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

# File Reader Overview

> Transform documents and data files into structured information for your automation workflows

File Readers understand how your files are formatted and extract data to use in automations. They process different document types and transform unstructured content into structured information that can be used throughout your workflow.

<Info>
  **Maximum File Size**: File Readers can process files up to **250MB** in size. This limit applies to all file types and reader types. Processing time may vary based on file size and complexity.
</Info>

## File Reader Types

Elementum provides specialized File Readers for different document formats and use cases:

<CardGroup cols={2}>
  <Card title="Text File Reader" icon="file-text" href="/workflows/text-file-reader">
    Extract raw text from images, text documents, and PDFs using OCR technology
  </Card>

  <Card title="Purchase Orders Reader" icon="receipt" href="/workflows/purchase-orders-file-reader">
    Process purchase orders with OCR trained specifically on PO fields and formats
  </Card>

  <Card title="Table File Reader" icon="table" href="/workflows/table-file-reader">
    Read Excel and CSV files with support for thousands of rows and multiple sheets
  </Card>

  <Card title="Elementum Intelligence Reader" icon="sparkles" href="/workflows/elementum-intelligence-file-reader">
    AI-powered document analysis for complex text, image, and PDF documents
  </Card>

  <Card title="JSON File Reader" icon="code" href="/workflows/json-file-reader">
    Parse JSON data with AI-assisted field detection and automatic type mapping
  </Card>
</CardGroup>

## How File Readers Work

### Integration with Automations

File Readers are used within [automation workflows](/workflows/automation-system) to process documents when specific events occur:

```
Document Uploaded → File Reader Processes → Structured Data → Automation Actions
```

**Example Workflow:**

1. **Email Received** trigger fires with PDF attachment
2. **File Reader** action processes the PDF
3. **Create Record** action uses extracted data
4. **Send Email Notification** confirms processing

### Processing Pipeline

<Steps>
  <Step title="File Input">
    Documents are uploaded or received through triggers like Email Received or Attachment Added
  </Step>

  <Step title="Content Extraction">
    File Reader analyzes the document using OCR, AI, or structured parsing
  </Step>

  <Step title="Data Mapping">
    Extracted information is mapped to defined fields with appropriate data types
  </Step>

  <Step title="Automation Integration">
    Structured data flows into subsequent automation actions
  </Step>
</Steps>

## Choose the Right File Reader

<AccordionGroup>
  <Accordion title="By Document Type">
    **Text-Based Documents**

    * **Text File Reader**: Simple OCR extraction
    * **Elementum Intelligence**: AI-powered analysis with custom prompts

    **Structured Business Documents**

    * **Purchase Orders Reader**: Specialized for PO processing
    * **Elementum Intelligence**: Custom business document analysis

    **Data Files**

    * **Table File Reader**: Excel and CSV processing
    * **JSON File Reader**: JSON data parsing
  </Accordion>

  <Accordion title="By Use Case">
    **Document Processing Automation**

    ```
    PDF Invoice → Purchase Orders Reader → Create Record → Start Approval Process
    ```

    **Data Import Workflows**

    ```
    CSV Upload → Table File Reader → Transform Data → Update Records
    ```

    **Intelligent Document Analysis**

    ```
    Contract PDF → Elementum Intelligence Reader → Extract Key Terms → Create Alerts
    ```
  </Accordion>
</AccordionGroup>

<Note>
  **Need flexible, prompt-driven analysis instead of fixed field extraction?** If your documents vary in structure or you need summarization, classification, or open-ended reasoning, consider using [Run Agent Task](/workflows/agent-task-automation#file-inputs) with a multimodal agent instead of a File Reader. File Readers are best when you need the same named fields extracted consistently from every document.
</Note>

## Performance Comparison

| Feature         | Text File Reader | Elementum Intelligence | Purchase Orders |
| --------------- | ---------------- | ---------------------- | --------------- |
| **Speed**       | ⭐⭐               | ⭐⭐⭐                    | ⭐⭐⭐⭐            |
| **Accuracy**    | ⭐⭐⭐              | ⭐⭐⭐⭐⭐                  | ⭐⭐⭐⭐⭐           |
| **Flexibility** | ⭐⭐               | ⭐⭐⭐⭐⭐                  | ⭐⭐              |
| **Cost**        | ⭐⭐⭐⭐⭐            | ⭐⭐⭐                    | ⭐⭐⭐⭐            |

