Electronic invoice processing is used for two quite different arrangements and the distinction is worth holding on to. One is handling invoice documents digitally: scanning, capture, workflow, digital approval. The other is exchanging structured invoice data with no document to interpret at all. The first mitigates the reading problem; the second removes it.
Digital document handling
The invoice is still a document meant for human eyes, but everything after it arrives happens digitally: capture, matching, routing, approval and payment. This is what most organisations mean and it delivers the majority of the practical benefit, because the delays being removed are handling and routing delays rather than reading ones.
Structured data exchange
The invoice arrives as fields in an agreed format. Nothing is read, so extraction accuracy stops being a question and the match can run immediately. It requires the supplier to participate, which is straightforward for large suppliers with mature systems and often unreasonable for small ones. Most organisations end up running both.
How to combine them sensibly
Target structured exchange at your highest-volume suppliers, since the benefit scales with document count rather than value. Handle everyone else with digital document processing. A handful of suppliers usually accounts for a large share of your invoice count, so a small programme captures most of the available gain.
Questions people ask about electronic invoice processing
Is emailing a PDF electronic invoicing?
It is digital document handling rather than structured exchange. The PDF still has to be interpreted, which is where extraction errors come from.
Which suppliers should we approach first?
Those sending the most invoices, not those with the largest values. The work is per document.
Does structured exchange change our record keeping?
The obligation is unchanged. Make sure the archive is readable independently of the system that received it, which matters more when there is no human-readable document.