Skip to content
ENPL
Log in

Live support

How to handle conversations waiting for an agent, take them over from the bot, and end them.

The Live support tab is your team's operations center. This is where conversations land when a customer has asked to speak to a person, or the bot has decided it can't manage on its own (see Human support).

What you see on screen

The screen is split into two columns:

  • Waiting: conversations where the customer is waiting for an agent, with the longest-waiting first.
  • My conversations: open conversations you've taken over personally and are currently handling.

Each card in Waiting includes:

  • the customer's name (if they introduced themselves),
  • the last message, so you know what it's about,
  • the waiting time (for example "Waiting since 4 minutes ago"),
  • a Take over button and an Open link to the full conversation view.

Cards in My conversations show the customer's name, the last message, and the number of messages. Click a card to open the conversation.

Real-time updates

The list refreshes every 5 seconds, so you don't have to reload the page to see new conversations. If someone on your team takes over a conversation in the meantime, it disappears from the Waiting column.

The sidebar also shows how many conversations need a human, next to Conversations, so you can spot new ones from any screen.

Email notifications

Once the configured waiting time passes, if no agent takes over the conversation, Fibly closes it, tells the customer that the team will get back to them by email, and sends an email with a full summary of the dialogue to your team. The conversation then leaves the Waiting column. Configuration details are in the article Human support.

If you're on duty, keep the Live support tab open in the background. The list refreshes by itself, so new conversations will appear without a manual reload.

Articles in this section

  • Conversations: the full list of conversations, including those resolved by the bot alone.
  • AI assist for agents: how to generate a draft reply based on the knowledge base.