Overview:

Workbench has partnered with an image recognition and processing service called Xtracta. This allows companies to train the interface to recognise fields in repeat supplier invoices resulting in time-saving automation for high volume purchasing. The image recognition ties in to Workbench's AP Voucher module, and as such they are often explained in tandem.

The goal of using Xtracta is to significantly reduce the time required when processing AP Invoices. By scanning header and line-level detail through Xtracta we can automatically create Vouchers. In the voucher, the user can then link the lines from the original pdf document to Purchase Order lines, Docket lines, or even create manual voucher lines. The original PDF is attached to the voucher screen removing the need for a user to print out the document or open the document in another window.

For the technical setup guide, please see Data Extraction Downloader.

This screen is used to review and process documents that have been emailed to the Xtracta email address.

• This list displays the status of all image documents received:
  Image Added
  
  
• The list provides entry to any remedial action for data extraction. There is no notification for rejected image documents as it is assumed that this is a significant role with a semi / dedicated user.
• The Usage chart depicts the number of document used from the current bundle and the balance available before top-up of the bundle is required. The email address is displayed for convenience.
• Various filters are available for the list, the default being any Rejected documents as these require attention to progress.
• Rejected documents display a [check] image link that displays the Image Document popup for extraction remedial actions.
• Validated image documents (currently only AP Invoices / credits) display a [view] document link that opens the voucher details popup. This has rows equivalent to the AP Vouchers List.
• A coloured indicator highlights those suppliers whose invoice format is anticipated to require training:
  • Red – Requires 3 or 4 further documents
  • Amber – Requires 2 further documents
  • Yellow – Requires 1 further documents
  • Clear – Should be fully trained
• The Refresh button gives you a view of the current state, instead of waiting for the scheduled downloader refresh.