Skip to content

Latest commit

 

History

History
175 lines (125 loc) · 5.96 KB

File metadata and controls

175 lines (125 loc) · 5.96 KB
layout default
title Intent Detection — Rypl Documentation
description How Rypl's intent detection system classifies Facebook ad comments into 8 categories for smarter AI replies and sales optimization.

Intent Detection

Rypl's intent detection system automatically classifies every incoming comment into one of eight categories. This classification drives smarter AI replies, enables targeted analytics, and helps you understand what your audience wants at scale.

Available on Pro and Ultra plans.


How Intent Detection Works

When a comment arrives on your Facebook ad, Rypl's AI analyzes the text, context, and sentiment to determine the commenter's intent. This happens in real time — before the AI generates a reply. The detected intent influences:

  • Reply strategy — A purchase-intent comment gets a different reply than a support question
  • Priority routing — High-value intents (Purchase, Price Inquiry) can be prioritized
  • Analytics — Track which intents are most common across your ads
  • Moderation — Different intents can trigger different moderation actions

The 8 Intent Categories

1. Purchase Intent (Green)

The commenter is ready to buy or is expressing strong buying signals.

Example comments:

  • "I want to buy this! How do I order?"
  • "Take my money!"
  • "Can I get two of these?"
  • "Is this still available? I need it"

Recommended actions:

  • Reply immediately with a direct link to purchase
  • Include any active discounts or promotions
  • Make the buying process as frictionless as possible
  • Offer to answer any last questions before they purchase

2. Price Inquiry (Yellow)

The commenter is asking about cost, discounts, payment options, or availability.

Example comments:

  • "How much is this?"
  • "Do you have any discount codes?"
  • "What's the price with shipping?"
  • "Is this on sale?"

Recommended actions:

  • Provide clear, accurate pricing from your knowledge base
  • Mention any active promotions or bundle deals
  • Include a link to the product page
  • If the price is not in your knowledge base, direct them to your website or DMs

3. Return Request (Red)

The commenter is asking about refunds, exchanges, or returning a product.

Example comments:

  • "I want to return this, it doesn't fit"
  • "How do I get a refund?"
  • "Can I exchange this for a different size?"
  • "This isn't what I expected, can I send it back?"

Recommended actions:

  • Respond empathetically and direct them to your return/exchange process
  • Provide your return policy details from the knowledge base
  • Offer to move the conversation to DMs for personal assistance
  • Flag for human review if the situation is complex

4. Positive Feedback (Teal)

The commenter is leaving a compliment, testimonial, or expressing satisfaction.

Example comments:

  • "Love this product! Best purchase ever!"
  • "Your customer service is amazing"
  • "Just got mine and it's even better than expected!"
  • "Highly recommend this to everyone"

Recommended actions:

  • Thank the commenter warmly
  • Encourage them to share their experience or tag friends
  • Ask if they would be willing to leave a review
  • Use the positive interaction to boost engagement visibility

5. Complaint (Orange)

The commenter is expressing dissatisfaction, frustration, or a negative experience.

Example comments:

  • "Still waiting for my order, it's been 3 weeks"
  • "The quality is terrible"
  • "Worst customer service ever"
  • "This is nothing like the pictures"

Recommended actions:

  • Respond empathetically and acknowledge their frustration
  • Offer a concrete solution or next step
  • Move the conversation to DMs to resolve privately
  • Flag for human follow-up if needed

6. Support (Blue)

The commenter needs help, has technical questions, or requires assistance.

Example comments:

  • "How do I track my order?"
  • "The app isn't working for me"
  • "Can someone help me with my account?"
  • "I need help choosing the right size"

Recommended actions:

  • Provide helpful information from the knowledge base
  • Link to relevant support resources or FAQ pages
  • Offer to continue the conversation in DMs for detailed assistance
  • Escalate to human support if the issue requires it

7. Product Info (Purple)

The commenter is asking about product specifications, features, compatibility, or details.

Example comments:

  • "What material is this made of?"
  • "Does this work with iPhone 15?"
  • "What are the dimensions?"
  • "Is this waterproof?"

Recommended actions:

  • Pull accurate specifications from the knowledge base
  • Be specific and detailed in the response
  • Link to the product page for full details
  • If the information is not available, acknowledge that and offer to find out

8. Shipping (Cyan)

The commenter is asking about delivery, tracking, shipping costs, or logistics.

Example comments:

  • "How long does shipping take?"
  • "Do you ship internationally?"
  • "Where's my order?"
  • "How much is shipping to Australia?"

Recommended actions:

  • Provide shipping details from the knowledge base
  • Include tracking information or how to find it
  • Mention free shipping thresholds if applicable
  • Direct them to DMs for order-specific tracking

Leveraging Intent Data for Sales

Intent detection is not just about replies — it is a powerful sales intelligence tool:

  1. Identify high-value ad creatives — Ads generating more Purchase and Price Inquiry intents are your best performers
  2. Spot knowledge gaps — A high volume of Product Info intents suggests your ad copy needs more detail
  3. Optimize reply templates — Tailor your knowledge base to address the most common intents
  4. Prioritize response speed — Purchase-intent comments should get the fastest replies to capture the sale

Next Steps

Learn how to configure Moderation Rules to control which comments are hidden and which get AI replies.


Back to Configuration | Back to Documentation | Visit Rypl