How do you handle Fin negative ratings?

Hi ​@Audrey  Ebenezer here from Engineering Support👋.

Great question! There are a few built-in ways to automate follow-up on negative experiences and uncover the “why” behind poor outcomes, so you can improve content without manual digging.

🔁 Automate review and escalation for negative experiences

You can use a Workflow trigger based on the CX Score:

• Trigger: “When a conversation receives a CX Score rating.”

• Condition: Score is 1–2.

• Actions: Reopen the conversation, tag it (e.g., Fin-DSAT), assign it to a human queue, and add an internal note with next steps.

🕓 Timing note: The CX Score is usually assigned 2–3 hours after a conversation closes, so your workflow will trigger the same day—no manual monitoring needed.

📊 Understand why DSAT happens (beyond CX Score)

If you want to analyze Fin thumbs-down ratings separately from CX Score:

• Build a custom report using the Conversation rating dataset.

• Add the metric “Fin AI Agent DSAT score” to track negative Fin ratings (😠 🙁 😐).

• Include the attribute “Fin AI Agent rating remark” to see customers’ written reasons for DSAT.

• Filter by “Fin AI Agent rating = negative” to focus on problem cases.

You can also use the CX Score rating explanation attribute for AI-generated insights summarizing why the rating occurred—even if no survey was sent.

💡 Use Optimize to fix what’s causing DSAT

In Optimize (Fin AI Agent → Analyze → Optimize):

• The dashboard highlights where Fin under-performs and offers AI-powered Suggestions to improve content (add, edit, or merge articles).

• These are based on comparing failed Fin responses to successful human replies.

In practice:

• Resolution rate view: Filter to Content gaps → check the Suggestions column for one-click edits/additions.

• CX Score view: Toggle Negative Fin CX → drill into conversations → apply content suggestions directly in Resolution rate.

If you have Optimize v2, you’ll also see detailed breakdowns (e.g., Missing content, data access, or action capability) plus Data and Action Suggestions with automation guidance. If you don’t have access yet, your Intercom contact can help enable the beta.

✅ A streamlined workflow you can set up today

1. Create a Workflow:

   • Trigger = “When a conversation receives a CX Score rating.”

   • Condition = score 1–2.

   • Actions = reopen, tag Fin-DSAT, assign to human queue, add an internal note.

2. Build a Custom Report:

   • Metric = Fin AI Agent DSAT score.

   • Breakdowns = AI Topic/Subtopic, Time.

   • Table fields = Conversation ID, Fin AI Agent rating, and remark.

3. Use Optimize weekly:

   • Resolution rate → Content gaps: Apply Suggestions on top issues first.

   • CX Score → Negative Fin CX: Review problem conversations and feed learnings back into content updates.

This setup automates triage, captures the “why” behind DSAT, and connects insights directly to fix-forward content improvements.