Actualizado el 23/05/2023

icon Facebook icon Twiiter icon RSS icon EMAIL
  1. Portada
  2. >
  3. TecnonewsWorld
  4. >
  5. The Machine-Readable PDF Invoice that is Minimizing the Need for OCR

The Machine-Readable PDF Invoice that is Minimizing the Need for OCR

Escrito por Redacción TNI el 05/06/2018 a las 20:10:06

Basware, the global leader in networked source-to-pay solutions, e-invoicing and innovative financing services, is introducing a new, smarter way to capture data from PDF invoices automatically with near perfect accuracy. Basware’s Smart PDF invoice capturing service easily extracts data from any machine-readable PDF, creating a touchless invoice automation process.


“With Smart PDF, we are taking the next step in helping our customers realize maximum efficiency and achieve touchless invoice processing. Suppliers simply email PDF invoices to Basware, and our customers receive invoices in their preferred format, without any need for a process change on the buyer or supplier side. Suppliers can email PDF invoices directly from their ERP system. Basware handles the data extraction, content validation and augmentation process automatically by utilizing a range of different technologies including deterministic algorithms, machine learning and optical character recognition,” explains Jussi Erjanti, Product Management Director, Basware.


An analysis of the millions of transactions flowing through the Basware system shows that 70% of PDF invoices are machine-readable, which means they can be processed automatically by extracting the invoice information directly from the PDF file without the need for optical character recognition (OCR) technology. The advantage over OCR is faster processing and less room for character interpretation that can lead to invoice exceptions and human intervention. With machine-readable PDFs, the exact information from the supplier’s ERP system is transmitted directly into the buyers’ purchase-to-pay solution – meaning errors will become a thing of the past. In cases where the PDF is not machine-readable, the invoices are automatically routed to OCR process to guarantee 100% invoice delivery.


“There is an increasing need to address the wide gap between invoices received as structured data in EDI/XML format, and paper invoices that are processed using OCR. We have seen organizations focus on building complex EDI/XML connections with their strategic or high-volume suppliers and then settle for paper invoices and OCR to connect to their long tail. We believe that a better, smarter way to process PDF invoices can address this gap. Sending PDF invoices via email is one of the most used invoicing methods around the world. With Smart PDF we can make it easy for both the buyer and supplier, while maximizing automation and efficiency. Buyers can avoid late payments and optimize their DPO (Days Payables Outstanding) resulting in happier suppliers enjoying shorter DSO (Days Sales Outstanding)” Erjanti continues.


The Basware Smart PDF service has several key benefits:


  • Buyers can connect to all their suppliers – large and small – including the long tail without changing anything on the supplier side.
  • This service can be used with machine-readable PDFs as well as invoice images.
  • Fast and accurate processing prevents late payments, which enables optimizing the buyer DPO (Days Payables Outstanding) and minimizing the supplier DSO (Days Sales Outstanding).
  • The service is self-learning, which means Smart PDF learns from transactions as it goes for continuous improvement, increased automation rates and more efficiency.


“All of this results in significant savings opportunities to our clients by eliminating exceptions. One of our clients estimates a savings of nearly 20 euros per invoice,” Erjanti adds.


The first version of Smart PDF will be available in 2018, with significant feature upgrades planned over the next year.


To learn more about how Basware innovation is empowering companies to simplify operations and spend smarter, visit