Electronic Document Management:
Challenges and Opportunities for Information Systems Managers

By:Ralph H. Sprague, Jr.
University of Hawaii
College of Business Administration
2404 Maile Way
Honolulu, HI 96822
U.S.A
sprague@hawaii.edu

ABSTRACT

Harnessing information technology to manage documents is one of the most important challenges facing I/S managers in this decade. It is important because most of the valuable information in organizations is in the form of documents such as business forms, reports, letters, memos, policy statements, contracts, agreements, etc. Moreover, most of the important business processes in organizations are based on, or driven by, document flows. Electronic Document Management (EDM) promises major productivity and performance increases by applying new technology to documents and document processing.

The purposes of this paper are to show the value of new technology for managing documents, to illustrate the variety of ways this value can be realized, to develop some structure for understanding this rapidly evolving field, and to suggest some actions I/S managers can take now to prepare for this revolution in information management. The paper argues that the I/S Department, as the developers and managers of the technical infrastructure for EDM, will be in a position to lead this evolution as major change agents as they did in the EDP and MIS eras, but some specific actions will be needed to assume this leadership role.

The first section of the paper explores the scope and importance of EDM in more detail, and illustrates how it expands our view of information management. The next three sections are designed to help structure the field by approaching it from three perspectives: technologies that are making EDM possible, the application areas in which business value are being realized, and the roles and responsibilities of several departments that will be involved in developing EDM. A final section suggests what I/S managers can do now to begin preparing for this major advancement in information management.

KEYWORDS: Document Management, Document Processing, I/S Management

ISRL Categories: CA14, DA06, DA07, DB05, EF02, HA6-12, HA14.

INTRODUCTION

For the most part, Computer systems have handled facts organized into data records. Far more valuable and important to organizations are the concepts and ideas contained in documents. Reports drawn from the computerized database fill important roles in status assessment and control, but frequently they must be accompanied by a memo or text report which explains and interprets the computer report. Indeed, in one study, CEO's rated computer reports least valuable for decision making from among a set of communication mechanisms. Meetings, phone conversations, news items, written memos, and noncomputerized reports rated much higher. (McLeod and Jones 1987) Technology applied to the handling of documents promises to improve these important forms of communication.

Until recently, however, technology for document processing has been mostly limited to better and faster ways to generate, print, and transport text documents. Now several trends and developments suggest that we are on the verge of a major advance in computer based information management. Electronic Document Management (EDM) is the application of new technology to save paper, speed up communications, and increase the productivity of business processes. From a broader perspective, EDM is a major expansion in the domain of information management, and a concomitant increase in the responsibilities of I/S managers and executives.

A recent Gartner Group Strategic Analysis Report forecasts that by 1995 document management functions will become the most important service on Local Area Networks after basic connectivity. (Popkin and Cushman 1993) The report argues that a strong business case can be made for investing in document management systems and the organizational leverage they provide. Ninety-eight percent of computer users employ word processing on their computers. Eighty to ninety percent of organizational information is in documents rather than structured databases. Dominant as they are, these figures will increase.

More than a decade ago Swanson and Culnan recognized the importance of document-based computer systems for management planning and control. (Swanson and Culnan 1978) They cited early references to its importance, but noted that "the role of document-based information systems in management has been relatively neglected." As a result of their study, they suggested that "there is some reason to believe that document-based systems may at last find their way into contemporary MIS thinking".

Unfortunately, their view proved to be optimistic. Although there has been continual progress in applying technology to documents, the majority of computer accessible information is still in data records, and the majority of document information is still on paper. But now. a strong combination of business forces and major technology developments may finally make it possible to apply technology to documents in a productive way.

Advances in handling information in documents are being driven by several forces in the business environment, and enabled by new technology developments. The business forces include the drive for increased quality on which to base global competitiveness, and the need for increased productivity to conserve scarce resources. Technology developments enabling these advances include digital image processing, large capacity storage, hypertext, multi-media documents, high bandwidth communication channels, electronic printing, electronic mail and fax, and improved techniques for information and text retrieval. These technologies are proving valuable for applications such as workflow management, communication between and within organizations, training and education, records management, and internal reporting.

The objectives of this paper are three fold. First is to establish a vision of the scope and potential impact of EDM in organizations, and the implications of these developments for I/S managers and executives. A second objective is to develop an understanding of the technologies that are enabling new ways of handling and processing documents and how these new technologies will be integrated into an expanded I/S infrastructure. These rapidly evolving technologies make possible functions that were previously infeasible, so they will force us to think about information and documents differently. A third objective is to provide some structure to guide the journey toward the EDM vision.

The first section of the paper explores the scope and importance of EDM in more detail, and illustrates how it expands our view of information management. The next three sections are designed to help structure this field by approaching it from three perspectives: A final section suggests what I/S managers can do now to begin preparing for this major advancement in information management.

SCOPE AND IMPACT OF EDM

Definition and Scope

A document can be described as a unit of "recorded information structured for human consumption." (Levien 1989) It is recorded and stored, so a speech or conversation for which no transcript is prepared is not a document. This definition accommodates "documents" dating back to cuneform inscriptions on clay tablets; what has changed lately are the ways the information is represented, and the ways the documents are processed. Information previously represented primarily by text is now also represented by graphical symbols, images, photographs, audio, video, and animation. Documents previously created and stored on paper are now digitally created, stored, transported and displayed.

