American Tax Return Processing Company again turns to ABBYY and chooses FineReader Engine – the only OCR product in industry to read data from over 900 document variations

“The speed, accuracy and consistency with which ABBYY’s software can recognize and extract data from taxpayer source documents can’t be matched by thousands of human operators”.

David Wyle, CEO, SurePrep

Looking back to 2004 when David Wyle, CEO of SurePrep, purchased his first two ABBYY products, FlexiCapture and FormReader, he knows he did the right thing. Not only did the solution pay for itself in 2 years, SurePrep now has the most advanced knowledge base in US tax form processing. And that is not all...

Background
SurePrep is an American company which specializes in processing tax returns, with clients from top 100 accounting firms. SurePrep's tasks are to analyze, organize and enter data from taxpayer source documents into tax software that generates the forms that are filed with the IRS (Internal Revenue Service). Brokerage statements, W-2s, 1099s, K-1s are just some examples of the hundreds of source documents submitted by a tax payer. Each form type has its own, often complex, structure, and data fields are arranged differently on each type.

Going through these document batches and classifying them isn't a walk in the park for any human data entry operator. And, it gets more complicated: all U.S. individual income tax returns are due on April 15th, whereas taxpayers generally do not receive their source documents until early February. That is why tax preparers have a very compressed workload from around February 15th to April 15th handling hundreds or even thousands pages per day. For companies such are SurePrep, it is imperative that they use all efficiencies possible to maximize the number of returns they can process.

SurePrep has a BPO, which handles data processing, located in India. In 2004 SurePrep phased out manual entry and invested in a data capture solution from ABBYY: FormReader Enterprise Edition and FlexiCapture, through ABBYY's local partner, NETSPIDER INDIA (NIIL). A detailed story about this part of the project (completed in 2005) is published on ABBYY's website.

David Wyle has always had an eye for the most advanced, reliable and potent technologies. He invested in ABBYY's data capture system that can classify hundreds of document types, and he requested training straight from ABBYY, which gave his company the kind of knowledge base that has no analogies in the world.

“ABBYY sent their tech training specialists over to our BPO in India about 7 times in 3 years, and that was great. The depth and quality of knowledge in data capture we received is unsurpassed. ABBYY's tremendous technological power plus their and NIIL's support of our project is invaluable,”comments Mr. Wyle, “The power of our data capture system and our unique knowledge puts SurePrep in a very advantageous position.“

SurePrep's initial investment in data capture software paid off in 2 years. David Wyle comments, “Not only did the solution pay for itself so soon, but it also improved our ability to consistently deliver top quality services, and, needless to say, now we can get much more work done. The speed, accuracy and consistency with which ABBYY’s software can recognize and extract data from taxpayer source documents can’t be matched by thousands of human operators.“ Mr. Wyle's vision and his virtuoso execution of the project gave SurePrep a huge competitive advantage as the company now has the most advanced OCR platform for automating U.S. tax return preparation.

New Project: SurePrep Gears up with ABBYY FineReader Engine OCR
Any highly-successful business has to keep abreast of the latest tech developments. SurePrep is no exception. In connection with the increasing document flow there was a necessity to customize its system for feature requirements, such as performing verification and correction in SurePrep's proprietary application. To solve this task, the company again turned to ABBYY, and this time, it procured FineReader Engine, a comprehensive SDK for data capture and document conversion. In addition to OCR, ICR, OMR and barcode recognition, FineReader Engine also offers image pre- and post­processing, document layout analysis, PDF conversion, and forms processing (for both fixed and semi-structured forms). This solution has been a favorite among developers and integrators because it offers all of the essential data capture and document conversion tools in a single toolkit. As a result, developers and integrators benefit from working with a single and efficient environment, and modular approach in its run-time licenses ensures that developers only pay for the technology they use in the end application. Also, FineReader Engine is flexible enough to support development of an application of any architecture and scale.

