Did you know that over 90% of the total number of invoices issued globally are still processed manually?
Yes, you read that right. Less than 10% of all invoices are automatically processed! This takes a huge toll on AP departments.
But it’s not only time (and thus, money) that is lost when processing invoices manually. You also lose out on tons of data.
One approach to automatically processing invoices is called template-based OCR. This means that lots of rules and templates are in place to automatically read data from business documents.
This works as long as the document scanned follows the specified rules EXACTLY. As soon as something is out of the ordinary… this approach fails.
Each individual alternation of the document requires a long and expensive configuration process.
Another and better approach uses artificial intelligence cognitive data capture. Based on deep neural networks, invoices can be scanned, and patterns can be learned.
The system can adapt to changes in layout and input. The technology infers the general structure of invoices quite like how a human mind does.
Imagine the possibilities: Invoices scanned and processed automatically. Without the need to continuously update the configuration.
#invoiceprocessing #ai #ocr #invoices
Imagine the possibilities: Invoices scanned and processed automatically
Without the need to continuously update the configuration