Automation and integration
Intelligent document processing
Intelligent document processing (IDP) is the use of OCR and AI models to read documents such as invoices, forms and contracts, extract the fields you need and pass them into your systems.
How it works in practice
Older template-based capture needed a separate set-up for each document layout. Current IDP uses language and vision models that can read a layout they have not seen before, find the supplier, dates, totals and line items, and return them as structured data with a confidence score for each field.
The confidence scores are what make it usable. Fields above a threshold flow through; low-confidence fields go to a review queue where a person checks only the uncertain part. Validation against your own records, such as matching a PO number or checking the totals add up, catches most errors before they reach the ledger.
A worked example
A 60-person contractor receives about 900 supplier invoices a month as PDFs and photos. An IDP pipeline extracts header and line data, matches each invoice to its PO, and sends only mismatches and low-confidence fields to the accounts assistant.
Examples in this glossary describe typical builds for companies of 1 to 1,000 people. They are not named client results.