This definition also accommodates a wide variety of documents used in organizations. Examples include: The application of technology for processing even the more traditional documents in this list is making a major change in what documents are and can accomplish in organizations. A definition more oriented to technology comes from BYTE magazine. (Michalski 1991): "A document is a snapshot of some set of information that can Another perspective suggests that a document is the "unit record" of conceptual information. A data record contains the attributes of an entity such as an employee in a personnel system or a part number in an inventory system; a document contains the information necessary to represent a concept or idea. Although most documents currently contain a cluster or set of these "concept nodes," future documents may be composed of a network or web of linked conceptual unit records. These "chunks" or "bundles" of information will have attributes that make them more useful and human than traditional data records. Context, tone, richness of representation media, and flexibility of structure will make the information in documents more consumable and accessible to humans. This perspective strengthens the understanding that document management is an expanded form of information management.

In spite of these broad definitions, the dominant connotation of a document is relatively structured and formal information, primarily text, printed on paper. Therefore the scope of Electronic Document Management must encompass the use of technology to handle paper documents or their electronic equivalent. Older technologies for document handling include micrographics, computer output microfilm (COM), and automated records center applications. A newer technology is digital image processing which represents a page of a paper document with a digital image of that page.

Increasingly, however, EDM will emphasize electronic documents and their management. An electronic document uses a variety of symbols and media to represent a set of ideas and concepts. In addition to traditional letters and numbers (text), an electronic document may contain graphical symbols, photographs and other images, voice, video clips, and animation. This clustered set of symbols can be stored, retrieved, and presented electronically as a "compound document". For example, an internal report on a product improvement may present, on a computer screen, the text explaining the feature, a photograph, an engineering diagram, a voice notation from the product designer, and a video clip of the product in use. Figure 1 shows the conceptual structure of a such a compound document. This is a richer, enhanced definition of what we have traditionally called a document. EDM requires us to expand our connotation of this old and comfortable word.


Figure 1. Structure of a Compound Document

The use of the term EDM in this paper encompasses both of the interpretations given above. Thus, the scope and definition of EDM can be defined as follows.

Business Value from Documents

Document management can generate business value in two ways. First, for some industries such as publishing companies, documents are a direct source of revenue as a product, or as support for a product. Second, for all organizations in all industries, electronic document management can improve the information management tasks that are needed to manage, control, and operate the organization.

In the first category, publishing companies produce books, magazines, newspapers, and other formal documents as products. Document processing facilities, procedures, and technologies are their "factory". Less obvious publishers are lawyers, insurance companies, consulting firms, and many government agencies for which the primary manifestation of their product or service is a document. Here again, electronic document management will improve the quality of their product or service and reduce the costs of producing it in the same way a modern factory does for a manufacturing firm. For some industries, documents are not the primary product, but they generate revenue by supporting the product. Examples include the owners manual for an automobile or major appliance, the reference manual for a software product (often in electronic rather than paper form), and a pharmaceutical firm's application to the government for new drug.

In the second category, the value of document management to support organizational performance can be grouped in three general categories -- as a mechanism for organizational communication (especially for concepts and ideas), as a vehicle for business process, and as a major component of organizational memory.
  1. Improved management and communication of concepts and ideas. A major value of EDM derives from its ability to expand the scope of information management from facts in the form of data records and databases, to concepts and ideas which are generally captured, stored, and communicated in the form of documents. EDM technology can thus improve the efficiency and effectiveness of documents in their role as a primary mechanism for storing and communicating concepts and ideas within and between organizations (and their groups and individuals). This set of benefits directly supports the expected structure of future organizations which are likely to be flatter, heavily based on teams, geographically distributed, and more dependent on their ability to handle rich and varied information.
  2. Upgraded, "reengineered" basic business processes. Most organizations have a substantial set of "paperwork" systems that have been resistant to computerization, at least partly because they are based on documents rather than (or in addition to) data records. Evolving EDM technologies will support more of these applications. Real benefit will result, not just from automating these processes, but from rethinking or reengineering them to take advantage of the advanced technology. (Kind and Eppendahl 1992) Several of these efforts suggest that the most effective reengineering approaches are document based or document driven. (McDonnell and Somerville 1991) In a recent project to reengineer one of six major business processes, Xerox identified 263 subprocesses with 2300 links connecting them A full 2070 of the links, or 92%, were documents.
  3. Leveraging Organizational Memory. Documents form a important part of organizational memory, and EDM can improve the ability of the organization to utilize it. In the short run, major value derives from merely storing paper documents in electronic form. In the long run, EDM will provide the means to access and analyze organizational memory to improve productivity and performance.
These general areas of use and value of documents illustrate why documents are so pervasive and fundamental to an organization. Figure 2 shows some additional roles and purposes for documents with an example of each.

In summary, the overall potential impact of applying technology to document management is significant. Because documents contain concepts and ideas, EDM promises to advance the management of conceptual information in organizations. Because most of the activities of information workers at managerial and professional levels deal with concepts and ideas, EDM promises improved support and productivity improvement at these levels. Because documents are a major vehicle for exchanging information in business processes, EDM promises to make a major contribution to process redesign and improvement efforts. Finally, with documents forming a major part of organizational memory, EDM will support enhanced utilization of that resource.

ROLES

EXAMPLES

To record or to "document" contracts and agreements employment contracts, maintenance agreements, consulting contracts, purchase agreements, leases, mortgages, loans,etc.
To record policies, standards, and procedures procedure manuals, standards specifications, instruction handbooks, executive memos and letters that state corporate policy, etc.
To represent a view of reality at a point in time (reports and plans) status reports, problem analyses, operational, reports, staff recomendations, budgets, strategic plans, etc.
To create an image or impression annual reports, marketing brochures, TV or radio commercials, etc.
To generate revenue as a product a book for sale by a publisher, a report by a consulting firm to be sold to its client, a news item from a wire service, a reference from a bibliographic service, etc.
To support revenue by adding value to a product a user's manual for a car or appliance or a software product, a warrantee form, a catalog, a discount coupon for the next purchase, etc.
To act as a mechanism for communication and interacton among people and groups memos, letters, presentations, email, messages, minutes of meetings, etc.
To act as a vehicle for organizational process orders, invoices, approval letters, most business forms, etc.
To provide a discipline for capture and articulation of concepts and ideas nearly all the kinds af documents that carry concept and ideas.

