### Answer The "chat successful" determination in post-call/chat analysis is controlled by the

analysis_successful_prompt

field on your agent. [(1) ]() This prompt defines the criteria the system uses to evaluate whether the interaction was successful. For example, the default prompt is: > "The agent finished the task and the call was complete without being cutoff." You can customize this prompt to match your own success criteria. [(2) ]() To update it: - In the Dashboard: Navigate to the agent's

Post-Call Data Extraction

section and edit the prompt that evaluates whether a call or chat was successful. - Via API: Use the

analysis_successful_prompt

field when updating your chat agent. [(1) ]() After updating the prompt, you can rerun the analysis to regenerate results based on your revised criteria. [(2) ]() Note: When you rerun analysis, the system always uses the prompts from the latest draft version of the agent—even if the chat was originally handled by an older published version.

​ To learn more, see the sources I considered: [[1] Update Chat Agent](https://docs.retellai.com/api-references/update-chat-agent) [[2] Rerun Post-Call/Chat Analysis](https://docs.retellai.com/features/rerun-call-analysis) ​

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

### Answer No, the

analysis_successful_prompt

(which determines whether a chat is marked "successful" or "unsuccessful") does not control whether post-call/chat data extraction works. Post-call analysis will still run and extract data (Boolean, Text, Number, Selector fields) regardless of the success determination. [(1) ]() The "chat successful" field is just one of the built-in analysis categories—it doesn't gate or prevent other data extraction from occurring. However, there is one important note: Retell AI will not populate custom post-call analysis fields for calls/chats that were not connected or where no conversation took place. So if there's no actual conversation, extraction won't run—but that's unrelated to the success/unsuccessful determination. In short: - "Chat successful" prompt → determines the success/failure label only - Data extraction → runs independently as long as a conversation occurred

​ To learn more, see the sources I considered: [[1] Post Call Analysis Overview](https://docs.retellai.com/features/post-call-analysis-overview) ​

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

### Answer Based on the available sources, here are the common reasons post-chat data extraction may not work: 1. No conversation took place Post-call analysis will not populate custom fields for chats that were not connected or where no conversation occurred. [(1) ]() 2. Model selection issue One community member reported getting "Error in parsing JSON for call analysis" with certain models. Switching from Gemini 3 Flash to GPT 4.1 or GPT 4.1 mini resolved the issue. [(2) ]() 3. Check if the field exists You should verify whether the field exists before using it, as fields won't be populated for calls/chats without actual conversation. [(1) ]() To troubleshoot further, I'd need more details: - Are you seeing a specific error message? - Did a conversation actually take place in the chat? - What model are you using for

post_chat_analysis_model

? (Options include

gpt-4.1-mini

,

gpt-4.1

, etc.) [(3) ]() - Can you share a chat_id where extraction failed? If you're seeing the JSON parsing error specifically, try changing your post-chat analysis model to GPT 4.1 or GPT 4.1 mini. [(2) ]()

​ To learn more, see the sources I considered: [[1] Post Call Analysis Overview](https://docs.retellai.com/features/post-call-analysis-overview) [[2] Post-call extraction error](https://discord.com/channels/1186424376300679209/1465667130187059382) [[3] Get Chat Agent](https://docs.retellai.com/api-references/get-chat-agent) ​

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

### Answer Based on the information you've provided (no error, conversation occurred, using

gpt-4.1-mini

, specific chat ID) , the available sources don't contain specific troubleshooting steps for this exact scenario. However, here are a few things to verify: 1. Check your

post_call_analysis_data

configuration Ensure you have properly defined the extraction fields on your agent. The

post_call_analysis_data

field should contain your custom extraction definitions with the correct

type

,

name

, and

description

. [(1) ]() [(2) ]() [(3) ]() 2. Verify the analysis model setting Confirm

post_call_analysis_model

is set correctly—`gpt-4.1-mini` is a valid option. 3. Try rerunning the analysis You can rerun post-call/chat analysis from the dashboard to regenerate results. Since the sources I have access to don't cover the specific case where extraction silently fails with no error message despite a valid conversation, I'd recommend: - Check out these potentially useful sources: [Post Call Analysis Overview]() , [Update Chat Agent]() , [Rerun Post-Call/Chat Analysis]() Would you like me to try to answer generally based on my knowledge, or would you prefer to reach out to Retell support with your specific chat ID for investigation?

​ To learn more, see the sources I considered: [[1] Update Voice Agent](https://docs.retellai.com/api-references/update-agent) [[2] Get Voice Agent](https://docs.retellai.com/api-references/get-agent) [[3] Create Voice Agent](https://docs.retellai.com/api-references/create-agent) ​

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

### Answer Based on our previous conversation, you're using a Chat Agent but I was previously referencing Voice Agent API documentation. For Chat Agents, you need to configure

post_chat_analysis_data

(not

post_call_analysis_data

) . [(1) ]() [(2) ]() Here's what to verify for your Chat Agent: 1. Confirm your extraction fields are configured correctly The

post_chat_analysis_data

field should be set on your chat agent with proper definitions: [(1) ]()

json
"post_chat_analysis_data": [
  {
    "type": "string",
    "name": "customer_name",
    "description": "The name of the customer.",
    "examples": [
      "John Doe",
      "Jane Smith"
    ]
  }
]

2. Check data storage settings One Discord user discovered that having data storage disabled prevented post-call analysis from working. When they disabled the "opt out of recording" setting, analysis started working. [(3) ]() Check your

data_storage_setting

field—if it's set to

basic_attributes_only

, transcripts won't be stored and analysis may not run properly. [(1) ]() [(2) ]() 3. Try rerunning the analysis From the dashboard, you can rerun post-chat analysis to regenerate results. [(4) ]() If none of these resolve the issue, I'd recommend reaching out to Retell support with your chat ID (

chat_fc2935ce5a3e32defc7a5f26f97

) for direct investigation.

​ To learn more, see the sources I considered: [[1] Create Chat Agent](https://docs.retellai.com/api-references/create-chat-agent) [[2] Get Chat Agent](https://docs.retellai.com/api-references/get-chat-agent) [[3] Post-Call Analysis](https://discord.com/channels/1186424376300679209/1339583616665255968) [[4] Rerun Post-Call/Chat Analysis](https://docs.retellai.com/features/rerun-call-analysis) ​

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