ABBYY Technologies in SurePrep's Workflow

1. Collection of documents. All source documents necessary for tax report are collected. These documents show a range of income and expenses and appear on various forms. Most of these forms come as hardcopies.
2. Classification of document type. Sets of source documents pass through SurePrep’s solution, called 1040Scan™, where ABBYY FineReader Engine classifies it to fit to certain document template and performs OCR.
3. Data capture. It involves data capture solution called ABBYY FlexiCapture. Every form image received from OCR runs through defined “ID” template. Document structure is analyzed by ABBYY technology. It uses over 400 specific Flexi templates and has the capacity to process hundreds of document types with potentially millions of variations within a type. Additionally, irrespective of client scanning variations the system locates all data fields and finds their exact coordinates on the page. Thus, this software-hardware complex enabled SurePrep to process thousands of semi-structured documents in a manner that is the least labor-intensive yet the most accurate and fast.
4. Validation. In an event when image recognition is questionable or when certain business rules are not met, a subset of the extracted data is sent for verification and correction in SurePrep's proprietary application.
5. Export. Verified data is exported to the tax software which generates a PDF that has the OCR'd pages organized and bookmarked to follow the flow of the tax return contents.

Main advantages of using FineReader Engine

1. Per David Wyle, the highlights of the new system are “greater flexibility and customizability.” It became possible to determine what data to verify and who should verify it regardless of location. Verification is done in an industry-specific context that makes sense to the end user.
2. More reliability. The system is efficient enough not to require 24/7 attendants during peak periods.
3. Better performance. More pages are processed in less time. Mr. Wyle comments: “FineReader Engine is the only product in the industry to read data from over 900 document variations and export it to the tax return. This generally eliminates 50% or more of the file organization and data entry that would otherwise need to be done manually.

Notably, SurePrep was the first, and, at the time of preparing this material, is the only company to successfully automate this process using ABBYY FineReader Engine.

Summary

With the new system up and running, SurePrep was able to accomplish its goals:

1. Maximize tax return volume per preparer by eliminating need to manually enter 50%+ of the source document data
2. Provide a technology solution that allows firms to reduce tax preparation data entry without having to outsource

Some elaboration on point #2 and its significance: When SurePrep started, they only provided tax outsourcing services. The infusion of ABBYY's OCR not only automated tax outsourcing operation, but also allowed the company to offer a service to other U.S. firms which did not want to outsource. Because of this service, those firms can now do their own verification in SurePrep's proprietary verification application. The ability to offer this option essentially provided SurePrep with a key to unlock the door to a market segment which they previously could not reach. Hence, ABBYY made SurePrep's existing operations more efficient, and it provided a platform to develop a new, leading edge solution that expanded product offering, target market, and customer base.

"One other thing that deserves a special mention is how devoted and accommodating ABBYY has been to our needs," says David Wyle in conclusion, "They did custom development work for FormReader Enterprise Edition in 2006, and then, in 2007 they made sure that FineReader Engine and FlexiCapture results synced up. It was very important for us to have these changes, and ABBYY did them within the time frame that was essential to our business. Having a software company like ABBYY which was willing to work with us towards what we needed to accomplish meant a world of difference."

About SurePrep the leading provider of tax productivity solutions, is a U.S.-based company that focuses on providing technologies and services that help make accounting firms more profitable. The company offers U.S. accounting firms a web-based file cabinet where tax return data is stored on servers at the AT&T data center in Irvine, California. SurePrep offers numerous advantages including increased availability of qualified personnel, faster turnaround, secure online access and lower cost. For more information, please visit http://www.sureprep.com/
	About NetSpider India (NIIL) NETSPIDER INDIA specializes in ABBYY channel distribution and giving end-to-end solutions in various fields like imaging, networking, scanning, connectivity, Project consultancy, and doing developments around ABBYY technology. NETSPIDER INDIA has experienced technical skills in implementing ABBYY FineReader OCR, FormReader Enterprise, Desktop, FlexiCapture Studio and other ABBYY products. More details are available at http://www.niil.in/