Skip to content
Pridonu

Article · March 2026

From invoice intake to the ERP: why handovers fail in practice

Recognition rates get plenty of attention; the last mile rarely does: the handover of checked data to the ERP, accounting or line-of-business systems. Yet this is exactly where it is decided whether automation becomes an end-to-end process – or an island from which someone ends up retyping data after all. The typical stumbling blocks are rarely the interface itself, but the details of the data:

  • Master-data mismatch: the supplier on the invoice is named differently from how it appears in the ERP. Without a match against your creditor master data at intake, every handover becomes guesswork
  • Mandatory fields that are not on the invoice: cost centre, property or order number must be added by the intake process – automatically from the master data or in clearing, not later in the target system
  • Formats and rounding: tax keys, line-item totals and cash discounts must arrive in the way the target system calculates – otherwise discrepancies only surface at posting
  • Exceptions without a return channel: if the target system rejects a record, there needs to be a defined route back into clarification – otherwise a blind spot arises

For the route itself there are three proven options – an API, a software robot working through the existing input screens, or an agreed file format via SFTP. More important than the technology is the agreement about the data: which fields, to what quality standard, and who deals with the exceptions. The litmus test of a handover is not the demo record, but the hundredth invoice with no cost centre.

Three routes into your system: API, RPA, SFTP →

Please note: this article reflects the legal position and administrative guidance as at the date stated and is no substitute for tax or legal advice.

← All articles

Questions about your invoice intake?

We show you, on your own invoices, what an automated intake looks like.