a contact centre?
Yes, and on a quarterly basis. The correlation between employee experience (EX) and customer satisfaction (CSAT) is well documented and observable in the field. Engaged, well-trained teams who see real career progression deliver a better customer experience. EX is a predictive indicator of turnover and absenteeism, two of the main hidden costs of a contact centre.
Armatis is a European specialist in customer relations and business process outsourcing (BPO), operating across multiple continents with thousands of employees serving companies of all sizes and sectors. The company designs and manages end-to-end customer service operations: multichannel contact centres, complaints handling, technical support, back-office and digitised processes. Backed by integrated technology infrastructure and the ability to adapt to any sectoral and regulatory context, Armatis helps its clients combine operational performance, quality of experience and cost control, wherever they need it.
Contact our teams to discuss your challenges and find out how we can support you
Join the leaders who trust our multilingual and technological expertise.