Cognitive OCR Solution for Invoice Processing System



A leading procurement platform provider that offers full procurement exchange solutions for customers in the healthcare industry.

The Business

The customer provides a procurement exchange platform for hospitals and suppliers (vendors). They take the purchase orders issued by the hospitals and invoices raised by the suppliers, and transform these documents through their online procurement platform. The procurement exchange platform allows the buyers and suppliers to transact their business seamlessly.

the Challenge

  • Vendors submit their invoices to the Procurement Platform in different formats. Normalization of the multiple document formats was a challenge.
  • In spite of the available electronic solutions, around 80% of all invoices, remittances and payments continue to be on paper.
  • Often 40-60% of invoices come by fax or email and the rest come in paper format.
  • The customer used manual processes to identify invoices from each vendor and review the invoice details against the respective purchase orders.
  • Manual invoice verification created unnecessary delays in completing the payment cycle.
  • Frequent occurrence of invoice data mismatch and errors increased the manual dependency.

the requirement

  • Reduce manual work by automating the invoice identification and verification workflow.
  • Allow the vendors to submit their invoices in their preferred format. That means, even if they submit paper invoices, they should be able to scan and send it via email or FTP server as PDF files or image files.
  • The solution must recognize all selectable and non-selectable characters in the invoice, such as text as well as images to support rapid invoice processing.
  • Ability to understand any invoice duplication.
  • Differentiate the invoices from two vendors having the same name.
  • Highly customize and train the Abbyy Optical Character Recognition software for seamless invoice processing.


  • Implemented automatic invoice identification and verification workflow using the Optical Character Recognition software - Abbyy.
  • Customized and trained the Abbyy software to enable smarter invoice processing.
  • That is, each time a new vendor is added to the procurement system, the system is trained to understand the vendor’s invoice format.
  • Next time onwards, the system self-performs the invoice verification of that vendor.

When a vendor submits an invoice to the customer, the invoice is processed through an integration hub. Here, the invoices submitted in any format by the vendors are normalized by the system to capture the details. At this point, the OCR feature performs a thorough scanning of the invoice to recognize the vendor, the line items, vendor ID, price, authorized signature, any images (logo) and other invoice specifics, thus enabling the system to perform the invoice verification automatically.

Along with the automatic invoice identification and verification, the flow is designed to allow manual review, at certain steps, as required.


  • Enables hospitals to take full advantage of the early payment discounts.
  • Vendors get the maximum advantage with shorter invoice processing cycle.
  • Operational expenditure reduced with the automation of invoice processing.
  • Savings in costs and time improved the efficiency of admins in handling the invoices.
  • Eliminated the problem of misplaced, mishandled or forgotten invoices.
  • Automatically capture invoices regardless of their format or medium.
  • Ensured complete visibility at the hospital accounts payable.

Results at a Glance

  • 50% reduction in cost per invoice
  • 100% streamlined invoice approval cycle
  • 5x optimized invoice capture with minimal investment & disruption

Download pdf

Please fill the form to receive the case study in your mailbox.

Thank you!!!

The case study will be mailed to you shortly.

Contact us
Close Get In Touch

Interested in our services? Let us get in touch to offer customized services !

Thank You. Your message has been successfully submitted.
Oops, an error occurred! Please try again.

© 2005-2018 Zerone Consulting Private Limited. All Rights Reserved