I’ve recently written about our latest advances in document ingestion in our Digital Mailroom. It’s definitely the biggest buzz right now – we’re excited about all the cutting-edge technology we’re bringing to your accounts payable processes. But one part of document ingestion that I didn’t deeply dive into is document indexing, which is admittedly a less flashy process, but no less important. In fact, indexing digital documents with minimal exceptions is one of our most critical steps to ensuring a seamless AP workflow for your team.
Other resources you might find helpful:
Invoices and Documents: From Receipt to Digital Delivery
When we receive your documents – whether by electronic or physical means – we kick off the ingestion process. We retrieve, sort, and prep physical mail and scan the documents. At the same time, we receive your digital documents via email. This is when the indexing process begins. In our Invoice Processing Machine™ (IPM), we classify each document by PO invoice, non-PO invoice, credit, and non-invoice. We then send the digital documents through IPM for OCR data capture of header and line information. This validates the information pulled from the documents and serves it up for human verification by our highly trained experts. These steps are all completed in our 5-step QA process.
Our goal is to provide accurate invoice data so any exceptions can be caught in the automation process and worked by qualified members of your accounts payable staff. We are able to do that by going the extra mile beyond industry best practices for data capture. Our 25 years of experience has given us the tools and know-how to successfully transition from the old-school method of AP processor data capture to artificial intelligence data capture, which provides the highest level of document indexing consistency. This AI capture process not only reads the full text of the image, but also determines what the data is and (through a mix of standard AP rules and tailorable rules) decides if it is valid. This program learns over time to produce better results in this stage.
Consistency is extremely important in document indexing. We use standard naming conventions for document images and file paths, and we use standard data capture on a variety of document types. Data security is also essential to us, and we make sure that all security laws, such as HIPAA, are followed. We protect sensitive information by keeping physical documents under lock and key and storing images on secure servers.
We Know Document Indexing
At DataServ, we process more than 3,600,000 pages of documents per month – and climbing! If we printed all of those documents and stacked them up, that’s about 500 yards of paper, or five football fields! Suffice it to say, we know what we’re doing. And with the launch of our latest invoice processing machine, the risk of human error is greatly reduced, while the true automation potential is growing.
We’d love to see what we can do for your AP documents and workflows.
Want to learn more about our document indexing processes and standards? Contact us.