GLOSSARY

Customer service metrics and how to calculate them

Each metric below has its own page, with the formula, a worked example and benchmarks wherever a named source publishes them. Most lists of customer service KPIs are written for a support team with a manager and a helpdesk. If you answer support yourself, start with ticket volume, first response time and deflection rate. They show how many questions reach you, how long customers wait for a reply and how many get answered without you.

Deacon is an AI customer support agent that answers your customers in seconds from your own docs, so fewer questions reach you, and it gets better with every answer you write. Its dashboard counts messages, people and the share it answered, and Topics show what customers are stuck on and what they want next. There’s a free plan, and it needs no card.

• Average handle time

  Average handle time (AHT) is how long you spend working on each call, chat or email, on average, including the follow-up once it ends.

• Average resolution time

  Average resolution time is the average time from a customer’s first message to their problem being solved, usually counting every hour in between.

• Containment rate

  Containment rate is the share of conversations a chatbot or IVR phone menu finishes on its own, without passing the customer to a person.

• Cost per contact

  Cost per contact is what support costs you in a period divided by the number of times customers got in touch in it, by phone, email or chat.

• Customer satisfaction score (CSAT)

  CSAT, or customer satisfaction score, is the share of survey responses that say the customer was satisfied, usually a 4 or 5 on a five-point scale.

• Customer effort score

  Customer effort score (CES) asks customers how much work it took to get their problem solved, in one survey question sent after a support conversation.

• Customer experience (CX)

  CX stands for customer experience, everything customers go through with a company, from the first visit to renewing or cancelling, and how they feel about it.

• Customer self-service

  Customer self-service is every way customers find an answer or fix a problem without contacting you, from a help centre to a billing portal or an AI chat.

• Deflection rate

  Deflection rate is the share of customer questions that self-service, such as a help centre or a chatbot, answers before they reach your team.

• Escalation rate

  In customer support, escalation rate is the share of conversations passed up to a person or a more expert team, such as a chatbot handing a chat to you.

• First contact resolution

  First contact resolution (FCR) is the share of customer problems solved in the first contact, so nobody has to get in touch again about the same thing.

• First level resolution

  First level resolution (FLR) is the share of support tickets solved by the first line of support, without passing them up to a more expert level.

• First response time

  First response time (FRT) is how long a customer waits between sending their message and getting the first reply from your team.

• Human in the loop

  Human in the loop means people review, correct or take over from an AI at set points, so the AI handles what it can and a person decides the rest.

• Knowledge base

  A knowledge base is one searchable place for the answers to your customers’ questions. What goes in one, how tools organise it, and how to start small.

• Resolution rate

  Resolution rate is the share of customer problems that end up solved, whoever solves them. For an AI agent, each vendor decides what counts as solved.

• Support tiers

  Support tiers sort who answers a customer’s question, from self-service at tier 0 up to your engineers at tier 3 and outside vendors at tier 4.

• Ticket backlog

  Ticket backlog is the number of support requests still unsolved at a given moment, whether they are waiting on you, on the customer or on a fix.

• Ticket volume

  Ticket volume is the number of support requests your customers send in a period, by email, chat, phone or form, with each request counted once.

• Ticketing system

  A ticketing system is software that turns each customer request into a ticket, with its own number, a status and an owner, so a team can share the work.