Contact Us

If you still have questions or prefer to get help directly from an agent, please submit a request.
We’ll get back to you as soon as possible.

Please fill out the contact form below and we will reply as soon as possible.

  • Contact Us
  • Give Feedback
  • Home
  • Deploying Your Agent

Deploying Outlearn to Intercom

AI-drafted replies for every Intercom conversation - your agents review and send with one click.

Written by Outlearn Documentation

Updated at April 24th, 2026

Contact Us

If you still have questions or prefer to get help directly from an agent, please submit a request.
We’ll get back to you as soon as possible.

Please fill out the contact form below and we will reply as soon as possible.

  • Getting Started
  • Your Agents
  • Knowledge Sources
    Understanding Sources Managing Sources Knowledge Base Integrations Cloud Storage Ticketing Systems Team Chats Manual Imports
  • Actions
    Understanding Actions Communication Actions Calendar & Scheduling CRM & Sales Custom & Advanced
  • Handoffs
  • Deploying Your Agent
    How Deployment Works Customizing Your Chat Widget
  • Analytics
  • Plans & Billing
  • Terms & Policies
+ More

Table of Contents

What the Intercom Copilot Does How to Add Outlearn to Intercom What Your Agents Will See Best Practices

Your agent gets a message: "Hi, I was charged twice this month." They know the answer. They open the billing docs, find the right section, copy the relevant part, rewrite it in a tone that fits the conversation, add a closing line, and send. Four minutes for something they have done thirty times this week.

With Outlearn as a Copilot in Intercom, they open the conversation and the draft is already there - pulled from your knowledge base, written in your voice, ready to send. They read it, adjust the one line that needs changing, and click send. Thirty seconds. Same quality. The thinking is still theirs. The drafting is not.

In this article, you'll learn:

  • What the Intercom Copilot does
  • How to add Outlearn to Intercom from the Deploy tab
  • What your agents will see inside Intercom

What the Intercom Copilot Does

The Copilot is an AI assistant that works alongside your human agents inside Intercom - it doesn't replace them, it makes them faster.

For every conversation in your Intercom inbox:

  • Your agent reads the full conversation context
  • It drafts a suggested reply based on your knowledge sources and instructions
  • Your Intercom agent sees the draft, reviews it, edits if needed, and sends it in one click

Your agents stay in control. The AI handles the first draft.

How to Add Outlearn to Intercom

  1. Go to the Deploy tab.
  2. Scroll to the Helpdesk Copilot section.
  3. Find the Intercom card and click Add to Intercom.
  4. Log in to your Intercom account when prompted and authorize Outlearn.
  5. Follow the on-screen steps to complete the installation.

What Your Agents Will See

Once installed, your agents will see AI-drafted replies appear inside Intercom for each conversation. The exact experience depends on your Copilot configuration.

The Copilot behavior - including whether drafts appear as direct replies or internal notes, which admin they appear as, and assignment rules - is configured in the Handoffs tab. See  Connecting Intercom (Handoff + Copilot) for configuration details.

Best Practices

  • Start with Leave Note Instead of Replying enabled in your Copilot configuration - this keeps AI drafts internal until your team has reviewed quality and built confidence.
  • Keep your knowledge sources well-maintained. The Copilot drafts based on what your agent knows - outdated or thin sources produce weak drafts.
  • Gather feedback from your agents after the first week - they're the best judges of whether the drafts are saving time or needing too many edits.
integrate enhance

Was this article helpful?

Yes
No
Give feedback about this article

Related Articles

  • Connecting Intercom (Handoff + Copilot)
  • Creating Your Account

Copyright 2026 – Outlearn.

Knowledge Base Software powered by Helpjuice

Expand