Which webhook configurations enable automated question extraction from Zendesk tickets and Gong recording transcripts?

Automated question extraction requires configuring Zendesk ticket triggers on status solved with JSON payloads alongside Gong call transcribed webhooks to route raw customer text and conversation transcripts into ingestion endpoints.

Customer inquiries highlight high-intent buying signals, yet agency research shows manual content extraction costs between $100 and $300 per hour in labor. Connecting support tickets and sales recordings directly to structured question workflows captures recurring customer queries before they disappear in closed threads and unreviewed calls.

If you only do one thing: Target closed conversations by triggering webhook payloads strictly on Zendesk `Status changed to Solved` and Gong `Call Transcript Completed` to prevent duplicate draft generation.

1. Zendesk Trigger Setup: Create a webhook in Zendesk Admin Center pointing to your ingestion endpoint, configured to fire when `Ticket > Status` changes to `Solved` and `Comment > Is Public` equals `true`.
2. Zendesk Payload Structuring: Define a JSON (JavaScript Object Notation) payload containing `ticket.id`, `ticket.title`, and `ticket.latest_public_comment` to capture the customer inquiry and final resolution pair.
3. Gong Webhook Subscription: Register a webhook in Gong API (Application Programming Interface) settings using the `call.transcribed` event, which fires automatically once call audio finishes processing.
4. Gong Transcript Retrieval: Use the webhook payload `callId` to query the Gong `v2/calls/transcript` endpoint, filtering by speaker track to isolate prospect questions from sales rep responses.
5. Ingestion Pipeline: Pass the parsed text into an answer engine like Dae, which structures queries against your verified documentation across a 3-step ingest, plan, and generate workflow.
6. Approval Gate Routing: Route drafts into a moderation queue requiring team review before publishing answers on your domain with structured schema markup.

• Watch out for: Unfiltered Zendesk triggers that fire on every internal ticket comment, creating redundant drafts and exceeding webhook rate limits.
• Watch out for: Missing PII (Personally Identifiable Information) scrubbing rules, which risk transmitting sensitive customer records from raw transcripts into content queues.
• Watch out for: Publishing unverified drafts directly without a two-layer moderation review to ensure technical pricing and policy accuracy.

Map your top 5 support ticket tags in Zendesk and run a test payload with 3 sample tickets to verify that raw customer questions format correctly.