Optical Character Recognition (OCR) refers to both the technology and process of reading and converting typed, printed or handwritten text into machine-encoded1. The initial application of OCR can be traced back to technologies involving telegraphy and reading devices for the blind2. Over the last two decades OCR has come a long way and is now used in converting text from paper documents to electronic formats and recognizing hand writing on electronic devices such as smartphones and tablets.
The insurance industry, which continuously accumulates staggering amount of paper and electronic documents, has been one of the biggest beneficiaries of OCR. Without the use of OCR technologies, employees of insurance companies will have to read information on screens or paper documents and manually copy it into other business processing software for further processing. This process involves the risks of human error due to issues such as fatigue, misinterpretation, typing errors etc. Whereas, OCR technology that runs on computers can work non-stop on picking out information from an endless number of documents. It eliminates the need for expenditure on manual data entry human resources and lets employees quickly access data to make business decisions.
In the insurance industry, access to OCR technology was primarily limited through scanning machines or certain large insurance business software in the first decade of the 21st century. Most of the investments came from insurance technology providers that had the ability to invest in the development of their own OCR technology (also known as OCR engines).
However, recently, large technology companies such as Google, Amazon, Microsoft and IBM have opened doors to their OCR technology through public APIs. These APIs are available to all for free for up to a set number of monthly transactions and $1 - $2.50 per thousand transactions from there on depending on volume. For example, Google’s vision API can recognize multiple languages, faces, logos, and landmarks along with performing sentiment analysis and flagging of inappropriate content3. This creates opportunities for upcoming insurance technology disruptors to leverage a cutting-edge complex technology in developing customized and more sophisticated use cases for the industry.
With the greater accessibility afforded, the insurance technology industry focus is shifting towards determining what could be the next possible applications of OCR. Couple that with the democratization of AI along with decreasing prices of faster cloud-based computing and storage capabilities, and the applications are manifest. We could see insurance agents and brokers acquire sophisticated document comparison capabilities to quickly analyze policy documents and report faster to client queries. Claim examiners could quickly scan claim forms and supporting documents to extract key information required to make a decision on a claim. Insurance customers could possibly compare the pricing, coverage descriptions and sub-limits on complex insurance agreements.
Are we on the cusp of the next generation of use cases that are enabled through the use of OCR with other technologies?
BOOK: Schantz, Herbert F. (1982). The history of OCR, optical character recognition. [Manchester Center, Vt.]: Recognition Technologies Users Association. ISBN 9780943072012.↩