The live chat metrics that show whether your team is doing well

• Which live chat metrics should you track?
• What counts as a good first response time?
• How many chats should you resolve in one go?
• What targets are realistic for a small team?
• What should you change when a number is off?
• Which number should you fix first?

You can tell whether your chat team is doing well from four live chat metrics. The median first reply on live chat was 47 seconds across 2.4 million tickets, according to Visionary Marketing’s 2026 analysis. Here’s how to measure each number and set targets a small team can hit.

None of this needs a data team. You need a week of chats, a spreadsheet or your chat tool’s reports, and the habit of looking at the same four numbers every Monday.

Which live chat metrics should you track?

Track four customer service KPIs, the key numbers that show whether support is working: first response time, resolution, missed chats and customer satisfaction. Each one catches a problem the others miss. A team can reply in seconds and still send customers away with half an answer, and only the resolution number will show it.

First response time

First response time, the time between a customer’s first message and your first real reply, is the number customers notice first. Lorikeet’s 2026 guide to first response time leaves automatic acknowledgments out of it, because a thanks-for-waiting message doesn’t answer anything.

Report it as a median, the middle value when you line up every reply time, not as an average. A few chats left waiting overnight pull the average up and hide what most customers actually saw. Lorikeet also suggests checking the 90th percentile, the wait that nine in ten customers beat, so you can see your worst experiences.

If a bot answers some chats, count human replies on their own. Blending bot replies into first response time lowers the figure and hides how long people wait for a person, the same Lorikeet guide warns. If you’re still deciding who should answer first, this comparison of when a chatbot or a person helps a small business more walks through the tradeoff.

Resolution time and first contact resolution

Resolution time is how long a chat takes from the first message to a solved problem. First contact resolution, or FCR, is the share of customers whose issue gets fixed in that one conversation, with no follow-up needed. Of the two, FCR tells you more, because a quick chat that sends the customer back to you isn’t resolved.

To measure FCR, check whether the same customer comes back about the same issue, or ask them directly in your satisfaction survey. SQM Group, which benchmarks call centers, measures it through customer surveys after the contact. That catches chats your team marked solved but the customer didn’t consider solved.

For chat resolution time, there’s no chat-only benchmark to compare against. Visionary Marketing’s 2026 study reports a median of 6 hours 18 minutes, but that mixes chat with every other channel. A 2025 benchmark of more than 220 million chats from a chat software vendor, summarized by CO Consulting, puts the average chat session at 11 minutes 55 seconds. Track your own resolution time week by week and watch which way it moves.

Missed chats

A missed chat is one where the customer gave up before anyone replied. Call centers call this the abandon rate: abandoned contacts divided by all the contacts that came in. SQM Group’s 2022 call center benchmark puts the average at 6% and rates anything under 5% as good.

Leave out chats closed in the first few seconds. SQM Group recommends ignoring contacts abandoned within 5 to 10 seconds, because these false abandons can make up as much as 2% of volume.

In LenoChat, a chat nobody has picked up waits under the Unassigned filter in the inbox, and the Dashboard shows Owners and Admins total conversations, response times and agent workloads. Checking that filter at the end of each shift tells you how many chats are still waiting for a person.

Customer satisfaction

Customer satisfaction, or CSAT, is how happy customers say they were with the chat, usually on a five-point scale. SQM Group sends its survey within one business day of the contact and scores only the top box: very satisfied answers divided by all answers.

SQM Group’s 2022 benchmark puts the average share of very satisfied customers at 78%. It rates 75% to 84% as good and 85% or higher as its top tier.

Visionary Marketing’s 2026 study found a median live chat score of 4.3 out of 5. Speed did more for satisfaction than anything else it measured, and problems too complex for chat caused the most friction.

What counts as a good first response time?

A median first response time under 120 seconds is a strong target for a small team, and past 8 minutes you’ve gone beyond what the typical customer will tolerate. SQM Group’s 2022 call center research found no drop in satisfaction for waits inside 120 seconds, and Visionary Marketing’s 2026 survey puts the median consumer’s patience for a chat reply at 8 minutes.

Most customers expect a reply well before that limit. In Visionary Marketing’s survey of 5,000 consumers, 64% expected a live chat reply in under 5 minutes and 89% expected one within 15 minutes. The study doesn’t say which countries the people surveyed live in, so treat these as a broad signal rather than a local one.

Businesses already reply much faster than customers ask. Across the 2.4 million tickets Visionary Marketing analyzed, the fastest quarter of brands replied in a median of 8 seconds and the slowest quarter took 6 minutes. The overall median dropped 52% in a single year, so the bar keeps rising.

