1. Extract
The service reads embedded PDF text first and uses OCR only for scanned pages. It recognises common labelled and section-based invoice layouts, including textual dates and right-aligned totals. Files are limited to 10 pages and 5 MiB.
Free local PDF and OCR extraction
Check a PDF invoice for Irish eInvoicing readiness
Upload a PDF invoice to extract an editable canonical JSON draft using local English and Irish text recognition. No account or API key is required.
2. Review
Compare every extracted value with the original. Confidence indicators and warnings point to fields that need particular attention. Sparse but recognisable invoices may produce an incomplete draft instead of being rejected.
3. Validate
The edited draft can be checked with the same versioned readiness and arithmetic rulesets used for canonical JSON.
Why PDF results are different
A PDF is primarily a visual document, while Revenue describes eInvoices as structured data that supports automated processing. The checker therefore reports on the extracted interpretation rather than treating the PDF as an EN 16931 syntax. Read the structured invoice and PDF guide and follow Revenue’s current VAT Modernisation guidance.
Extract and check a PDF Use the PDF extraction API
Peppol document readiness: UBL invoices and credit notes can be checked with the official Peppol BIS Billing 3.0.21 rules through auto-detection or an explicit API/UI profile. Network transport remains outside the service.
Testing workspace: the browser accepts up to ten documents sequentially, exports an aggregate CSV, and publishes a versioned fictitious scenario pack. The standards status page distinguishes EN 16931-1:2026 from the currently deployed validation artefacts.