### Answer To make a single prompt chat agent gather information like first name, date, car, etc., you need to structure your prompt with clear task instructions that guide the agent to ask one question at a time. [(1) ]() Here's how to set it up: 1. Use Sectional Prompts with Task Instructions Structure your prompt to explicitly tell the agent what information to collect:

## Identity
You are a friendly AI assistant for [Company Name].
Your role is to collect customer information for scheduling.

## Style Guardrails
Be concise: Keep responses under 2 sentences unless explaining complex topics.
Be conversational: Use natural language, contractions, and acknowledge what the caller says.

## Response Guidelines
Ask one question at a time: Avoid overwhelming the caller with multiple questions.
Confirm understanding: Paraphrase important information back to the caller.

## Task Instructions
1. Greet the customer
2. Ask for their first name
3. Ask for the date they need
4. Ask what car they have
5. Confirm all collected information before proceeding

2. Extract the Collected Data Retell AI supports extracting dynamic variables from conversations. You can configure Post-Call Analysis to extract specific data fields like: [(2) ]() - **Text**: For names, car models - **Date**: For appointment dates - **Enum**: For predefined lists (services, products, etc.) 3. Key Prompt Engineering Tips - Explicitly list the information you need collected in your Task Instructions section [(1) ]() - Include "Ask one question at a time" in your Response Guidelines to prevent overwhelming users - Add "Confirm understanding: Paraphrase important information back to the caller" to verify accuracy

​ To learn more, see the sources I considered: [[1] Prompt Engineering Guide](https://docs.retellai.com/build/prompt-engineering-guide) [[2] Platform Changelogs | Retell AI](https://www.retellai.com/changelog) ​

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

### Answer Post-call/chat data extraction automatically analyzes conversations after they end to pull out specific information you define. [(1) ]() Step 1: Navigate to Post-Call Analysis Go to the agent detail page and click on the "Post-Call Analysis" tab. [(2) ]() Step 2: Choose Your Analysis Category Type You can extract the following types of data: [(1) ]() - Boolean (True/False) : Simple yes/no determinations. Example: Whether the customer is a first-time caller - Text (String) : Detailed textual information. Example: Call summaries, action items, or key discussion points - Number (Numerical value) : Quantitative measurements. Example: Transaction amounts or satisfaction scores - Selector (Enum) : Categorization from a fixed list. Example: Issue types, product categories, or resolution status Step 3: Configure Each Field For each piece of information you want to extract, provide: [(2) ]() - **Name**: A short name to reference this variable (e.g.,

first_name

,

car_model

,

appointment_date

) - **Description**: A brief explanation of what this value should be Example Configuration for Text: - **Name**:

first_name

- **Description**: Extract the customer's first name from the conversation Accessing the Extracted Data The extracted values will appear in the webhook payload at the end of the webhook data. [(3) ]() Note: Post-call analysis fields will not be populated for calls/chats that were not connected or where no conversation took place. Check whether the field exists before using it. [(1) ](

​ To learn more, see the sources I considered: [[1] Post Call Analysis Overview](https://docs.retellai.com/features/post-call-analysis-overview) [[2] Define the information you want to extract](https://docs.retellai.com/features/post-call-analysis-create) [[3] Post Call Extraction](https://discord.com/channels/1186424376300679209/1447875703813963787) ​

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