Table 1. Roles Documents Play

Driving Forces

In the short run, EDM will be useful in dealing with the "paper problem." Many organizations are literally drowning in paper. A study by records managers estimates that there are 318 billion paper documents on file, with 92 billion new pages added each year. Computers print 775 billion pages each year. In 1990, these new documents required 3.1 million tons of paper. (Allerding 1992) Business forms are a large portion of this paperwork. A research firm estimates that U.S. companies spend more than $6 billion every year on preprinted forms, and throw away more than $2 billion, and they spend from $94 to $120 billion per year to distribute, store, and process forms. (Skapinker 1991)

Although EDM technologies have already begun to have an impact on this problem, it is unlikely that merely replacing paper with an electronic equivalent will yield lasting productivity improvement. Major productivity improvements will require radical organizational change as we abandon decades-old processes designed around paper documents. There is now widespread agreement that traditional work processes and practices must be redesigned or "reengineered", taking advantages of new technological capabilities, in order to reap true productivity increases. (Hammer 1990) (Davenport 1993) (Gleckman 1993)

However, the "paperless organization" is not a realistic idea in the foreseeable future. Paper has too many advantages and technology too many limitations to expect complete elimination of paper any time soon. It is likely, however, that the primary role of paper will change. Rather than serving mainly as the storage medium for documents, paper will act primarily as an interface medium. The official "original" document will exist in electronic form, to be printed when and where it is needed. The paper "copy" is used and discarded (recycled), and printed again later if needed.

In the long run, continuous productivity improvement leads to an enterprise view of reengineering. (Hammer 1993) The ability to use technology to manage all organizational information, in documents as well as data records, will support or enable the changes in organizational structure that are forecast by organizational theorists. Galbraith suggested that organizations are literally information processing systems. (Galbraith 1979) Viewing organizations as sets of information processing units which need to interact and communicate, illustrates the importance of the communication media.

While Galbraith stressed the role of information processing in reducing uncertainty, later work added another major function -- to reduce ambiguity and equivocality. (Daft and Lengel 1986) This work led to a consideration of additional important attributes of information and communication media within organizations. (Daft and others 1987) EDM applies technology to enable traditional mechanisms such as documents, to employ richer media, a richer structure, and therefore to support organizational communication and information processing more effectively. (Meier and Sprague 1993)

In summary, the primary driving forces for EDM are the business need for productivity improvement in the short run, and improved organizational effectiveness in the long run. But this organizational revolution will require a systems revolution driven by a paradigm shift in information management and the use of technology to implement it.

Opportunities and Challenges

Managing this revolution will be one of the most important challenges facing I/S executives in the rest of this decade. It will be important and challenging for several reasons. The "bottom line" is that organizations need improved productivity and enhanced performance. EDM promises to leverage new technologies to generate these productivity and performance improvements, but realizing these benefits will present major challenges and opportunities, especially for I/S managers.

EDM From Three Perspectives

The above discussion suggests that the topic of EDM is huge, pervasive, important, and technology-intensive, with significant potential value to organizations. It would be valuable to have some structure to begin working toward this potential. There are three perspectives which can be used to organize this effort. The next three sections of the paper explore these perspectives. Technology is first because it is proving to be the facilitating force. The second section explores a variety of application areas for which EDM is valuable, beginning with a case example that illustrates the variety of departments in which applications might evolve. The third section examines the roles and responsibilities of the several departments that must work together on a coordinated EDM effort.

TECHNOLOGIES FOR DOCUMENT MANAGEMENT

Underlying Infrastructure

The rapid developments in EDM are partly the result of advances in basic technology infrastructure. These underlying, enabling technologies improve the handling of information in any form, but several have attributes that support document processing and management. These enabling technologies can be organized in five major categories.

Stronger desktop workstations.

Powerful desktop computers based on RISC technology are equipped with large, high resolution color screens. These workstations permit the display of documents, a full page or two at a time, delivering (and capturing) non-text media such as voice, video, and animation. (Herr and Rosebush 1991)

Storage media.

High capacity storage media hold the large volume of bits required for multi-media documents. The capacity of magnetic media (hard disks and diskettes) in workstations and file servers has been increasing rapidly but is still barely adequate. Optical storage media such as CD/ROMs and laserdisks, perhaps in clusters called jukeboxes, provide orders of magnitude increases in storage capacity. (Harvey 1990) Holographic storage devices increase the amount of readily available storage capacity by several more orders of magnitude (Baran 1991)(Yam 1993)

Networks.

Networks will interconnect the workstations of most, if not all information workers, within and between organizations. These connections have increasingly high bandwidth to transmit the large volume of data contained in electronic documents and forms. Relevant technologies include FDDI, broadband ISDN, and Asynchronous Transfer Mode (ATM). (Sproull and Kiesler 1991)

User friendly software.

The continued growth of graphic user interfaces (GUI), is enabling the multitude of people who handle paper documents, many of whom are not yet computer literate, to deal more easily with documents on computers. (Seymour 1989) Even for experienced computer users, however, interface software must continue to advance so users can move beyond managing hundreds of files, to managing thousands of documents on the desktop workstation. Two examples are the Information Visualization project at Xerox PARC, (Card and others 1991) (Clarkson 1992) and the Piles Interface at Apple. (Mander and others 1992)

Operating Systems.

