Frequently Asked Questions

General Questions

What document types do you support?

DocVision supports any document type and format, including:

  • Financial documents: Invoices, receipts, bank statements, purchase orders
  • Business documents: Contracts, forms, reports, statements
  • Formats: PDF, images (PNG, JPG, TIFF), scanned documents
  • Languages: 50+ languages with automatic detection
  • Currencies: Multi-currency support with automatic recognition

The AI Vision models automatically understand document structure regardless of format or layout.

How accurate is the extraction?

Vision OCR+ MAX achieves 99.8% accuracy on average. The models understand document semantics, not just text extraction, ensuring reliable structured data output.

The accuracy varies based on:

  • Document quality and clarity
  • Model selected (MAX vs Lite)
  • Document complexity and structure

For best results, ensure documents are clear and readable. Both models handle scanned documents, photos, and digital PDFs effectively.

How fast is processing?

Average processing time is 30 seconds per document page. Processing speed depends on:

  • Document size and number of pages
  • Model selected (Lite is faster for simple documents)
  • Current system load

High-volume processing is supported with scalable infrastructure that handles millions of documents efficiently.

Do I need to create templates?

No. Both Vision OCR+ models work without templates, manual rules, or training. The AI automatically understands document structure and extracts data based on your Extraction Templates.

However, you can create custom Extraction Templates to define exactly which fields to extract and how the data should be structured. This gives you control over the output format while still leveraging AI-powered extraction.

What output format do I get?

You receive structured JSON data that matches your defined Extraction Template schema. The output includes:

  • All extracted fields with proper data types
  • Field-level confidence scores
  • Document metadata (date, type, etc.)
  • Ready-to-integrate format for your systems and workflows

The JSON structure matches your Extraction Template definition, making it easy to integrate with downstream systems.

What languages are supported?

We support 50+ languages with automatic language detection. The models can process documents in:

  • Major European languages (English, Spanish, French, German, Italian, Portuguese, etc.)
  • Asian languages (Chinese, Japanese, Korean, Hindi, etc.)
  • Middle Eastern languages (Arabic, Hebrew, etc.)
  • And many more

Language detection happens automatically - you don't need to specify the document language.

Pricing Questions

How does pricing work?

DocVision uses simple, transparent per-page pricing:

  • Vision OCR+ MAX: $0.04 per page
  • Vision OCR+ Lite: $0.02 per page

You pay only for pages processed - no monthly minimums, no setup fees, no hidden costs.

What counts as a page?

A page is counted each time a document page is processed:

  • PDF documents: Each page in the PDF counts as one page
  • Image files: Each image file counts as one page
  • Multi-page documents: Each page is billed separately

Is there a free tier?

Yes. New users receive $5 in free credits when you sign up. No credit card required to get started.

Use your free credits to test the platform and process your first documents. Upgrade when you're ready to scale.

Can I switch between models?

Yes. You can choose which model to use per extraction request via the API, allowing you to optimize cost and performance based on document complexity:

  • Use Vision OCR+ MAX for complex documents requiring maximum accuracy
  • Use Vision OCR+ Lite for simple, high-volume documents where cost efficiency matters

This flexibility lets you balance accuracy and cost based on your specific needs.

Do you offer volume discounts?

Yes. For high-volume processing needs, we offer custom enterprise pricing with volume discounts. Contact Sales to discuss enterprise pricing options.

Technical Questions

Is my data secure?

Yes. DocVision provides enterprise-grade security:

  • SOC 2 compliant - Meets industry security standards
  • Regional data hosting - Keep your data in your region (available for enterprise)
  • Encrypted in transit and at rest - All data is encrypted
  • No data sharing - Your documents and data are never shared with third parties

Can I self-host?

Currently, DocVision is available as a cloud-based service. Self-hosting options may be available for enterprise customers. Contact Sales to discuss your requirements.

How do I integrate with my existing systems?

DocVision offers multiple integration options:

  • REST API - Upload documents and retrieve extracted data programmatically
  • Webhook callbacks - Receive extracted data automatically via webhooks
  • Email integration - Send documents to a dedicated email address for processing
  • Structured JSON output - Ready to integrate with any system

See our API Integration documentation for detailed integration guides.

What happens if extraction fails?

If extraction encounters issues:

  • Error notifications - You'll receive detailed error messages via API response or webhook
  • Retry support - Failed extractions can be retried automatically or manually
  • Support assistance - Our team can help troubleshoot extraction issues

The platform includes built-in error handling and validation to ensure data quality.

Model Selection

When should I use Vision OCR+ MAX?

Use Vision OCR+ MAX for:

  • Complex document structures with multiple data types
  • Documents requiring maximum extraction accuracy
  • Irregular layouts or non-standard formats
  • Critical business documents where accuracy is paramount

When should I use Vision OCR+ Lite?

Use Vision OCR+ Lite for:

  • High-volume processing with consistent document types
  • Simple, straightforward extractions
  • Cost-sensitive applications
  • Documents following standard patterns

Can I use both models?

Yes. You can use both models within the same organization and even within the same workflow. Choose the appropriate model per document based on complexity and requirements.

Get Help

Still have questions? We're here to help:

  • Documentation: Browse our documentation for detailed guides
  • Support: Contact support for technical assistance
  • Sales: Contact Sales for pricing and enterprise inquiries