> ## Documentation Index
> Fetch the complete documentation index at: https://www.bolna.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Testing extractions

> Test Bolna extractions against sample or real call transcripts, read the test results, and refine prompts before deploying to production.

Before deploying extractions to production, test them against sample or real transcripts to validate accuracy and refine your prompts.

<Frame caption="Test Extractions panel with a sample transcript loaded">
  <img src="https://mintcdn.com/bolna-54a2d4fe/K8nkTcLHvQdEi_0R/images/extractions/test-extractions.png?fit=max&auto=format&n=K8nkTcLHvQdEi_0R&q=85&s=94e3684f0306df322009588f8020a0d7" alt="Test Extractions side panel with Sales Call, Support Call and Appointment sample transcript buttons, Paste and Import tabs above a transcript text area, an Override LLM model for test run checkbox and a Run Test button" width="2880" height="1800" data-path="images/extractions/test-extractions.png" />
</Frame>

## How to test extractions

<Steps>
  <Step title="Click Test Extractions">
    In the Extractions section, click the **Test Extractions** button to open the testing modal.
  </Step>

  <Step title="Choose a Transcript">
    Select from sample transcripts (Sales Call, Support Call, Appointment) or provide your own:

    * **Paste** - Paste a transcript directly into the text area
    * **Import** - Upload a transcript file
  </Step>

  <Step title="Run Test">
    Click the **Run Test** button to process the transcript through all your extraction templates.
  </Step>

  <Step title="Review Results">
    View extraction results organized by category. Each extraction shows:

    * **SUBJECTIVE** - Free text responses generated by the LLM
    * **OBJECTIVE** - Pre-defined values selected by the LLM
  </Step>
</Steps>

<Frame caption="Extraction Results after running a sample transcript">
  <img src="https://mintcdn.com/bolna-54a2d4fe/vlPn4lzPQ372t9yX/images/extractions/test-extractions-result.png?fit=max&auto=format&n=vlPn4lzPQ372t9yX&q=85&s=aa228e684077d99dc9b8646c9d745ae4" alt="Test Extractions panel scrolled to an Extraction Results section showing Extracted Data for the Lead Qualification category and Lead Quality extraction with subjective null, objective hot_lead, confidence 0.9, confidence label High and a reasoning objective paragraph" width="1952" height="5680" data-path="images/extractions/test-extractions-result.png" />
</Frame>

## Understanding test results

Extraction results are displayed hierarchically by category. Each extraction shows its full result including confidence and reasoning:

```
CATEGORY NAME
  ├─ EXTRACTION NAME
  │   ├─ SUBJECTIVE: [Free text response]
  │   ├─ OBJECTIVE: [Pre-defined value or null]
  │   ├─ CONFIDENCE: 0.92 (High)
  │   ├─ REASONING (SUBJECTIVE): [LLM explanation]
  │   └─ REASONING (OBJECTIVE): [LLM explanation]
```

**Example:**

```
VISIT DETAILS
  ├─ RESCHEDULED VISIT TIME
  │   ├─ SUBJECTIVE: 14:00
  │   ├─ OBJECTIVE: null
  │   └─ CONFIDENCE: 0.89 (High)
  ├─ RESCHEDULED VISIT DATE
  │   ├─ SUBJECTIVE: 24/03/2026
  │   ├─ OBJECTIVE: null
  │   └─ CONFIDENCE: 0.91 (High)
```

<Info>
  `null` appears when an extraction type isn't configured or no matching value is found. Low-confidence results (below 0.5) are worth reviewing manually.
</Info>

## Testing best practices

<AccordionGroup>
  <Accordion title="Test with Multiple Transcripts" icon="list">
    Test your extractions against various conversation types to ensure they work across different scenarios.
  </Accordion>

  <Accordion title="Validate Both Answer Types" icon="check-double">
    If using both free text and pre-defined answers, verify both are extracting correctly.
  </Accordion>

  <Accordion title="Refine Based on Results" icon="pen">
    If results aren't accurate, edit your extraction prompts and conditions, then test again.
  </Accordion>

  <Accordion title="Use Real Transcripts" icon="file-lines">
    For production validation, test with actual call transcripts from your agent.
  </Accordion>
</AccordionGroup>

## Sample transcripts

The Test Extractions interface provides three sample transcript types:

* **Sales Call** - Conversation about products and services
* **Support Call** - Customer support interaction
* **Appointment** - Scheduling and booking conversation

These samples help you quickly validate extraction logic without needing your own transcripts.

## Next steps

<CardGroup cols={2}>
  <Card title="Output format & results" href="/docs/prompting/extraction-output" />

  <Card title="Extract data from conversations" href="/docs/prompting/using-extractions" />
</CardGroup>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.