White Paper

ABBYY FormReader And FlexiCapture Studio Save Your Time, Nerves And Money

Click Here To Download:
•Product Brochure: ABBYY FormReader 6.5 Desktop Edition
•Product Brochure: ABBYY FormReader 6.5 Enterprise Edition
•Product Brochure: ABBYY FormReader 6.5 Developer Edition
•Product Brochure: ABBYY FlexiCapture Studio 1.5

SurePrep is an American company which specializes in processing tax returns. Most of its clients are accounting firms. SurePrep's job is to extract data from clients' source documents to tax returns and to make statistical and other calculations.

In the USA there is no general standard regulating the structure and appearance of such documents. Therefore, a batch of documents may include forms which are absolutely different in their structure, with the same data fields arranged differently on each form or located on different pages (such forms are usually referred to as "semi-structured forms"). SurePrep's back processing office (BPO) is located in India, and all forms to be processed are securely accessed from the Company's US servers in PDF format. Initially, a group of operators was organized whose task was to manually extract information from received files. But soon this practice had to be abolished because neither labor productivity nor the quality of data extracted were satisfactory: people will make mistakes, and at the end of the working day, when the operators get tired, the number of errors skyrockets.

In September 2004 SurePrep contacted NETSPIDER INDIA, ABBYY's Indian partner, for consulting on creation of a data capture system. After thorough consultations with ABBYY and NETSPIDER INDIA specialists, SurePrep bought the ABBYY FormReader Enterprise Edition and ABBYY FlexiCapture Studio systems. On the basis of these two systems, a software-hardware complex was built. Its hardware part consisted of the Data and License Server, and the Administration, Recognition, Verification, Correction, and Export and Monitoring Stations. There was no need to set up a Scanning Station or install a scanner as all the forms to be processed were already in electronic form. A typical workstation was Pentium IV 2.5 GHz, 512 MB RAM.

After the first phase of the project, NETSPIDER INDIA organized training and consulting courses involving the presence of ABBYY consultants. That was the step where ABBYY's involvement in the project was maximal. SurePrep specialists completed the course on FormReader Enterprise Edition and FlexiCapture Studio 1.0 and within the two months following the completion of the course created, tested and set up flexible templates. After some consultations with ABBYY and additional training, the flexible templates were further modified to ensure the best processing quality. The subsequent technical support and consulting services were provided mostly by NETSPIDER INDIA itself.

The use of the ABBYY data capture systems has allowed SurePrep to cope with several important tasks at once. First of all, the capabilities of the software-hardware complex allow for a quick classification of forms: FormReader uses user-defined properties to detect the type of each document (which may contain up to 50 pages) in the batch, and sends the document to the corresponding workstation.

Second, FlexiCapture technology enables the system to analyze document structure. The system locates all data fields — it can even find their exact coordinates on the page. Thus FormReader and FlexiCapture Studio enabled SurePrep to process semi-structured documents.

Third, once all necessary data are captured and converted to digital format, they undergo multi-step verification, and are then automatically routed to SurePrep's tax return processing system. SurePrep's specialists have created several programs for review managers which allow them to easily check the quality of captured data. Since ABBYY technology can establish the coordinates of data fields, it was possible to create applications for visual verification of processed data through highlighting recognized fields on original images.

Integration of ABBYY FormReader Enterprise Edition into SurePrep's complex was performed in a rather simple fashion. The FormReader complex was organized as an isolated system in SurePrep's dataflow which took images from a dedicated folder automatically. After data capture processing, FormReader provided resulting XML file with all the necessary information which was automatically grabbed by SurePrep's software. This kind of integration scheme is also widely used in indexing and classification systems. Such approach made administration easy and allowed for deep integration of the ABBYY technologies without additional programming efforts.

The software-hardware complex built on ABBYY's technologies not only solved all SurePrep's tasks and raised the quality of captured data, but significantly reduced SurePrep's data capture expenses.