​ To learn more, see the sources I considered: [[1] Consume the analysis data](https://docs.retellai.com/features/post-call-analysis-consumption) [[2] Webhook Overview](https://docs.retellai.com/features/webhook-overview) ​

### Answer --- Post-call analysis fields will not be populated for calls that were not connected or where no conversation took place. [(1) ]() However, based on what you're describing (user hangs up after 5 seconds, but analysis fields still get populated) , the system may still be running analysis on very short calls where technically a "conversation" occurred (even if minimal) . A few approaches to handle this on your end: 1. Check field existence before using — The documentation recommends checking whether the analysis field exists before using it. 2. Use call metadata for filtering — When you receive the

call_ended

or

call_analyzed

webhook, you can filter based on call duration or disconnection reason before processing the analysis data. The

call_ended

webhook includes all fields from the call object. [(2) ]() 3. Filter by disconnection reason — Calls with disconnection reasons like

dial_failed

,

dial_no_answer

, or

dial_busy

won't trigger

call_started

but will still trigger

call_ended

and

call_analyzed

webhooks. For your specific use case, I'd recommend adding logic in your webhook handler to check the call duration or transcript length before processing the post-call analysis responses—this way you can discard analysis from calls that ended too quickly to have meaningful conversation.

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.