Client/Server operating systems and network management systems are increasingly document oriented. In fact, new operating systems shift focus from the application to the document. They are also Object Oriented. This approach or paradigm is gaining popularity for improved software design, and for the design of operating systems. (Rymer 1989) It is also the approach used by most of the work on compound documents. (Stewart 1992)

Document Management Technologies

In addition to the underlying technology infrastructure, there is a set of technologies aimed directly at handling documents. Often called document "middleware", these technologies provide the functionality for the processing and management of documents, both electronic and paper. There are actually two sublayers in document middleware: functions for document processing, and functions for document management. Summarized below is the set of document processing technologies, organized by the major steps in a document life cycle, and the document management functions which together form the document technology infrastructure.

Capture and Creation.

These are basically technologies to digitize information. For documents already on paper, hardware and software digitizes an image of a page, and then electronically handles that image. Scanners capture the image while algorithms convert it to digital form, frequently with compression to save storage space. (Daniels 1993) (Wallace 1992) (Datapro 1991) After a document page is scanned and digitized, it can be further analyzed to recognize the characters. Current software can capture full text in editable form in a variety of fonts, sizes, and formats. Extensions of these pattern recognition techniques can recognize voice, some images, and patterns in graphics, animation, and video. (O'Gorman and Rangachar 1992)

Other technologies for creation of documents include a wide variety of word processing and graphic software, joint (and group) authoring tools, version control and access control software. Digital cameras, audio capture boards, and computer graphics systems which produce animation are used for digitizing non-text information.

Storage and Organization.

Several technologies determine how documents are stored and organized. The primary developments are the compound document architecture, distributed storage management software, the integration of documents and databases, and hypertext.

Compound Document Architecture. Such an architecture is required for the different objects that make up a compound document to be handled together. In several implementations, the compound document consists of objects (e.g. a text object, a graphics object, a spreadsheet object, a digital photograph object) which may be stored on different devices, brought together logically through the use of pointers. Several vendors are working on different definitions of a compound document, although the Compound Document Architecture (CDA) from Digital Equipment claims over 200 compliant applications from 50 vendors. (Williams 1993) (Travis 1990)

Distributed Storage. Documents are stored on local PC hard drives, servers (including large capacity document servers), mainframes, and large repositories. A recent survey by the Gartner Group found that 80% of the documents stored in a PC networked environment are stored on the local hard drives, not on the server. (Popkin and Cushman 1993) This underscores the importance of distributed document management software to provide organization and access to this valuable resource.

Integrating Documents and Databases. Making documents an integral part of the information resources of an organization requires integration of document collections and databases. So far, most approaches have been to extend the database to accommodate documents, or vice versa. One approach is to define a BLOB (binary large object) as part of a tuple in a relational data base. (Shetler 1990) A column is defined as a large binary object which can contain a document image or compound document in bit form. A document-centered approach is to reference a data record or entity in the document. This cross reference is used by the application to link the document with a data record. These early approaches are helpful in the short run, but eventually we will need an approach which integrates the data and document resources by their content, instead of just linking documents and data records. (Gilbane June 1993) Developing these approaches promise to present major challenges to researchers and practitioners. Organizing, cataloging, and retrieving concepts in documents is likely to require an entirely different architectural approach than those which have been used for facts in data records.

Hypertext. Software which implements a hypertext structure enables non-linear access to the logical structure of text within a document, and multiple cross references between documents. Hypermedia technology provides the same functionality with multi-media or "compound" documents. Hypercard by Apple, and Notecards by Xerox are examples of software to support hyperstructure. (Conklin 1987) (Halasz 1988) (Bieber and Kimbrough 1992)

Retrieval and Synthesis.

Information retrieval selects documents from a collection according to the presence or absence of keywords assigned by an indexer. Text retrieval uses algorithms that eliminate the need for an assigned index. All content bearing words are indexed. (Lundeen 1992) A further enhancement called concept retrieval, uses thesauri and word co-occurence analysis to select documents that use similar, but different words to represent a concept. (Chen and others 1993) Queries can result in a list of selected documents ranked in order of likely relevance. An extension of this approach allows automatic synthesis or summarization of documents.

Transmission and Routing.

Email systems are moving beyond simple text messaging to become the primary transport mechanism for electronic documents and forms. (Butler-Cox Foundation 1991) Object independence allows transmission of compound documents consisting of a variety of objects (text, graphic, image, audio, video). Other functionality required for business transport of electronic documents includes:
  1. authorization - assuring that the correct user is accessing the workstation and documents
  2. authentication - assuring that the "digital signature" of the user is valid
  3. encryption - coding and decoding documents for security
  4. filtering - automatically routing messages or documents according to their content
Other relevant technologies for routing include workflow management software, access control mechanisms, and intelligent documents. Documents which are "intelligent" or "smart" contain mechanisms to sense who should receive them and in what form. The document literally sends and displays itself to accomplish its purpose. "Active documents" by Interleaf is an example. (Weinberger 1991) (Michalski 1991) (Beal 1991)

Print and Display.

Most documents will be printed at some time in their life cycle, so an important technology is the wide variety of digital printers and copiers on the network. These printers, along with text handling software, page layout languages, and WYSIWYG displays (what you see is what you get) put high quality printed output within reach of nearly everyone. Laser printers significantly reduce the need for preprinted forms. Desktop printers permit distributed printing of richly formatted documents. Xerox's new production publisher operates on a network, accepts Postscript files, allows printed tab inserts, and offers a variety of covers and binding. The result is a new form of distributed printing and "print-on-demand" services that can print small or large runs of complex documents at remote sites under the direction of a workstation. Color copiers and printers also lead to an increased use of color in printed documents. More than just an increase in attractiveness, color is providing a major increase in the communication power of documents. (Vienzenu 1988) (Shank and LaTarce 1990)

This network of printers requires a set of software to manage the distributed printing. Print services on the network will include, for example, usage and activity reports, downtime and repair reports (perhaps with automatic diagnostics and repair calls, checks to be sure documents are being printed on the most appropriate printer, automated job tickets, and an accounting log. For documents that may not need to be printed, electronic display/delivery takes the place of printing, but the lack of a truly universal standard (other than ASCII) is slowing the growth of paperless publishing. (Gilbane September 1993)

Document Management Functions.

