Extractor design

How do I decide which document types need separate extractors?

Direct answer for teams evaluating document automation workflows.

Short answer

Use separate extractors when document types require different fields, validation, review rules, or downstream destinations; reuse one extractor when only the layout changes.

Direct answer

A separate extractor is useful when the business meaning of the document changes, not merely because the layout is different.

If documents need different fields, row structures, validation rules, review queues, or export destinations, they likely deserve separate extractors or workflow branches.

When one extractor is enough

If many senders provide the same kind of document with the same required fields, one extractor with good instructions and validation may be easier to maintain than many templates.

Use classification to route truly different documents and custom instructions to handle variations within the same document type.

How Lido helps

Lido can classify documents and route them to different extractors while still allowing similar documents to share a common schema.

That helps teams avoid both extremes: one overloaded extractor for everything or hundreds of unnecessary templates.

Example workflow

  1. Compare the required fields for each document type.
  2. Check whether validation, review, or export requirements differ.
  3. Use one extractor for layout variation within the same business document type.
  4. Use separate extractors or workflow branches for materially different document types.

Built for real document workflows

Need to turn messy documents into clean spreadsheet-ready data?

Lido helps teams extract, review, and automate data from PDFs, forms, invoices, statements, and other recurring document workflows.

Talk to Lido