When scan-and-populate software first surfaced in the tax and accounting industry more than fifteen years ago, it was viewed as an innovative, paperless solution to help automate PDF bookmarking of scanned documents and decrease data entry into tax software. Although scan and populate solutions have delivered on that promise, the increased efficiency comes with a cost. Since OCR is imperfect, someone must verify OCR-extracted source document data for accuracy (for example, a 5 might be read as an 8). When those OCR errors occur, they need to be corrected.
Firms can identify and correct OCR errors in one of two ways:
- Manually verify the OCR data before exporting to tax software
- Export the OCR data as-is, including errors, and rely on the preparer to find and correct errors in the tax software.
The recommended approach to verifying OCR data
All scan and populate solutions provide a user interface to facilitate the manual verification of OCR data before exporting to tax software. The user glances back and forth between the scanned document and the OCR data that’s being extracted and corrects errors after a visual check. We strongly recommend this approach for two main reasons:
- Since the data is presented to the user, there is no time wasted trying to hunt for errors.
- Since the data is presented on a document by document basis, the verification is a simple, visual check that can be done by interns or administrative staff. The verification requires no knowledge of tax software or how to prepare a tax return.
The manual verification process can take as little as a few minutes for a small return with a handful of source documents to as long as several hours for a return with hundreds of documents. The verification process takes time, but the verification user interface supports an efficient task that can be performed by lower paid staff.
Identifying and correcting OCR errors in the tax software is an inefficient process
Although we discourage it, some firms choose to export the OCR data to tax software without first verifying it for accuracy. This process requires the preparer to search through each form and field in the tax return and compare them to the source documents. We discourage this approach for two main reasons:
- The preparer must waste time hunting for errors, rather than benefitting from the scan and populate interface designed for the verification process where the data is presented to the user.
- Since this process requires knowledge of the tax software it must be performed by the preparer rather than a lower paid administrative staff person or intern.
Preparers end up bearing the entire responsibility for verifying OCR data regardless of how much work they already have on their plate. It’s an inefficient procedure that significantly increases the chance of human error, extends the preparation time , and negatively affects the bottom line.
Next generation scan-and-populate eliminates human verification
While verifying OCR data before exporting to tax software is certainly better than the alternative, neither is ideal. Both processes require firms to bear the hidden costs of compensating for the fact that scan and populate technology isn’t 100% accurate.
1040SCAN’s patented Auto-Verification technology reduces or eliminates the need for time consuming human verification of OCR data pulled from native PDFs. This is the first and only solution on the market that can automatically verify the accuracy of OCR data.
1040SCAN’s next generation scan-and-populate technology cuts down the OCR verification time for 50 pages of source documents from 30 minutes to 11 minutes or less for a firm. The momentum-killing bottleneck of human verification is completely eliminated.
1040SCAN’s Auto-Verification technology is one of the 5 Keys of a Successful Scan and Populate Process.