The second sublayer of the document technology infrastructure consists of document management functions that cut across the phases of document processing. This set of functions is what enables managing documents as an information resource rather than as a collection of files. These document management functions include:
  1. Status Reporting - Who has a document? What is its recent activity?
  2. Access Control - Who "owns" it? Who can read it? Change it?
  3. Version Control - What is the current version? What previous versions are still needed?
  4. Retention Management - What are the legal retention requirements? Corporate policy requirements? How do we destroy paper and electronic versions?
  5. Disaster Recovery - How and where are backup copies kept? What are the recovery procedures?
This technology section can be summarized with the conceptual layered architectural diagram shown in Figure 3. The lower level is the basic infrastructure and the middle layer (with two sub-layers) is the document infrastructure that is required for EDM. The top layer is the application layer, which shows the four main areas of business value described earlier. A more detailed set of application areas is described in the next section.

Figure 2. Technology Architecture for EDM

APPLICATION AREAS

The second perspective which helps organize EDM is the applications which generate value to the organization. The word "application" is used here in the broad sense of benefit, impact, usage, and value, rather than a specific program or system.

Case Example

As technology and organizational processes evolve, EDM applications will be developed in several areas, and for several purposes. To illustrate the areas in which EDM can be applied, consider the case of a medium-sized manufacturing firm that discovered several EDM applications evolving in separate areas. These application areas, and the departments in which they evolved, are summarized in Figure 4.

This case illustrates that the EDM approach and technology are turning up in several application areas. Generally, the departments which install them are not aware of the developments in the other areas. These "first generation" EDM applications generate business value by improving customer service, revising business processes, speeding the distribution of documents, reducing storage costs or improving access to documents. They are different enough in structure, purpose, and users, that they are separately developed, but they use similar technologies and approaches. Imaging, for example, is a technology used in several of the applications. A document server with multi-media storage and a strong search engine is needed for several. And the concept of "just in time" (printing, learning, forms processing) pervades the design philosophy in all areas. Without some planning in the development of these applications and their extensions, however, incompatibilities will limit the effectiveness of the applications in all areas.

Department

Application

Records Management An imaging system for archival storage and access to legal and tax documents. Replaced an aging microfilm system. Implemented on a network to elimate physical shipment of paper documents among several offices in different cities.
Manufacturing An extended version of a CAD/CAM system to use imaging to manage the blueprints and engineering diagrams.
Human Resource Management An imaging system to support the hiring process. Candidate's resumes are scanned into the the system when they apply, then circulated in image form among the many people involved in the hiring process.
Systems and Procedures A plan to improve the process of printing and distributing the procedure manuals to secretaries and administrative assistants.

Currently- manuals printed centrally and mailed to all users; revised yearly with interim modification sheets.

Phase I - Print manuals over the network on high speed remote printers at each major site (distributed printing).

Phase II - Allow secretaries and administrative assistants to print sections of the manual on their local printer as needed(print on demand).

Phase III - Add retreival and reference capability so users can access relevant parts of the manual online as needed.
Customer Services A new system for publishing and distributing owner's manuals, repair manuals, product descriptions, and products specifications. In the past these have always been printed on paper and mailed to customers, distributors, and sales personnel. Recently they began distributing them on CD/ROM.
Administrative Services Development of a work flow system utilizing electronic forms for such tasks as office supply orders from stores, check requisitions, internal office equipment orders, telephone change requests, etc. A new version of the system will include some features such as authorization,encryption, and signature verification that will permit the use of electronic forms for larger and more important processes also.
Training and Education

A plan to evolve the process of teaching administrative assitants and secretaries.

Currently a classroom course, based heavily on the procedures manual, which uses multi-media presentation materials to explain the steps in these procedures and show the forms that must be used.

Phase I - Convert the multi-media course to a Computer-Based Training course for use on a work station instead of in the classroom.

Phase II - Structure the software so each procedure module can be accessed as needed rather than as part of an entire course.

Phase III - Use real forms instead of just sample forms as part of the course material. These forms can be filled in on the work station and sent over email so that the system becomes a real workflow system that actually performs the tasks. Access to reference material and training/education are additional built in features.

Table 2. Case Example

Document Management Application Areas

A taxonomy of application areas will assist the planning for integration of applications over time. This section describes some of the application areas that are particularly susceptible to EDM. They are generic functions in organizations that: EDM applications which generate value in supporting the organization can be organized into seven generic categories.
  1. Improving the publishing process
  2. Supporting Organizational Processes
  3. Supporting Communication Among People and Groups
  4. Improving Access to External Information
  5. Creating and Maintaining Documentation
  6. Maintaining Corporate Records
  7. Promoting Training and Education


Improving the Publication Process. Technology is enabling a major restructure of the process of publishing and distributing paper documents. For those organizations which produce documents as a product or as support for a product, this change is reengineering their document production processes. The stages of the traditional process, designed primarily for high volume and high quality documents is shown in Figure 5. The document is created, generally with the use of electronic tools, and a photographic plate is made for an offset printing press. The offset press requires long print runs to amortize the extensive set-up costs. Thus, a large quantity of documents is produced and stored in a warehouse, then shipped to their destination when they are required.

