Guides

Practical writing on accounts payable automation — accuracy, duplicates, matching, fraud — from the team behind DVAP. Numbers over adjectives.

• PINT A-NZ e-invoice validation: what gets rejected, and why

  A UBL e-invoice that breaks the PINT A-NZ rules is rejected after it leaves your system. The rules that actually fail, why arithmetic causes most of it, and how to check first.

  2026-07-30 · DataVance

• How to validate an ABN properly (the checksum is only half of it)

  An ABN can pass its checksum and still belong to a business wound up years ago, or to someone other than the supplier who gave it to you. The three separate questions worth asking.

  2026-07-30 · DataVance

• Invoice OCR API accuracy: what the numbers mean, and why almost nobody publishes them

  “99% accurate” means nothing without a named benchmark and per-field numbers. How to read invoice OCR accuracy claims — and the numbers DVAP publishes.

  2026-07-24 · DataVance

• Duplicate invoice detection: how businesses end up paying twice, and how to stop it

  Most duplicate payments aren't fraud — they're the same invoice arriving twice through different doors. Where duplicates come from and how to catch them before the payment run.

  2026-07-24 · DataVance

• 2-way vs 3-way PO matching, explained with the exceptions that actually fire

  Matching an invoice to its purchase order is how you stop paying for things you never ordered. What 2-way and 3-way matching each catch, and when you need which.

  2026-07-24 · DataVance

• How to catch invoice fraud before the payment run

  Invoice fraud succeeds because the document looks legitimate. The patterns that get invoices paid to the wrong account — and why screening must happen before the payment run, not after.

  2026-07-24 · DataVance