OLYMP LogoOLYMP
Contact
Reference

Customer support KPIs and how to calculate them

Twelve metrics cover almost everything worth measuring in customer support. Each one below has a plain definition and the formula most teams use. Pick a few that match your goals rather than tracking all of them.

Talk about your KPIsSee the formulas
SLA guideAI quality assuranceGlossaryChoosing a partner

Which customer support KPIs matter most?

Most support teams need four core KPIs: customer satisfaction (CSAT) for how customers feel, first contact resolution (FCR) for whether issues are solved, a responsiveness measure such as first response time or service level, and a quality-assurance score for how well agents follow your standards. Add occupancy, shrinkage and attrition for planning.

Customer outcome
CSAT, FCR, customer effort
Speed
First response time, service level, resolution time
Quality
QA score on a shared scorecard
Planning
Handle time, occupancy, shrinkage, attrition

Twelve customer support KPIs

Formulas use the most common definitions. Write down your own exact rules so numbers stay comparable over time.

  • Customer satisfaction (CSAT)

    How satisfied customers are with a specific interaction, usually from a one-question survey after the contact.

    Satisfied responses ÷ all responses × 100

  • First contact resolution (FCR)

    The share of issues solved in the first contact, without the customer having to come back.

    Issues solved on first contact ÷ all issues × 100

  • First response time (FRT)

    How long customers wait for the first human reply on email, tickets or messaging.

    Sum of first-reply times ÷ number of tickets

  • Resolution time

    How long it takes from opening a ticket to solving it. The median is often more useful than the average.

    Solved time − opened time, per ticket

  • Service level

    The share of calls or chats answered within a target time. The main input for staffing plans.

    Answered within target ÷ all offered × 100

  • Average handle time (AHT)

    The average time an agent spends on one contact, including talk or chat time, hold and follow-up work.

    (Talk + hold + after-contact work) ÷ contacts

  • Quality-assurance score

    How well conversations meet your standards on a shared scorecard: accuracy, resolution, tone and process.

    Points achieved ÷ points possible × 100

  • Net Promoter Score (NPS)

    How likely customers are to recommend you, on a 0–10 scale. Reflects the whole relationship, not only support.

    % scoring 9–10 − % scoring 0–6

  • Customer effort score (CES)

    How easy customers found it to get their issue solved, usually rated on a short scale after the contact.

    Average score, or share of easy ratings

  • Occupancy

    The share of logged-in time agents spend handling contacts. Too low wastes cost; too high causes burnout.

    Handling time ÷ logged-in time × 100

  • Shrinkage

    The share of paid time agents are not available for contacts: breaks, training, meetings, leave and absence.

    Unavailable time ÷ paid time × 100

  • Attrition

    How many people leave the team over a period. High attrition means constant hiring, retraining and uneven service.

    Leavers ÷ average headcount × 100, per year

Using support KPIs without gaming them

Metrics shape behaviour. These principles keep them pointed at the customer.

How many KPIs should a support team track?

Report on many, target few. Three to five targets are enough for a service level agreement or a team goal. Track the rest in reports, so you can see side effects without asking agents to optimise a dozen numbers at once.

Which KPIs should be read together?

KPI pairs that keep each other honest
If you pushAlso watchBecause
Average handle timeFirst contact resolutionShorter contacts can mean more repeat contacts
Service levelOccupancyFast answers with overloaded agents lead to burnout
CSATSurvey response rateA small, self-selected sample can mislead
Resolution timeReopen rateTickets closed too early come back

How do you make the numbers comparable over time?

  • Write down each definition, including edge cases such as abandoned contacts
  • Keep one system as the source of truth
  • Change definitions only with a note and, where possible, a recalculated history
  • Compare like periods, since seasonality changes every number
Continue reading

How do KPIs work with an outsourced team?

Use the same KPIs and the same scorecard for in-house and outsourced agents, measured from the same systems. Share dashboards rather than exchanging spreadsheets, and review real conversations together each week, so both teams discuss the same evidence.

Last reviewed: . Written and reviewed by the OLYMP BPO team in Prishtina, Kosovo. About the company

Support KPIs: common questions

What are the most important customer support KPIs?

For most teams, the core set is customer satisfaction (CSAT), first contact resolution (FCR), a responsiveness measure such as first response time or service level, and a quality-assurance score. Together they show whether customers are helped quickly, correctly and pleasantly. Add operational metrics such as occupancy and attrition for planning.

How do you calculate first contact resolution?

Divide the number of issues solved in the first contact by the total number of issues, then multiply by 100. Define what counts as a repeat contact, for example the same customer and topic within seven days, and keep that rule fixed so the number stays comparable.

What is the difference between CSAT and NPS?

CSAT measures satisfaction with a specific interaction, usually right after it. NPS measures how likely a customer is to recommend the company overall, on a 0–10 scale. CSAT is better for judging support quality; NPS reflects the whole product and relationship.

Is a low average handle time good?

Not on its own. A falling handle time can mean better tools and knowledge, or rushed contacts that come back later. Read it together with first contact resolution and quality scores, and avoid using it as an individual target, which encourages agents to end contacts too early.

Which KPIs should be in an outsourcing contract?

Put three to five measurable targets in the service level agreement, usually a responsiveness target per channel, a quality-assurance score and, once a baseline exists, customer satisfaction. Track the other metrics in regular reports without making them contractual targets.

Measuring the wrong things?

Tell us what you track today. We will suggest a small set of KPIs that shows whether customers are well served.

Talk to usRead the SLA guide