This process has several inefficiencies. The offset presses are large, expensive, and use toxic chemicals. The infrequent long print runs require storage of documents which become obsolete between runs. And transportation is an inordinately large part of the total cost of the process. In fact, R & R Donnelly, the country's largest publisher, estimates that 60% of the total cost of delivering these documents is in storage and transportation.

Figure 4 shows the steps in the revised publishing/distribution process using newer technologies. Documents are stored electronically, shipped over a network, and printed when and where they are needed. The major benefits result from the reduction of obsolescence (revisions are made frequently to the electronically stored version), elimination of warehouse costs, and reduction or elimination of delivery time.


Figure 3. Traditional Prublishing Process


Figure 4. Reengineered Publishing Process

Supporting Organizational Processes - Documents are the vehicle or mechanism through which most processes in organizations are accomplished. Typical examples include processing a claim in an insurance company, hiring a new employee, or making a large expenditure. The documents are primarily forms, which flow through the organization carrying information, and accumulating input and approval from a sequence of people. These "workflow systems" are still heavily based on the physical circulation of paper forms in most organizations.

The use of technology to support these processes generates significant value in reduced physical space for the handling of forms, faster routing of forms (especially over geographical distances) and managing/tracking of forms flow and overall workload. Two new trends in organizations are increasing the importance of these workflow systems: quality improvement processes and process reengineering, both of which tend to be heavily dependent on documents.

In addition to transaction-oriented business processes which can be improved with EDM, many organizations are finding that documents are important to management processes of reporting, control, decision making, and problem solving. (Stanat 1988) Several executive information systems (EIS), now supply documents to supplement the more traditional data-based reports. Organizations with a custom developed EIS are also adding so-called "soft" information, in the form of documents. (Watson and others 1993)

Supporting Communication Among People and Groups - The purpose of applications in this area is to facilitate communication among people, and groups of people, in organizations. In the broadest sense, all EDM applications support this function, but included here are specific systems to support the transfer of information among people across time and space. Communication can take place without documents, of course. The conversation in the hall, a phone call, a video conference, a presentation -- all are communication events that do not necessarily require a document. However, if the concepts, ideas, and information are to be communicated over time, they must be captured in a document. If they are to be communicated over distance, without voice or video connection, they must be captured in a document. Even when communication takes place between people at the same time and in the same place, a document might be used to improve the articulation or formation of the concepts. The primary value of EDM applications in this category derives from the richer communication offered by multimedia or compound documents, and the reduced time needed for the electronic distribution of documents. An additional value results from the sharing of documents among a group of people simultaneously, coupled with the rapid feedback and interaction that ensues. The popularity of Lotus NOTES to support a variety of organizational communication illustrates this set of benefits.

Improving Access to External Information - The purpose of applications in this area is to provide better access to external information resources. Two general kinds of external resources are time-critical information (news) and reference material. The documents include news wire items, newspapers, periodicals, magazines, electronic bulletin board items, books, video tapes, research reports, proceedings of conferences, etc. Traditionally performed by a library, these applications are increasingly computer-based with on-line card catalogs, direct user access to on-line text databases (e.g. DIALOG, NEXUS), circulation of full text research papers, etc.

A major emerging document reference and access service is the wide area information system (WAIS). (Kahle and Medlar 1991) It consists of a consortium of Universities, government agencies, research institutes, and other organizations to share access to the full text of a document collection through a client/server network. Other document collections are available on the Internet. FTP, Gopher, and Archie allow access and transfer of text files from servers across the network. The World Wide Web is a multimedia, hypertext collection of documents managed by the Center for European Nuclear Research (CERN) in Switzerland. To access and explore the WWW collection, the National Center for Supercomputing Applications (NCSA) has developed Mosaic. NCSA Mosaic is a distributed hypermedia system designed for information discovery and retrieval on the global Internet. Mosaic provides a unified interface to various protocols, data formats and information archives using powerful methods for discovering, using and sharing information.

Creating and Maintaining Documentation - Another cluster of applications in EDM involves the creation, maintenance and distribution of documentation. The purpose of these applications is to maintain documents that contain policies, procedures, reference material, product descriptions, etc. They differ from records management applications which capture and store documents for archival purposes, accessing them infrequently, usually by request from an internal user. Documentation applications, on the other hand, maintain the "current version" of documents which must be updated and accessed frequently by a wide variety of requesters. Documents tend to be reports, manuals, drawings, and reference material; they have been mostly text in the past, but are increasingly multi-media. Examples include:



Access to documentation can be provided in several ways. For internal users, on-line access through a workstation is most common. For external users, access to documentation improves customer service by providing answers to customer queries or solving problems with the right reference material.

When access is provided to reference material, not on-line, but through periodic distribution, these applications become "electronic publishing". When the reference material goes beyond basic status reports or reference material, it may become an "information product" of value to customers or as a marketing tool usable by dealers or distributors.

The benefits of EDM for these applications are (1) quicker access to the documents, (2) more efficiency in the search process, (3) simultaneous access by several people to the most current version of the document, and (4) reduced cost of printing and distributing documents.

Maintaining Corporate Records - Organizations must maintain official documents and records concerning their obligations, agreements, and financial performance, primarily to satisfy legal requirements. Traditionally the responsibility of the Records Management Department, this application area involves storage and retrieval of contracts, financial records, internal reports, and other important corporate documents. These are "documents" in the traditional sense, mostly text, mostly on paper. The role of EDM applications in this area is to manage this set of official corporate records by providing archival storage, and occasional retrieval. The methodologies, approaches, and technologies have evolved from a tradition of library operations, from an earlier emphasis on automated records center applications, micrographics (film and fiche), and computer output microfilm (COM), to an emphasis today on digital image processing. For records management and documentation applications, a government initiative is becoming important. CALS (Computer-aided Acquisition and Logistic Support) has been defined as a requirement for documentation by the U. S. Government . (Carter and others 1989) Starting with the Defense Department, it has been adopted by many other government agencies. (Zurier 1992) Eventually, it is likely that any organization with government business will need to be CALS compliant.

