First Response Time: What It Is, How to Measure and Improve It

First response time is the time between a customer's first message and the first reply from your team. In live chat it's the metric customers feel most directly: they're sitting on your site, looking at the chat window, waiting. This guide explains how to calculate first response time properly, whether an AI assistant's reply should count, what a reasonable target looks like for chat, and ten practical ways to bring it down.

What first response time means in live chat

In email, first response time is often measured in hours. In chat, it's measured in seconds and minutes, because the visitor expects a conversation, not a ticket.

The clock starts when the visitor sends their first message and stops at the first meaningful reply from someone on your side. Two details matter:

• Automatic messages don't count. A greeting that appears when the chat opens or a "we'll be with you shortly" auto-reply isn't a response — nothing has been answered.
• Each conversation counts once. First response time is about the start of a conversation. How fast you reply later in the chat is a separate metric, usually called average response time.

How to calculate it: average vs median

The basic formula is simple: for each conversation, subtract the time of the first customer message from the time of the first operator reply. Then summarize.

Average

Add up all the first response times and divide by the number of conversations. It's easy to understand, but a few very long waits — say, a chat that arrived just before lunch — can pull the average far up.

Median

Sort all response times and take the one in the middle. The median tells you what a typical customer experienced and isn't distorted by outliers. Many teams track both: the median for the typical experience, the average (or a high percentile such as the 90th) to catch the long tail.

What to include and exclude

Decide these rules once and write them down, so the number stays comparable from month to month:

• Missed chats. Conversations nobody answered have no first response time at all. Track them separately — a low response time with many missed chats is a misleading picture.
• Spam and tests. Exclude them.
• Business hours. Messages sent at 3 a.m. when nobody works will look like eight-hour waits. Most teams either measure only within working hours or report in-hours and out-of-hours separately.

Should an AI bot's reply count as a first response?

If an AI assistant answers first, it's tempting to count its reply and celebrate a near-zero response time. That number is real for the customer — they did get an answer — but it hides how quickly your people respond.

A cleaner approach is to track two metrics:

1. First response time — including the bot, which describes what visitors experience.
2. Wait time after handoff — the time between the bot transferring the chat and an operator's first reply.

The second number often matters more. A bot that hands off quickly, followed by a long silence, feels worse than no bot at all. ChatDeck's analytics shows both the average first response time and the wait time after bot handoff, so you can see each part separately. For setting up the transfer itself, read about chatbot-to-human handoff.

What is a good first response time for chat?

There's no universal benchmark, and published "industry averages" vary widely depending on who measured what. Use these rules of thumb as a starting point instead:

• Under a minute feels like real-time chat to most visitors.
• A few minutes is acceptable if the visitor got a quick acknowledgment first.
• Beyond several minutes, many visitors leave the page or open a second channel, so treat it as a sign something needs attention.

Your own trend matters more than any outside number. If your median moves from two minutes to one, that's progress, whatever anyone else reports.

Why first response time drifts up

When the number creeps up, the cause is usually one of these:

• Notifications aren't reaching anyone. The inbox tab is closed or muted.
• Peaks in volume — a sale, a newsletter, a Monday morning — with the same number of people online.
• No clear ownership. Everyone assumes someone else will take the chat.
• Operators juggling other work, such as phone calls or order processing.
• Long first replies. An operator spends three minutes writing a perfect answer instead of acknowledging first.
• Schedule gaps — lunch breaks, shift changes, time zones.

10 ways to reduce first response time

1. Install desktop and mobile apps for operators so new chats trigger a native notification even when the browser is closed.
2. Acknowledge first, answer second. A quick "Got it, checking now" stops the clock and reassures the customer.
3. Use canned responses for frequent questions. Our live chat scripts are a ready starting set.
4. Make the queue visible. A shared "waiting" view with a counter makes unanswered chats hard to ignore.
5. Assign ownership. Decide who picks up new chats during each part of the day.
6. Staff for your peaks. Look at when messages arrive and shift schedules to match, as covered in our guide to building a live chat support team.
7. Let an AI assistant answer common questions so operators only handle the chats that need them.
8. Set working hours so out-of-hours visitors are told when you'll be back instead of waiting in an empty chat.
9. Show operators context — the current page, the referrer, previous conversations — so they don't spend the first minutes asking basic questions.
10. Review slow chats weekly. Look at the slowest few conversations and ask why each one waited.

How to track it week over week

A single number tells you little; a trend tells you a lot. Set up a simple routine:

• Pick one main figure — median or average — and stick with it.
• Compare to the previous period. A week-over-week or month-over-month view shows whether changes are working.
• Break it down by operator to spot who needs help or a different schedule, not to rank people.
• Look at it alongside volume, missed chats and ratings. Faster replies that hurt quality aren't a win.
• Check the team presence timeline to see whether slow periods line up with gaps in coverage.

ChatDeck's analytics covers these out of the box: average first response time, response rate, wait time after handoff, per-operator response times, a comparison with the previous period and a timeline of when team members were online or away. For the bigger picture, read our overview of live chat metrics.

Frequently asked questions

What is a good first response time for live chat?

As a rule of thumb, aim for under a minute during working hours, and treat anything beyond several minutes as a problem to investigate. Your own trend is a better guide than external benchmarks.

Is first response time the same as average response time?

No. First response time measures only the first reply in a conversation. Average response time looks at every reply, including those later in the chat.

Do automated greetings count toward first response time?

They shouldn't. A greeting or auto-reply doesn't answer anything. An AI assistant's real answer is different — count it, but also track how long customers wait for a person after handoff.

How do I measure response time outside business hours?

Either exclude out-of-hours conversations from the main figure, or report them separately. Mixing them in makes the number jump for reasons that have nothing to do with team performance.

Measure it from day one

The easiest way to improve first response time is to see it clearly. Create a free ChatDeck workspace and the analytics tracks response times, handoff waits and ratings automatically, with no per-operator fees. For more on the numbers worth watching, see live chat metrics. And so new chats reach you away from your desk, set up notifications and apps.