Capture reads a PDF invoice and returns its content as structured data: header data such as invoice number, dates, amounts and parties, and the individual invoice lines. Posting suggestions and classification use this data when an invoice does not arrive as a Peppol or other XML document.
What is extracted
| Level | Fields |
| Header | Invoice number, invoice date, due date, total amount, VAT amount, currency, payment reference and payment information, supplier details, customer details, invoice type |
| Invoice lines | Product number, product or service name, description, quantity, unit price, line total, tax per line, tax rate, unit code |
Ways to use capture
- In the Enrichment API, as the step before posting suggestions and classification.
- On its own through the OCR API, which returns the data as JSON or as Peppol XML data fields.
- In the PDF capture test tool, where you can upload an invoice and see the result.
A captured PDF is still a PDF invoice
Capture gives you structured data from a PDF you have received. The PDF is not delivered as a Peppol e-invoice, and capture does not replace asking suppliers to send e-invoices. See how to get your suppliers to send e-invoices.
Related: Invoice OCR · What is the Enrichment API?