For large records management applications, the savings from image processing in storage space and ease of retrieval alone, are impressive. Other sources of value from the application of technology to records management include: Promoting Training and Education - The purpose of the applications in this area is to teach or train people in an organization. The documents are curricular training materials or reference materials, and the use of multi-media documents, perhaps with hyperstructure, are proving to be extremely effective. A primary characteristic of these applications is the continuous, sequential interaction between the user and the information through the learning process over time, rather than a specific search and retrieval event to obtain a document.

Training and education applications are good early examples of the use of multi-media documents and hypertext. As mentioned earlier, hypertext is the most promising approach to structuring conceptual information. The body of knowledge to be learned or understood consists of "concept nodes" which are linked or cross-referenced to form a "web" of ideas and concepts. An excellent example is the Intermedia research project at Brown University. (Yankelovich and others 1988)

Converging Application Areas

These categories of applications illustrate the benefit and value of EDM. The good news is that there are many opportunities in many different areas. But these applications use many common approaches and technologies, and, as the earlier case example illustrates, it will be desirable for them to converge eventually. If they have been developed separately, without a plan to integrate them, it will be very difficult to reap the potential benefits.

ROLES AND RESPONSIBLILITES

Because of the pervasiveness of documents, all organizational units will find it important to use technology for document management. There are, however, several departments that have had primary responsibility for one or more functions in document management. It is these "document support" departments that will find it especially important to work together and coordinate their efforts. It will become their responsibility to develop the infrastructure needed to enable electronic document management. Thus, the third perspective for structuring EDM is identifying and defining the roles and responsibilities of these document support departments, so they can accommodate the document processing needs of all other departments in the organization. The primary document support departments are: The result of these developments is that all the groups above will be facing significant changes in their traditional work and responsibilities. In addition, new ways of handling documents will affect the work practices of almost everyone in an organization. The I/S Department, as the developer and manager of the technical infrastructure for EDM, will be in a position to lead this evolution and become the organization's major change agents as they have been in the EDP and MIS eras. However, assuming this leadership role will require the I/S department to take some specific steps, such as those described below.

CONCLUSION

The previous sections of this paper have argued that EDM is a potentially significant development in organizations, but that it will create some major changes in the ways organizations process information and conduct business. The technology is new, powerful, and rapidly evolving, but that rapid rate of change will make it difficult to build a compatible technical infrastructure to support document management.

This effort is justified, however, because there are applications in several business areas that can benefit from EDM tools and techniques. Because these early applications are springing up in diverse areas, there is a danger that they will evolve separately although they benefit from being integrated. Therefore, planning will be important to build an integrated document technology infrastructure. In this process, several departments or organizational units with a history of document management will need to work together, even though they have different history, background, and perspectives.

The I/S department has the opportunity to play a leadership role in coordinating the efforts of these user departments and document support departments, in order to evolve the infrastructure and applications needed to support EDM. Playing this leadership role may be more difficult than it has been in the past. In this era of distributed systems and distributed responsibility for systems initiative, I/S managers will need to educate a variety of user departments and document handling departments, convincing them to cooperate in the development of an EDM strategy and technology infrastructure. Here are some steps that I/S managers can take now to prepare for these developments.

Form a "Document Council". Form a council consisting of representatives from each of the document support departments identified above which have been charged with managing some part of the document processing cycle in the past Their first assignment would be to identify mission-critical documents and work back to applications and departments that depend on them. An initial set of applications will undoubtedly evolve from the members of the group. They have probably been responsible for producing and managing these documents in the past.

There will also be important applications which have been developed directly by the user departments, so the group should develop a way to find important applications of which they are not aware. Mechanisms for doing this include examining the areas and examples mentioned earlier in this article, finding examples in journals and trade publications, distributing surveys and questionnaires in the organization, etc. The work of this group and the technology tracking group described below should proceed in parallel, with periodic joint meetings for coordination and status reporting.

Form a Document Technology Group.Assign the task of tracking and forecasting the emerging document technologies to a small group with technical proficiency in several areas. If there is an advanced technology group, this assignment would probably fit into their charter. The assignment should cover both the infrastructure technologies as well as document technologies.

Prioritize Applications. The application group, perhaps in consultation with people who have been using the documents, should then prioritize the applications by business value and technical feasibility. There might be a difference in long and short range perspectives, so both should be considered. The prioritization should also include consideration of fit or linkage between applications, especially when two or more applications can use the same technology or approach.

Develop an EDM Plan. As a result of their regular joint meetings, the document council and the document technology group can jointly develop a plan for adding the necessary technology to the infrastructure and developing the applications. These applications might not be developed by the I/S department, or even by the departments' represented in the group, but their development and approximate time schedule should be included in the plan. As it is refined and developed over time, this plan becomes an integrated EDM architecture and a plan for implementing it.

Revise Responsibilities. By this time, it may become clear that some of the roles and responsibilities of some departments may need to be revised. The council can develop recommendations to management concerning these changes. By performing this step last, any shifts in responsibilities will result from discussions based on the evolution of the applications and technologies, measured by business value. This will reduce the probability of a "turf war" that could result from the changes in the way documents are managed.

The benefits of EDM will evolve as the technology and our ability to use it evolve over the next several years. It is not too early, however, for I/S managers to begin the planning processes to build the technology infrastructure for document management, and to harness these new technologies to improve the performance of their organizations.

