Document capture technology is used to distill information from a paper or digital document and integrate it into a workflow. The idea is to simply capture the data to get where it needs to be to perform an operational task such as approval of invoices, sales orders or customer maintenance. In this blog we will go through the phases to achieve this.
The automated process starts with a scanned, imported or e-mailed image of a document, such as a PDF. Documents can be scanned at a central location or from different locations - whichever fits best with the existing business model.
The software checks the incoming e-mails, automatically identifies each document type and captures important data elements to sort and distribute documents and their attachments to the right person or network location.
For example, in the case of invoices, suppliers can email them as PDF attachments to be automatically diverted to the queue to process supplier administration.
The information is therefore not stuck in someone's mailbox or inbox, avoiding bottlenecks or delays. The time previously lost searching or waiting for information can now be spent on more valuable tasks.
Documents can also be automatically identified, sorted and routed to the right person or location by applying various classification techniques. This includes recognizing specific characteristics of the document which correspond to a library of images, or even 'reading' the document to determine its content.
For example, it is possible to distinguish invoices from credit notes, to separate an identity card from a contract, to separate successive documents in 1 pdf, ... Thanks to this technology, your staff no longer has to do manual sorting.
Previously, the use of templates was the only way to automatically recognise data. Today, our technology also processes unstructured or semi-structured documents (where the location of the data on each document is different). The software scans the entire document and automatically records data such as an invoice number, a customer code, an amount, a file number, article details, terms and conditions, and so on.
Current software solutions train themselves via Machine Learning and can therefore remember where the data is located when, for example, you receive a new invoice from the supplier.
Once localised, the software solution applies the correct extraction logic to extract the data automatically.
The automatically recognized data can then be validated to ensure accuracy, based on the applicable business rules within the organisation. For example, it is determined whether the data complies with the expected format or with the information known within the company's own systems (such as suppliers, order numbers, deliveries received, open files, ...).
User-friendly solutions make it possible to apply different rules. The solution is therefore easy to adapt to existing processes so that integration with an accounting solution is not a problem.
The best solutions also make it possible to look up data already contained in a company's existing databases, such as financial systems, ERPs and existing ECM systems. This link automatically replenishes missing data to make further processing much easier and faster.
Once all relevant data of the documents are available and correct, they can be transferred to the back-end or ERP system, with or without the accompanying documents. Often the documents are also stored in a document management system (such as SharePoint).
This normally happens automatically at the right time. In this way, everything is available to the right people at the right time.
At this moment, a workflow can be started. For example, invoices can be routed immediately after verification for approval and fast/correct payment.
From a paper document in the letterbox or received e-mail to the final processing in your computer system, a lot of time-consuming operations are involved in the phases that documents go through. You can save money by automating these processes. RecoMatics uses smart software like EasyForm Capture to make your business processes run smoother, more correct and faster.