https://retellai.com/?utm_source=discord&utm_medium=seo logo
I have a question regarding webhooks vs get reques...
# ❓・ai-evy-support
h
@AI Evy I have a question regarding webhooks vs get requests. I’m building post-call artifacts (structured PCA + generated PDFs) from Retell calls, including long calls in the 30–60 minute range. We currently consume the call_analyzed webhook and run our own normalization on that payload, but in testing we’ve observed that replaying the same call via the GET /call/{id} endpoint can sometimes return a more complete or updated representation of the call (e.g. additional analysis fields or enrichment), especially when analysis is rerun or for longer calls. There doesn’t appear to be an explicit contract in the docs that webhook payloads are always identical to the GET response at that point in time. For correctness and auditability, would Retell recommend treating the webhook purely as a trigger and fetching the call via GET after call_analyzed before generating final artifacts, or is the call_analyzed webhook intended to be a fully authoritative final state suitable for downstream compliance-critical processing?
a
Thanks for tagging me @huskinn. I'll look through relevant sources and get back to you shortly.
### Answer --- The documentation doesn't explicitly state whether the
call_analyzed
webhook payload is guaranteed to be identical to what the Get Call API returns at that moment. [(1) ]() Here's what the docs do clarify: Webhook payload for `call_analyzed`: The
call_analyzed
event is triggered when call analysis is complete and includes the call object with the
call_analysis
field populated. [(1) ]() [(2) ]() Get Call API: The Get Call endpoint returns all fields from the call object including
call_analysis
. [(3) ]() Documented consumption methods: The docs present three methods to consume analysis data—Dashboard, Webhook, and Get Call API—as separate options, noting for the API method: [(2) ]()
Copy code
json
{
  "call_id": "123",
  "call_analysis": {
    // Analysis results object
  }
}
Given the behavior you've observed (GET sometimes returning more complete data, especially for longer calls or rerun analysis) , and the absence of an explicit contract guaranteeing webhook/API parity, treating the webhook as a trigger and fetching via GET before generating compliance-critical artifacts would be the safer approach for auditability. For a definitive answer on whether this is the recommended pattern, I'd suggest reaching out to support@retellai.com or [Contact Sales]() .
​ To learn more, see the sources I considered: [[1] Webhook Overview](https://docs.retellai.com/features/webhook-overview) [[2] Consume the analysis data](https://docs.retellai.com/features/post-call-analysis-consumption) [[3] Get Call](https://docs.retellai.com/api-references/get-call) ​
Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.
Thanks for the feedback, glad I was helpful.