DE EN

Solutions

Invoice intake through to a posting proposal

An agent handles PDF and e-invoices, Swiss QR-bills and supporting documents. It matches suppliers, accounts and cost centres, applies your rules and hands the result to your ERP.

feld.ai review view: invoice side by side with the extracted fields — invoice number, amounts and IBAN stay traceable to their source in the document

More than reading fields

feld.ai brings different input formats into one workflow. The agent reads header and line-item data, matches suppliers and references, checks required information, and creates a traceable posting proposal. Clear cases can be passed on automatically. Missing or conflicting information is presented for review with its source and an explanation.

How it works

1. Bring every input together: Invoices and supporting documents arrive by upload, email, scan folder, DMS or API. feld.ai handles PDF and image invoices, electronic invoices and Swiss QR-bills.

2. Understand the invoice: The agent captures the supplier, invoice number, date, line items, amounts, IBAN and payment information from the QR code.

3. Apply master data and rules: Suppliers, accounts, cost centres and other reference data are matched. Required information, totals and agreed business rules are checked.

4. Post or review the exception: Clear cases go to the ERP or accounting system as posting proposals. Exceptions remain traceable, with an explanation and source reference.

What sets us apart

One workflow for several formats

PDFs, e-invoices and Swiss QR-bills enter the same downstream process.

Master data matching

Mapping to existing data (supplier IDs, accounts, cost centres)

Review only the exceptions

Missing or conflicting information is presented for targeted review.

Deployment to fit the use case

On our infrastructure in Austria, as a dedicated instance or in your environment.

In German production, 96% of invoices were fully reconciled and more than 90% of line items were posted correctly.

German logistics group · invoice processing, live in Germany since May 2025

View reference

Related solutions

Automated Document Verification: Check invoices for completeness, correctness and plausibility.

Document Classification: Automatically identify and sort incoming documents.

Data Extraction from Files: Extract structured data from any document type.

Industry Solution: Logistics & Freight: Process freight invoices, CMR notes, and customs documents.

Frequently asked questions

Which invoice formats are supported?

feld.ai processes PDF and image invoices, electronic invoices and Swiss QR-bills. Credit notes, delivery notes, reminders and payment advices can enter the same workflow.

What happens to incomplete or ambiguous invoices?

Required data and matching results remain traceable. Clear cases can continue automatically; missing or conflicting information is presented for review with its source and an explanation.

Is there a DATEV integration?

DATEV can be connected through suitable export formats or existing middleware. The appropriate route depends on your current system landscape.

How long does a pilot project take?

We begin with real invoices and agree the review set, metric and acceptance criteria in advance. Timing depends on the formats, rules and target system.

Where is data processed?

The service can run on our infrastructure in Austria, as a dedicated instance or in your environment. Data flows, access and retention are defined for the chosen setup. It does not require an external model API.

Test your own invoice intake

Request a demo
Talk to the founder