How long a customer waits to hear that a human is on it — and how to shorten the wait without faking it.
First response time. First response time (FRT) is the time between a customer’s first message and the first reply from a human agent. Automatic acknowledgements do not count as a first response. First response time is usually reported as a median, and often calculated within business hours rather than calendar hours.
The clock starts when the customer’s message arrives and stops at the first genuine human reply. Two measurement choices shape the result. First, business-hours adjustment: a message sent Friday evening and answered Monday morning is a weekend on the calendar but minutes on a business-hours clock, and which clock you use should match the hours you actually publish. Second, the average versus the median: response times have a long tail, and a handful of forgotten threads can drag an average far from what the typical customer experienced. The median tells the truth about the middle; track the 90th percentile separately to keep an eye on the tail.
The wait before the first reply is the worst wait, because it is undefined — the customer doesn’t know if the answer is coming in an hour, a day, or never. A first response converts that open-ended uncertainty into an expectation, which is why FRT correlates with satisfaction even when total resolution takes just as long.
The metric’s known failure mode is the empty first response: “we’re looking into this” sent purely to stop the clock. It improves the dashboard and helps no one — the customer’s actual wait just continues under a different name. If first replies are getting faster while resolution times aren’t moving, that is usually what’s happening.
FRT calculation. First response time = timestamp of the first human reply − timestamp of the customer’s first message, excluding auto-replies, aggregated as a median across conversations — typically counted inside business hours. Report it per channel: expectations on live chat are minutes, on email hours.
Lowering it honestly is a routing problem more than a typing problem. Most first-response delay is queue time — the message waiting for anyone to look at it — and shrinks when conversations land pre-sorted with an owner, the job triage and routing exists to do. Set different targets per channel, and let an honest auto-reply state the hours you keep so the expectation is set even when the team is asleep.
Give your customers faster answers and your team their evenings back.
Desktop and mobile. Free for solo.