Customer Service Metrics Explained: FRT, Resolution Time, CSAT, NPS, CES

Customer service metrics are numbers that show how fast and how well your team helps customers. Without them, the only signal an owner gets is the loudest complaint of the week, which is rarely a fair picture of how the team is doing.

This explainer covers the five metrics most businesses should know, plus SLA compliance, with a plain-English definition, the formula, and a worked example from a Nairobi boutique for each. It is written for owners and support leads who want a handful of numbers they can actually act on.

What are the main customer service metrics?

The main customer service metrics fall into two groups: how fast you are and how customers feel. Here is a summary.

What is first response time (FRT)?

First response time is the time between a customer's first message and the first reply from a person on your team. It is the metric customers feel most directly on WhatsApp and live chat.

Formula: FRT = sum of (first reply time − first message time) ÷ number of conversations.

Count only business hours, and do not count instant auto-acknowledgements as a reply.

Worked example. A Nairobi boutique receives five WhatsApp enquiries one morning. First replies come after 2, 4, 6, 3 and 35 minutes.

• Average FRT = (2 + 4 + 6 + 3 + 35) ÷ 5 = 10 minutes.
• Median FRT = 4 minutes (the middle value when sorted: 2, 3, 4, 6, 35).

The average suggests customers wait 10 minutes; the median shows most waited four, and one chat was simply missed. That missed chat is the thing to fix. For practical fixes, see how to reduce first response time.

What is average resolution time?

Average resolution time is how long it takes, on average, to fully solve a customer's problem, from when the request arrives to when it is marked resolved. It is usually measured on tickets rather than quick chats.

Formula: ART = sum of (resolved time − opened time) ÷ number of resolved tickets.

Worked example. In one week the boutique resolves four tickets: an exchange (3 business hours), a refund (5 hours), a delivery mix-up (20 hours) and a faulty zip repaired by a tailor (52 hours).

• ART = (3 + 5 + 20 + 52) ÷ 4 = 20 business hours.
• Median = (5 + 20) ÷ 2 = 12.5 business hours.

Break ART down by category. Here the tailor repair dominates, which points to a supplier conversation rather than slow agents. A helpdesk with categories makes this split easy; see what a helpdesk ticketing system is.

What is CSAT and how is it calculated?

CSAT (customer satisfaction score) measures how satisfied a customer was with a specific interaction, usually from a one-question survey sent after a conversation or ticket is resolved: "How satisfied were you with our help today?" on a scale of 1 to 5.

Formula: CSAT = (number of 4 and 5 ratings ÷ total ratings) × 100.

Worked example. In October the boutique sends a short rating question on WhatsApp after resolved conversations. 120 customers answer; 96 give a 4 or 5.

• CSAT = 96 ÷ 120 × 100 = 80%.

Read the comments behind low scores; they are often more useful than the score. Our guide on collecting customer feedback on WhatsApp covers how to ask without annoying people.

What is NPS (Net Promoter Score)?

Net Promoter Score measures how likely customers are to recommend your business, on a scale of 0 to 10. It was developed by Fred Reichheld with Bain & Company and Satmetrix, and it reflects the overall relationship rather than a single chat.

Respondents are grouped as promoters (9 or 10), passives (7 or 8) and detractors (0 to 6).

Formula: NPS = % promoters − % detractors. The result ranges from −100 to +100.

Worked example. The boutique asks 200 customers "How likely are you to recommend us to a friend?" 110 answer 9 or 10, 60 answer 7 or 8, and 30 answer 6 or below.

• Promoters = 110 ÷ 200 = 55%.
• Detractors = 30 ÷ 200 = 15%.
• NPS = 55 − 15 = +40.

Passives count in the total but not in the subtraction. Ask NPS occasionally, such as quarterly, not after every chat.

What is CES (Customer Effort Score)?

Customer Effort Score measures how easy it was for a customer to get their issue handled. The idea was popularised by research from the Corporate Executive Board (now part of Gartner), which argued that reducing effort builds loyalty more reliably than trying to delight customers.

A common version asks customers to rate the statement "The business made it easy to handle my issue" from 1 (strongly disagree) to 7 (strongly agree).

Formula: CES = sum of scores ÷ number of responses.

Worked example. After returns are completed, 50 customers answer. Their scores add up to 290.

• CES = 290 ÷ 50 = 5.8 out of 7.

If CES drops, look for friction: being asked to repeat details, being moved between channels, or having to chase for updates. A private ticket progress link that customers can check themselves is one simple way to cut effort.

What is SLA compliance?

SLA compliance is the share of tickets handled within your agreed response and resolution targets. It turns your service promises into one number.

Formula: SLA compliance = (tickets within target ÷ total tickets) × 100.

Worked example. Of 200 tickets in a month, 180 met both their response and resolution targets: compliance is 90%. Look at the 20 breaches by category and priority to find the cause. Our guide on setting up customer support SLAs explains how to choose targets.

Which customer service metrics should a small business track?

Track three or four metrics you will act on, not a dashboard of twenty. A sensible starting set for a Kenyan SME:

1. First response time on WhatsApp during business hours, as an average and a median.
2. Average resolution time for tickets, by category.
3. SLA compliance for urgent and high-priority tickets.
4. CSAT after resolved conversations, or NPS each quarter if you prefer a relationship view.

In chat.or.ke, managers can see response-time reports per channel, and the helpdesk records priorities, categories and SLA alerts on every ticket. For metrics such as CSAT and NPS, a short question sent on WhatsApp after resolution works well; keep it to one tap where possible.

What mistakes do businesses make with metrics?

• Chasing one number. Pushing only FRT encourages agents to send quick "hello, checking" replies that do not help.
• Using averages alone. Always look at the median and the slowest cases.
• Ignoring channels. Mixing email and WhatsApp in one FRT figure hides problems in both.
• Surveying too often. Asking NPS after every chat annoys customers and lowers response rates.
• Measuring without deciding. Every metric should link to an action, such as a rota change, a new saved reply or a supplier conversation.