Sales · n8n

Classify Cold Email Replies With AI and Update Your CRM

Automatically read replies to your cold outreach, classify them as interested, not interested, or out-of-office with AI, and update the deal stage accordingly.

difficulty Advancedsetup 40 minresult Every reply is auto-triaged and the CRM stage updated, so reps spend time only on genuinely interested prospects.
  1. 1

    Trigger on Replies

    Add a Gmail Trigger node filtered to your outreach inbox or a specific label so only campaign replies are processed.

  2. 2

    Classify the Intent

    Add an OpenAI node prompted: Classify this reply as interested, not_interested, or auto_reply and return only the label.

  3. 3

    Match the Deal

    Add a Pipedrive node to find the deal by the sender's email address.

  4. 4

    Update the Stage

    Add a Switch node routing on the label — move interested deals to Engaged, mark not-interested as Lost, and skip auto-replies.

  5. 5

    Notify the Rep

    For interested replies, add a Slack node pinging the owning rep to follow up quickly. Activate and test with sample replies.

Frequently asked questions

How do I handle nuanced replies?

Add a fourth `needs_review` label for ambiguous replies and route those to a human instead of guessing the stage.

Can it draft a follow-up?

For interested replies, chain another OpenAI step to draft a contextual response for the rep to review and send.

About this recipe. Recipes on FlowRecipesHub are written for business owners, not developers, and are tested before publishing — how recipes get made. Some ingredient links are affiliate links that cost you nothing — full disclosure.