## File Reader Configuration

### Basic Setup

1. **Navigate to File Readers** in your App
2. **Click + File Reader** to create a new reader
3. Enter a **name**
4. **Select Document Type** based on your needs
5. Follow the steps for File Reader setup (uploading a file, configuring fields, etc.)
6. **Test with Sample data** to validate accuracy

<Info>
  **Always test File Readers** with sample documents before using in production automations. This ensures accurate field extraction and proper data mapping.
</Info>

### Field Configuration

All File Readers support configuring fields for extraction:

<AccordionGroup>
  <Accordion title="Standard Fields">
    Available across all reader types:

    * **Text** - String values
    * **Number** - Integer values
    * **Decimal** - Floating point numbers
    * **Date** - Date values with format detection
    * **Boolean** - True/false values
  </Accordion>

  <Accordion title="Advanced Fields">
    Specific to certain reader types:

    * **Address** - Structured address information
    * **Phone** - Phone number formatting
    * **Email** - Email validation
    * **Currency** - Monetary values
  </Accordion>
</AccordionGroup>

## Common Integration Patterns

<AccordionGroup>
  <Accordion title="Document Processing Workflow">
    Automate document handling from receipt to completion:

    ```
    Email Received → File Reader → AI Classification → Create Record → Route for Approval
    ```
  </Accordion>

  <Accordion title="Data Import Pipeline">
    Process batch data uploads systematically:

    ```
    File Upload → Table File Reader → Transform Data → Validate → Update Records
    ```
  </Accordion>

  <Accordion title="Intelligent Document Analysis">
    Extract insights from complex documents:

    ```
    Document Upload → Elementum Intelligence Reader → AI Summarization → Create Insights
    ```
  </Accordion>
</AccordionGroup>

## Best Practices

### Choose the Right Reader

Select specialized readers for better accuracy: Purchase Orders Reader for POs, Table Reader for spreadsheets.

### Test with Real Data

Use actual business documents during testing to ensure accurate field extraction.

### Validate Field Types

Ensure extracted data types match your automation requirements.

### Handle Errors Gracefully

Use IF conditions in automations to handle missing or invalid data.

## Advanced Features

<Tabs>
  <Tab title="Multi-Sheet Processing">
    Table File Readers can process multiple Excel sheets:

    * Configure field mapping across sheets
    * Use cell references for dynamic data
    * Handle complex workbook structures
  </Tab>

  <Tab title="AI-Powered Analysis">
    Elementum Intelligence Readers provide:

    * Custom system prompts for specific extraction needs
    * Multimodal AI support (text, images, PDFs)
    * Advanced field detection and validation
  </Tab>

  <Tab title="Batch Processing">
    Process multiple documents efficiently:

    * Use **Repeat For Each** actions for document batches
    * Configure error handling for failed extractions
    * Implement progress tracking for large volumes
  </Tab>
</Tabs>

## Error Handling

### Common Issues and Solutions

<AccordionGroup>
  <Accordion title="Poor OCR Quality">
    **Symptoms**: Inaccurate text extraction from images or PDFs

    **Solutions**:

    * Use higher resolution source documents
    * Consider Elementum Intelligence Reader for better accuracy
    * Implement validation checks in automations
  </Accordion>

  <Accordion title="Missing Field Values">
    **Symptoms**: Expected fields return empty values

    **Solutions**:

    * Verify field names match document structure
    * Test with multiple sample documents
    * Use IF conditions to handle missing data
  </Accordion>

  <Accordion title="Type Conversion Errors">
    **Symptoms**: Data type mismatches in automation actions

    **Solutions**:

    * Review field type configurations
    * Use Transform Data actions for type conversion
    * Implement data validation steps
  </Accordion>
</AccordionGroup>

## Next Steps

<CardGroup cols={2}>
  <Card title="Automation System" icon="gear" href="/workflows/automation-system">
    Learn how to integrate File Readers with automation workflows
  </Card>

  <Card title="AI Services" icon="brain" href="/ai-agents/ai-services">
    Configure AI providers for Elementum Intelligence File Readers
  </Card>

  <Card title="Best Practices" icon="star" href="/getting-started/best-practices">
    Follow proven patterns for document processing workflows
  </Card>

  <Card title="Troubleshooting" icon="life-ring" href="/support/troubleshooting/date-time-handling">
    Get help with common File Reader issues
  </Card>
</CardGroup>

***

*File Readers transform your documents into structured data that powers intelligent automation workflows. Choose the right reader type for your documents and integrate them seamlessly with your business processes.*
