Back to Templates

Reply to Facebook ad comments with OpenRouter AI, Google Docs and Slack

Created by

Created by: Salman Mehboob || salmanmehboob
Salman Mehboob

Last update

Last update 9 hours ago

Share


Every comment on your Facebook ad is a sales opportunity.
But manually replying to dozens of comments every day - in the right tone,
in the right language, at the right speed - is simply not scalable.

This workflow deploys an AI agent that watches your boosted Facebook posts 24/7,
classifies every incoming comment, and instantly posts a smart public reply -
so your ads stay engaged, your leads get answered, and your team only steps in
when it truly matters.


🚨 The Problem This Solves

When businesses and agencies run Facebook Ads, comments pile up fast:

  • Queries go unanswered for hours, killing purchase intent
  • Positive reviews get ignored, wasting social proof
  • Negative reviews stay public with no response, damaging brand trust
  • Spam clutters the thread, making the brand look unmonitored
  • Manual replies are slow, inconsistent, and cost agency time

This workflow eliminates all four problems automatically — with zero manual effort
for routine engagement and instant alerts for anything that needs a human touch.


⚙️ How It Works

  1. Webhook fires every time someone comments on your Facebook page
  2. Filter skips comments by the page owner and ignores non-boosted organic posts —
    only real comments on active ad posts continue
  3. Get Comment Details fetches the full comment text from the Facebook Graph API
  4. Skip Attachments drops GIF, sticker, and image-only comments —
    only text comments are processed
  5. Get Post Data fetches the original ad post text so the AI understands the context
  6. AI Classifier reads both the ad and the comment together, then classifies into:
    QUERY · POSITIVE_REVIEW · NEGATIVE_REVIEW · SPAM
  7. Switch routes each classification to the right action branch
  8. AI Reply Generator writes a warm, human-sounding public reply —
    automatically in the commenter's own language (English / Urdu / mixed)
    and searches your Google Docs Knowledge Base for accurate answers to queries
  9. Reply to Comment posts the reply publicly on Facebook via Graph API
  10. Slack Alert sends your team a full notification with commenter name,
    comment text, post link, and comment link — for every negative review
    that needs a manual response

✅ Key Benefits

  • Always-on engagement: Replies post instantly, day or night, no human needed
  • Context-aware AI: Reads the ad post AND the comment together before classifying —
    not just the comment alone
  • Multilingual replies: Automatically matches the language of the commenter —
    English, Urdu, or mixed — without any configuration
  • Knowledge Base powered: Queries are answered using your own pricing, FAQs,
    and service info stored in Google Docs — no hallucinated answers
  • Smart escalation: Negative reviews never get an auto-reply —
    your team is alerted on Slack immediately with full context
  • Spam-proof: Attachment comments and spam are silently ignored,
    keeping your workflow clean and cost-efficient

🛠️ Setup Steps

  1. Subscribe your Facebook Page to webhook feed events and paste the webhook URL
  2. Add your Facebook Graph API access token to the Query Auth credential
  3. Add your OpenRouter API credential (used for both classification and reply generation)
  4. Create a Google Doc with your agency's FAQs, services, and pricing —
    paste the document ID into the Knowledge Base node
  5. Connect your Slack account and update the Slack user ID
    in the Inform User node to the right team member
  6. In the Filter node, replace your page name with your actual Facebook page name

⏱️ Estimated setup time: 20–30 minutes


📋 Requirements

  • Facebook Page with webhook subscription (feed changes)
  • Facebook Graph API access token
  • OpenRouter account (LLM — supports all major models)
  • Google Docs account (Knowledge Base document)
  • Slack account (for negative review alerts)

For assistance and support: [email protected]
Linkedin: https://www.linkedin.com/in/salman-mehboob-pro/