Live chat
Live chat is a real-time messaging channel that lets customers talk with a support agent or AI agent through a chat window on a website, mobile app, or product. It's text-based like email, but it happens in the moment, with replies expected in seconds or minutes rather than hours.
That mix of speed and convenience is why live chat has become a default support channel for so many companies. Customers can get help without picking up the phone, leaving the page they're on, or waiting a day for an email reply.
How live chat works
A live chat session usually starts when a customer clicks a chat widget. That widget connects to a chat system behind the scenes, which decides where the conversation should go. It might go to an AI agent first, to a specific team, or to the next available person in a general queue.
On the other side, human agents work from an agent console where they can handle several chats at once, see the customer's account details, and look up past conversations. When an AI agent handles the first part of a chat and then hands off, the agent should get the full history so the customer doesn't have to repeat anything.
Most live chat services also record transcripts, track wait and response times, and connect to a help desk or CRM. That's what turns a chat window into a working support channel instead of just a box on a page.
Live chat vs. phone, email, and messaging
Phone is still the fastest way to sort out a complicated issue, but it ties up one agent per customer and makes people wait on hold. Email is easy to scale, but it's slow, and long threads can drag on for days. Live chat sits between the two, with the speed of a call and the flexibility of text.
It's also worth separating live chat from asynchronous messaging. With live chat, both sides are expected to stay in the conversation until it ends. With messaging, customers can leave and come back hours later, which works better when they need time to find an order number or a document.
Many teams offer all of these side by side as part of an omnichannel customer support setup. The goal is to let customers pick the channel that fits the problem, and to carry context with them if they switch.
Why customers and support teams use live chat
For customers, the main draw is speed. First response time in chat is usually measured in seconds, which matters most for time-sensitive questions about a purchase, a login problem, or a late delivery.
For support teams, live chat is often cheaper to run than phone because one agent can work several conversations at once. It also pulls volume away from the phone queue, since plenty of customers would rather type while they keep doing something else.
The quality of the experience still matters a lot. A study of banking customers using live chat found that the emotions people felt during chat shaped whether they'd recommend the company, and that customer service and support, followed by design, had the biggest effect on those emotions.
Live chat best practices
The basics start with being easy to find. In its guidelines for customer-service chat, Nielsen Norman Group found that people usually look for chat under Help or Contact Us, and that hard-to-find chat made them feel the company didn't want to hear from them.
Setting expectations helps too. Show clear hours if chat isn't always staffed, give a real time estimate before the chat starts, and keep customers updated if a reply is taking longer than usual. Even a minute can feel long when someone is staring at a screen.
Inside the conversation, avoid making people type the same question twice, be upfront when they're talking with a bot, and offer a transcript at the end. Answers should be specific and detailed rather than a string of canned replies.
How AI is changing live chat
Live chat used to mean a human on the other end of every conversation. Today, many chats start with an AI agent that can answer questions, look up orders, and complete tasks like refunds or plan changes on its own.
When this works well, AI resolves routine requests instantly and at any hour, and teams track the share of conversations it handles with metrics like chatbot containment rate. Human agents then spend more of their time on the complex or sensitive cases where judgment matters most.
The handoff is where many live chat setups still fall short. A good one passes the full conversation and context to a person, so the customer feels like they're continuing the same chat rather than starting over.
For a deeper dive, download Decagon's guide to AI and the next generation of CX.

