Scanning and storing an invoice is the easy 80% of invoice archiving. The valuable 20% is knowing which specific bank transaction actually paid it -- automatically, not by someone scrolling through a bank statement matching dates and amounts by eye every month.
Why this needs to be more than "amount matches, ship it"
A naive match on amount alone breaks constantly in practice: a marketplace payout that bundles dozens of orders into one lump sum, a payment that's slightly off due to a bank fee or an early-payment discount (Skonto), or two unrelated invoices that happen to share the same amount in the same week. A matching system that only checks "does this number appear in the bank statement" produces false matches often enough to be actively untrustworthy.
What real automatic matching needs
A priority-ordered strategy, not a single check: first try to find the invoice or reference number itself inside the bank transaction's description text (the strongest signal), then fall back to exact amount within a reasonable date window, then a looser vendor-keyword match within a wider window, then flag it for a manual pick if nothing confident matches. And it needs to catch net-vs-gross mismatches (a VAT rate that doesn't add up) and early-payment discounts as their own category, not silently treat them as an unexplained matching failure.
How Vedron handles this
Vedron's invoice archiving pipeline starts with OCR extraction on upload, with a fallback chain so a difficult scan still gets read rather than failing outright. Automatic bank matching then runs a priority-ordered strategy: extracted invoice/reference numbers matched directly against bank transaction descriptions first, then exact-amount matching within roughly two weeks, then a vendor-keyword match within a wider window, and a same-date-only fallback for manual review when nothing else is confident. Net/gross VAT mismatches and Skonto discounts are flagged specifically rather than silently causing an unexplained gap. The same matching pipeline is shared across invoice types (including receipts), so it's one consistent system, not a patchwork of separate logic per document type.
Frequently asked questions
What happens when nothing matches automatically?
It's surfaced for a manual pick rather than left silently unmatched -- the goal is confident automatic matches where the evidence supports it, and a clear fallback where it doesn't, not forcing a match that isn't actually reliable.
Does this handle invoices in different currencies?
The core exact-amount strategy is EUR-specific; other currencies rely more on the reference-number and vendor-keyword strategies rather than exact-amount matching.
Can OCR handle a poor-quality scan?
There's a fallback chain rather than a single OCR pass, so a scan that a simple regex-based extraction struggles with gets a second attempt through a more capable extraction method before being flagged for manual entry.
Have a question specific to your setup?
Every seller's marketplace mix and stock setup is a little different -- if you're weighing this against your own, book a short call and we'll walk through it with you directly rather than leaving you to figure it out from a blog post alone.
Beta offer: if you join as a beta client, custom integration setup -- configured for your actual marketplace and inventory mix -- is free for your first 2 months.
Still evaluating and need more time than the standard trial window? Just ask us for an extension -- we're happy to give you the room to actually test it properly before deciding.