Automating Tax Preparation With OCR Technology

OCR is changing the Tax Preparation Industry. Learn how you can take advantage of this amazing technology.

July 16, 2007
Sponsored by 1040SCAN™

by David Wyle, CPA, CITP, CEO SurePrep, LLC

Optical Character Recognition, usually abbreviated as “OCR,” is a type of software that transforms document images (usually captured by a scanner) into editable computer files. Traditionally this type of software has been used to read structured forms (i.e., forms with identical data types and layouts). Advancements in OCR technology, however, have now made it possible to read semi-structured forms (i.e., forms with identical data types but different layouts) with a high degree of accuracy. This advancement has made the technology applicable to tax preparation as most tax return source documents are semi-structured (e.g., all 1099-INTs contain interest income even though the location of that data may differ from one 1099 to another). This article will discuss how OCR technology can reduce tax preparation time and errors. It will also provide several tips for successfully implementing an OCR solution in your tax practice.

Organizing and entering data from source documents can take more than half of total tax preparation time. OCR technology can reduce this time by more than half resulting in overall savings of 25 percent or more. By recognizing organizer pages, brokerage statements, and hundreds of variations of source documents (e.g., W-2s, 1099s, etc.), OCR software can automatically organize your “shoe box” of client documents into a bookmarked PDF file that follows the flow of the return. It then reads the data from these documents and exports it to your tax software. This results in hours of saved file organization and data entry time. Of course not all documents are recognized by OCR software. Receipts, letters, hand written notes, income statements, etc. must still be entered manually. But OCR can generally capture any legible data provided on organizer pages, brokerage statements, and the great majority of source documents.

As a computer system, OCR can be programmed to perform numerous analyses on tax data and do so with 100 percent consistency and reliability. This enables it to catch errors that frequently are overlooked by human preparers. For instance, an OCR system can catch subtle changes in an address (e.g., an employer’s zip code) that may affect e-filing. It can verify that every document entered into the return has a Social Security number that matches either the taxpayer or spouse. It can determine when a dependent’s income should be reported on the parent’s return or separately. It can quickly identify foreign income transactions on a brokerage statement. And it can ensure the accuracy of numbers by applying business rules to them (e.g., ensuring Box 3 minus Box 4 equals Box 5 on Form 1099-SSA). The result of these automated analyses is improved accuracy, quality, and consistency.

To successfully implement an OCR solution your firm must have proper expectations and adequate scanning hardware/software. Regarding expectations, OCR does not recognize all data and sometimes makes mistakes. If the expectation is that recognition will always be perfect then you will be disappointed. Rather, you should expect that OCR will recognize 60 percent or more of the data that needs to be entered into the return. Any data that was not recognized or not recognized correctly should be highlighted by the system. This will make it easy to find and correct errors and enter the remaining data into the return. This will result in a net time savings of about 25 percent. But to achieve this time savings you must ensure that documents are scanned with proper image quality. Without this, no OCR system will work properly. The best way to ensure your images are optimized for OCR is to use image enhancement software such as Kofax’s VRS 4.1 Professional. This software integrates with most scanners and ensures that all images sent for OCR are right side up, without speckles, and without background shading or watermarks. This can improve OCR accuracy by up to 35percent, which often is decisive in whether an OCR implementation is successful.

If your tax practice is already paperless (i.e., you scan client source documents before you prepare the return), then utilizing OCR entails minimal process change. The documents you already scan will simply be sent for OCR before tax preparation begins. If you are not already paperless, then implementing an OCR system can be your first step in enjoying the wider benefits of a paperless workflow. Start your product research now so you have time to fine-tune your processes and ensure a successful tax season. For more information visit www.1040Scan.com, or call (800) 805-8582 and select option 3.