Menu

Scouring Documents Quicker with Optical Character Recognition

  • Written by Michael L. Starzec

starzec michaelIf you were to ask me, a court-going litigator, to discuss tech breakthroughs, this would be a dull article. I would extol e-filing, as it allows me those final, final, final edits well after 5:00 p.m. Next, I would mention remote access allowing me to work, as I am now, while waiting for court. Finally, I treasure the ease of converting Word documents to PDF (and vice-versa). As it turns out, that last function is the heart of a revolution in efficiency.

Make Many Documents Few

No, we’re not talking about the first impression you make when you walk in the room. Optical Character Recognition, or OCR, is simply the electronic conversion of typed or printed text into machine coded text. This, in its most basic sense, is what Adobe or other PDF conversion programs do. However, this is OCR’s most basic function. OCR use has not only made e-filing simple and easy, but in its broader applications, it makes possible text-to-speech and text mining. That sounds like technobabble and it is but OCR has changed the most important aspects of a collection firm: court filings, compliance and payment processing.

Our firm undertook a major effort to “tag” documents, having our OCR technology recognize key words in forms and court orders to trigger later events. For example, once an order is scanned, we can run a search by county and specific order type (judgment/alias) and tell the system to undertake a certain task – have an alias summons printed for attorney approval, prepare a report for the client of all the judgments from Washington County, etc. Staff time is saved from manually reviewing and coding on each individual file. This also allows us to check results. If we go to court with 25 files and postcourt we run a report and find we only have 23 orders, we can easily determine which are missing and what happened.

At the same time, once the OCR has determined an alias needs to issue, it creates the PDF form, telling the system to pull the previously filed complaint to accompany the alias. Then, it combines it with other cases set in the same county on the same date, depositing them in a file folder for review and approval by attorneys. Gone is manual data entry and physical printing of pleadings. Yes, manual reviews by staff and final approval by attorneys takes place, but the heavy lifting of printing, collating and matching documents is largely gone.

Payment Processing Enhanced

The same system affected our payment processing. OCR changed how we look up, present, and process payments. By reducing manual data entry, we sped up payment processing and increased accuracy. We also utilize the inverse of OCR for monitoring compliance in our phone conversations with consumers. Speech-to-text call auditing allows for control reporting to ensure compliance with legal and client standards for better call experiences and reduced audit findings. In addition, it catches items not coded properly – the consumer claimed fraud but we coded as a dispute – One requires an affidavit and investigation, the other requires validation.

However, for any system of data automation to work, you have oversight and controls. The key to this process is control reporting. Our internal audit process verifies everything that should have printed and every payment received is properly credited. Hence a robust internal compliance procedure and team is key to ensuring an OCR based system works.

In effect, for lawyers, words now serve two different purposes. On the one hand to persuade and on the other to provide a mechanical framework to ease the workload on your office.


Michael L. Starzec is a partner with Blitt and Gaines, P.C and is vicepresident of the Illinois Creditors Bar. He is a frequent speaker, writer and litigator on creditor’s rights.