Don’t chase the fastest quarter with a team of two. A blended number that includes bot replies can look that quick while customers who need a person still wait, as Lorikeet points out. A second target helps more: the share of chats answered within 120 seconds. Call centers have traditionally aimed for 80% of contacts answered within 20 seconds, and SQM Group recommends measuring the 120-second share instead.

How many chats should you resolve in one go?

Aim to resolve at least 70% of chats in the first conversation, and treat 80% as a stretch goal. Visionary Marketing’s 2026 study found a live chat first contact resolution rate of 71%. SQM Group’s 2022 call center benchmark puts the average at 70%, which leaves about 30% of customers getting in touch again, and rates 80% or higher as its top tier, a level only about 5% of call centers reach.

The effort pays off in satisfaction. In SQM Group’s 2022 data, each 1% rise in first contact resolution came with roughly a 1% rise in satisfaction. In its 2021 research, 25% of customers whose issue stayed unresolved said they didn’t plan to keep using the business.

Chat customers expect a little less than callers do. In SQM Group’s 2021 research, 72% of customers contacting a business through its website expected their issue resolved in one contact, compared with 93% of call center customers. That gap gives chat some room, but not much.

Keep in mind where these numbers come from. No government or academic benchmark exists for live chat, and almost every published chat figure comes from software vendors or surveys they commissioned, according to CO Consulting’s 2026 roundup. The SQM Group figures come from call centers. Use them as a direction, and judge yourself mostly against your own last month.

What targets are realistic for a small team?

A realistic target sits between the published median and the top performers: a median first reply under 120 seconds, about 70% of chats resolved in one conversation, missed chats under 5% and very satisfied customers above 75%. The table below pulls together the figures behind those targets from Visionary Marketing’s 2026 chat data and SQM Group’s 2022 call center benchmark.

Metric	Typical result	Strong result	Needs work
First response time, median	47s	8s	Over 120s
First contact resolution	71%	80%	Below 70%
Missed chats, abandon rate	6%	Under 5%	5% or more
Very satisfied customers	78%	85% or more	Below 75%

The typical and strong first response figures and the typical resolution figure come from chat data. The rest lean on call center benchmarks, so read them as a direction rather than a promise. If your medians sit close to the typical column, you’re doing fine, and if one lands in the last column, that’s the number to work on next.

What should you change when a number is off?

Change one process at a time, starting with how chats reach your team. A slow first response time is usually a queue and routing problem rather than slow agents, according to Lorikeet’s 2026 guide.

When first response time is slow

Look at where chats wait. Lorikeet suggests three fixes: route each chat to the right person the first time so it isn’t put back in the queue, take routine questions such as order status out of the human queue, and put urgent customers first.

Then cut the time each reply takes to write. A set of saved replies for the questions your team answers every day lets an agent send a correct first answer in a few clicks.

When resolution is low

Find the handful of reasons customers come back. In SQM Group’s 2021 research, the top five reasons for repeat contacts made up 43% of all repeat contacts, so fixing five answers can remove a large share of second chats.

Next, look at chats where an agent had to ask a colleague. SQM Group’s 2022 data shows contacts that needed help from a supervisor or colleague had 16% lower first contact resolution, and agents needed help on about 20% of calls. Write those answers down where everyone can find them, and use a starter set of canned responses for live chat for the common ones.

Watch speed while you do this. Lorikeet recommends tracking first response time next to first contact resolution, so faster replies don’t simply create more follow-up chats.

When missed chats climb

Check when the missed chats happen. If most arrive in the same hours, the fix is cover for those hours or an honest message about when you’ll reply, not faster typing.

Also check whether those customers simply moved channels. In SQM Group’s 2021 research, 42% of customers used two or more channels at the same time for the same issue, so a missed chat may turn up again as a message somewhere else. This guide to how customers choose between live chat, WhatsApp and Instagram helps you decide which channels to watch.

When satisfaction drops

Read the chats with low scores before changing anything. If they were slow, fix first response time first, because speed did more for chat satisfaction than anything else in Visionary Marketing’s 2026 study. If they were complex problems, offer a call or an email early, since issues too complex for chat caused the most friction in the same study.

Which number should you fix first?

Start with first response time, then check first contact resolution so faster replies don’t turn into repeat chats. Speed was the biggest driver of chat satisfaction in Visionary Marketing’s 2026 study, and Lorikeet pairs the two numbers for exactly this reason.

After that, look at all four live chat metrics once a week, as medians, and change one thing at a time. When a number moves, you’ll know what moved it.

Best for: teams of one to five people who answer chat alongside other work and want a short weekly check.

Not for: businesses that get only a handful of chats a week, where one slow reply swings every number, so read the chats themselves instead.