Due to the advancement of technology, many data entry clerks no longer work with hand-written documents. Instead, the documents are first scanned by a combined OCR/OMR system (optical character recognition and optical mark recognition,) which attempts to read the documents and process the data electronically. The accuracy of OCR varies widely based upon the quality of the original document as well as the scanned image; hence the ongoing need for data entry clerks. Although OCR technology is continually being developed, many tasks still require a data entry clerk to review the results afterward to check the accuracy of the data and to manually key in any missed or incorrect information.
An example of this system would be one commonly used to document health insurance claims, such as for Medicaid in the United States. In many systems, the hand-written forms are first scanned into digital images (JPEG, png, bitmap, etc.). These files are then processed by the optical character recognition system, where many fields are completed by the computerized optical scanner. When the OCR software has low confidence in a data field, it is flagged for review – not the entire record but just the single field. The data entry clerk then manually reviews the data already entered by OCR, corrects it if needed, and fills in any missing data by simultaneously viewing the image on-screen.
The accuracy of personal records, as well as billing or financial information, is usually of great importance to the general public as well as the healthcare provider. Sensitive or vital information such as this is often checked many times, by both clerk and machine, before being accepted.