From scanning paper documents to Capture as a Service (CaaS)
It is a fact, customers are increasingly connected. Accustomed to buying goods or services on the Internet from their smartphone, users want to be able to transfer all documents necessary for finalising end-to-end transactions on the same channel. Businesses are no exception. Encouraged by regulatory developments, they also seek to eliminate paper by digitalising their transactions. When documents are received in paper format, they are scanned as soon as possible to be processed in digitised format.
Faced with a multitude of reception modes, all organisations are now confronted with the same problem: how to uniformly deal with documents transferred via these different platforms?
Omnichannel capture software enables automatically processing any type of document, whether it is uploaded to a portal, received as an email attachment or digitised by an outsourcing agent. ADR* engines analyse documents to determine their type while ADC* engines automatically retrieve useful data from images. These processing operations can be carried out either in batch processing (backoffice) or in real-time when loading a document onto a portal. In the latter case, it is called CaaS (Capture as a Service).
*ADR: Automatic Document Recognition
*ADC: Automatic Document Content
ITESOFT supports NetworkRail on finance automation solutions
With a huge amount of invoices going through Network Rail every day and a long processing cycle, Network Rail decided it was time to explore new finance solutions. Their main requirements were a more automated and integrated approach and a similar way of working to limit the impact on users. Their huge efforts during this challenging project have been extremely successful.
Invoice rejection has been decreasing over the last three years, thanks to the information we extract on the ITESOFT platform.
Faced with the multiplication of communication channels (mail, FAX, paper, EDI), companies are interested in automating inbound document processes.
But how do you capture all these very varied documents from very different sources?
Primarily, an omnichannel capture solution must make it possible to scan all paper documents essential to managers, be it an outsourcing agent or various in-house departments. For this purpose we can use desktop scanners or multifunctional copiers. Some organisations have chosen to scan all their inbound documents from large production scanners installed in the digital mailroom for sorting into electronic processing trays.
However, an increasing proportion of documents are now transferred in digital format. These documents can be loaded onto a portal, sent by email with attachments or transferred in the form of image files or EDI (Electronic Data Interchange).
ITESOFT| W4 omnichannal capture solution makes it possible to standardise the process whilst capturing all these documents, enabling companies to offer their clients, employees, partners or suppliers a seamless omnichannel experience.
Automatic document recognition
Regardless of how they are digitalised, documents are typically used to support processing tasks for an entire transaction.
Upon receipt, the first check is to verify that the received document matches the type of document expected. Fortunately, this tedious, low-value task can be automated using the ADR (Automatic Document Recognition) engines.
Based on machine learning technologies, the ADR module automatically classifies a document from the structure or content (i.e. search of keywords, statistical analysis, etc.). The configuration of the engine, accomplished through machine learning, is very simple insofar as it merely supplies the solution with representative samples of each class or type of document.
The ADR function is available for batch processing as in the case of a digitalised mailroom or service mode (CaaS) for online use, for example from a contractual portal.
Automatic document content (ADC)
Documents received generally contain information necessary to process a file or complete the transaction. This data is most often entered directly from the paper document or from its digital image. These tasks are both resource intensive with little or no added value.
The ADC (Automatic Document Content) module automatically extracts the relevant data from a scanned document by combining multiple reading strategies.
In the case of a document structured as a form, “models” can be used to precisely indicate the area containing the data.
For unstructured documents such as invoices or bank account identity, the data may not always be in the same place. ITESOFT advanced full-text strategy uses several advanced algorithms to automatically retrieve document data: format, relative positioning via keywords, relative positioning of the zones between them, etc…
By combining these different reading strategies, the ITESOFT ADC module achieves reading ratios close to 100%, thus considerably reducing input tasks. The ADC function is available for batch processing, as in the case of a digitalised mailroom, or in service mode (CaaS) for online use, for example from a contracting portal or for automatically processing invoices delivered by an outsourcing agent.