Three messages in three seconds? One complete answer
When customers type in rapid bursts — a greeting, a question, a photo — the agent now reads the whole burst and replies once, to all of it.
Real customers don't compose essays. They type “hey”, then “is the canvas weekender back in stock”, then “the olive one” — three messages in as many seconds. The agent now treats a burst like that as one thought: every reply considers every message that hasn't been answered yet, attachments included, so a photo followed by “do you have this in olive?” keeps the photo in the picture.
And when a new message lands while the agent is mid-reply, the in-flight answer steps aside quietly instead of racing the new one — you never get two replies fighting over the same question. Escalation and sentiment detection read the full burst too.
One reply for the whole thought
Before composing an answer, the agent now gathers everything the customer has said that hasn't been answered yet — not just the message that happened to arrive last. A burst like “hey” / “is the canvas weekender back in stock” / “the olive one” produces one reply that greets, confirms the product, and answers the color question, instead of three half-answers or an answer to only the last line.
Attachments ride along with their question. A customer who sends a photo and then types “do you have this in olive?” is asking about the photo — the agent sees both together, so the answer is about that product, not a guess.
Never two competing replies
If a new message lands while a reply is already being composed, the in-flight reply steps aside quietly and a fresh one takes over — written against the full, updated conversation. The customer never sees two answers racing each other, and never sees a reply to a question they've already moved past. Escalation detection and sentiment analysis read the full burst too, so “this is the third time I'm asking” followed by the actual question registers as the frustration it is.
Every channel, including the API
Burst handling applies everywhere the agent replies — the widget, the chat page, messaging channels, and conversations created through the public API. The API channel also picked up a transcript fix along the way: earlier messages in a burst could previously be missing from the conversation your team sees in the inbox. Now every message a customer sends appears in the transcript, whether or not it arrived mid-reply.
See what's new in your workspace
Everything on this page is live today. Asks trains on your website and resolves customer conversations on every channel — free to try, live in minutes.