Capture Is Key To The Digital Mailroom

Businesses have used scanning devices and software packages to capture data from documents and forms for more than a decade. These systems have historically produced proven and rapid returns on investment for departmental scanning or isolated scanning applications. Based on these successes, progressive companies have been searching for a way to use the scanning and software model to digitally capture their entire incoming correspondence streams. This concept of a digital mailroom, while intriguing, was not possible using traditional forms processing methodologies.

"Most forms processing systems of a couple of years ago dealt with a limited number of structured documents, such as claims forms or tax forms, and used templates, patch pages, separator sheets, and ID bar codes to identify document types and extract desired data," says Jeff Toren, sales manager for SWT U.S., Inc. (San Diego). "In a digital mailroom, you're dealing with all types of documents. These documents can include structured forms, letters, and invoices, and can vary greatly in size and print [i.e. machine or hand]. To manually create a template for every document an organization receives is too time-consuming and infeasible."

The inability of traditional forms processing technologies to address the needs of the digital mailroom motivated several vendors to develop new recognition technologies that dynamically identify and process semistructured and unstructured content such as invoices and correspondence. These intelligent document recognition (IDR) technologies now make it possible to identify all document types that come into an organization and have made the digital mailroom concept a reality. A select few companies have actually implemented digital mailroom solutions, and their early successes are creating a strong case for other businesses to follow suit. Interest in the digital mailroom provides a new sales opportunity for resellers, but VARs must understand IDR's role in the digital mailroom and how to sell to the enterprise in order to capitalize on this trend.

Classification Enhancements Enable Document Identification
Recent advancements in document classification technologies have been among the most influential in establishing functional digital mailroom environments. Classification engines can now be configured to identify several different document types based on the customer's application requirements. For example, documents can be classified based on bar code recognition, dynamic identification or image analysis, image text classification, or electronic text classification. Dynamic identification or image analysis identifies a document type by comparing it to a list of predefined templates. This technology works well on structured documents such as questionnaires, surveys, and claims forms. Image text classification scans the content of an image to locate keywords that identify the document type. Electronic text classification identifies keywords in electronic document streams such as e-mail, Microsoft Word, XML, and PDF to classify information.

One way a VAR can immediately add value to a digital mailroom prospect is by helping conduct an in-house document analysis. "During the document analysis, unique differentiators between one document and another should be captured, including keywords, logos, and document formats," says Jim Vickers, chief marketing officer for Captiva Software (San Diego). "Once that information is gathered, you should research which technology best meets the end user's document classification requirements. Because the mailroom receives all types of documents, it is important to work with a vendor that provides a diverse classification set. Also, most classification technologies still have issues with hand-written correspondence and send these documents to a manual classification interface. It is important to have a flexible, user-friendly interface to process these documents."

Collect, Transform, And Deliver Information To Complete Transactions
The digital mailroom offers businesses more than the ability to receive, analyze, manage, and route all of their incoming communications from a single point. It also allows companies to employ information capture closer to core business applications and transform raw data into information needed to complete transactions. In other words, the digital mailroom is championing the idea of using captured information to actually accelerate other business processes, rather than just archiving data away for future retrieval.

In the past, information capture was disconnected from back end systems and other business applications. The digital mailroom provides a means for linking all of these processes together. To do so successfully requires the effective collection, transformation, and delivery of information.

Most businesses are increasingly aware that they live in a hybrid world and they receive vital information from a variety of different mediums including paper mail, e-mail, fax server, and the Internet. These businesses desire to have a single receptacle where they can gather information from all sources. The digital mailroom provides this collection point.

Capturing data by itself is useful for archival purposes, but to drive a business process, that data must be transformed into the critical information the back end system really needs. For example, with an invoice, this critical information includes the invoice number, line items, and total purchase amount. On a mortgage document, this critical information includes property costs, interest rates, and the terms of the mortgage agreement. Classification technologies provide the first step of the transformation process by identifying the document type (e.g. invoice, mortgage document, resume, claim form). Advancements in data extraction allow the second step of the transformation process to occur. These technologies can be configured to search select document types and pull out the information necessary to complete a transaction. Although classification and extraction technologies automate much of the transformation process, they aren't 100% accurate. Therefore, manual workflow systems should be included for validation and quality control purposes.

Once the data has been collected, transformed into useful information, and exceptions have been validated through manual workflow, the information must be delivered to the company's back end system. "The back end systems that use captured data are really different today from what they were 5 or 10 years ago," says Sameer Samat, CTO of Kofax (Irvine, CA). "This information used to reside only on content management systems. While these systems are still important, the need for line of business applications such as ERP [enterprise resource planning] systems has become a priority. Communicating with these systems requires sophisticated export and release capability, as well as Web service functionality."

Domain Expertise Can Lead To Enterprise Business
While new capture technologies have made the digital mailroom a reality and much interest exists in the business community, few companies have actually implemented digital mailroom environments. "Return on investment is still a big hurdle for the digital mailroom," says SWT's Toren. "While an automated mailroom is possible, the cost and time of implementation makes it impractical for many customers."

For these cost justification reasons, the immediate targets for digital mailroom technologies are large and midsize organizations within the financial services and insurance industries, as well as service bureaus that handle document capture for their customers. These businesses have a large influx of documents and can realize payback and efficiency gains on a digital mailroom solution more quickly.

Selling a digital mailroom solution to this demographic requires the ability to uncover and explore the needs of the entire organization, as opposed to a specific department implementation. This can be a change of pace for a VAR. However, VARs with vertical market expertise can use their specialization to their advantage. "Domain knowledge can be more important to selling the digital mailroom than technical expertise or enterprise selling experience," says Kofax's Samat. "VARs with knowledge of the invoicing process or mortgage and loan processing will have a huge advantage. A vendor partner can help a VAR get up to speed on the new technologies, but the reseller can add value by gaining vertical knowledge and applying these technologies in line with that knowledge."