ACKNOWLEDGEMENTS

The concepts and ideas in this paper evolved over the past several years from a review of literature in several contributing disciplines and technologies, and an analysis of more recent literature that is "converging" into EDM. (Sprague 1990) These reviews led to a series of interviews with over 100 managers, executives and professionals concerned with the development of this topic. The interviews, conducted over the past two years, sought information and opinion on the opportunities, importance, and "philosophy" of EDM, as well as examples of current and potential applications. The categories of business value, the conceptual technology architecture, the application areas, and the suggested action plan evolved from these interviews. Early versions of the paper were reviewed by CIOs who were members of the Dooley Group, and the Xerox Executive Advisory Forum. Their contributions and suggestions are acknowledged and appreciated.

REFERENCES

Allerding, Robert N. "Cost Savings Today : Records Management Tomorrow." Robert N. Allerding, CRM-FAI, 61 Summerset Rd., Delaware, Ohio 43015., 1992.

Baran, N. "Breakthrough in Holographic Memory Could Transform Data Access." BYTE 16 (January 1991):

Beal, Ellen. "Smart Documents: Document Based Programs Promise More Customized Publishing Systems." Computer Graphics World 14 (May 1991): 53- 59.

Bieber, Michael P. and Steven O. Kimbrough. "On Generalizing the Concept of Hypertext." MIS Quarterly (March 1992): 77-93.

Butler-Cox Foundation. The Future of Electronic Mail. Butler Cox Foundation, 1991. Research Report 82.

Card, Stuart, George G. Robinson, and Jock D. Mackinlay. "The Information Visualizer, An Information Workspace." In CHI'91 Conference Proceedings, 181-187. April 1991 ed., Vol. ACM Press, 1991.

Carter, F., B. Babcock, and L. Wylie. CALS: Its Intent and Impact. Corporate Publishing Strategies, Stamford, Conn.: The Gartner Group, 1989.

Chen, H., K. J. Lynch, K. Basu, and T. Ng. "Generation, Integration, and Activation of Thesauri for Concept-Based Document Retrieval." IEEE Expert (1993):

Clarkson, Mark A. "The Information Theater." BYTE (November 1992): 145- 151.

Conklin, Jeff. "Hypertext: An Introduction and Survey." Computer (September 1987): 17-40.

Daft, Richard L. and Robert H. Lengel. "Organizational Information Requirements, Media Richness, and Structural Design." Management Science 32 (May 1986): 554-571.

Daft, Richard L., Robert H. Lengel, and Linda Klebe Trevino. "Message Equivocality, Media Selection, and Manager Performance: Implications for Information Systems." MIS Quarterly (September 1987 1987): 355-366.

Daniels, Barry. "Introduction to Imaging Technology." DMR Vol. 3 (January 1993):

Datapro, ed. Document Imaging Systems: Technology Overview. Delran, N.J.: Datapro Research Group, 1991.

Davenport, Thomas H. Process Innovation: Reengineering Work Through Information Technology. Boston, Mass.: Harvard Business School Press, 1993.

Elam, J. and J. SvioklaThe Image Processing Project at USAA.. Case Study, N9-190-155, Boston, Mass.: Harvard Business School, 1990.

Galbraith, J. R. "Organization Design: An Information Processing View." In Readings in Interpersonal and Organizational Communication, ed. R. C. Huseman. Boston, MA: Allyn and Bacon, 1979.

Gilbane, Frank. "Document Management and Databases." In The Gilbane Report, 1. Arlington, Mass.: Publishing Technology Management, June 1993.

Gilbane, Frank. "Electronic Delivery - What are the Implementation Issues for Corporate Applications." In The Gilbane Report, Arlington, MA: Publishing Technology Management, Inc., September 1993.

Gleckman, Howard. "The Technology Payoff." Business Week, June 14, 1993, 56-68.

Gulden, Gary and Robert H. Reck. "Combining Quality and Reengineering Efforts for Process Excellence." Information Strategy: The Executive's Journal (Spring 1992): 10-16.

Halasz, Frank G. "Reflections on Notecards: Seven Issues for the Next Generation of Hypermedia Systems." Communications of the ACM 31 (July 1988): 836-852.

Hale, David P., William D. Haseman, and Frank Groom. "Integrating Islands of Automation." MIS Quarterly 13 (December 1989): 433-445.

Hammer, Michael. "Reengineering Work: Don't Automate, Obliterate." Harvard Business Review (July-August 1990): 104-112.

Hammer, Michael and James Champy. Reengineering the Corporation . New York, New York: HarperCollins Publishers, 1993.

Harvey, David A. "Optical Storage Primer." BYTE (Fall 1990): 121-130.

Herr, Laurin and Judson Rosebush, ed. The Future of Workstations. Delran, N. J.: Datapro Information Services Group, 1991.

Kahle, Brewster and Art Medlar. "An Information System for Corporate Users: Wide Area Information Servers." Online, September 1991, 56-60.

Kind, Joachim and Frank Eppendahl. "The Need for Office Analysis in the Introduction of Electronic Document Management Systems." Document Image Automation 12 (Summer 1992): 31-35.

Lasher, Donald R., Blake Ives, and Sirkka L. Jarvenpaa. "USAA-IBM Partnerships in Information Technology: Managing the Image Project." MIS Quarterly (December 1991): 551-565.

Levien, Roger E. The Civilizing Currency: Documents and Their Revolutionary Technologies. Rochester, N. Y.: Xerox Corporation, 1989.

Lundeen, Gerald W. "Text Retrieval Software for Microcomputers and Beyond: An Overview and A Review of Four Packages."Database (August 1992): pp. 51-63.

Mander, Richard, Gitta Salomon, and Yin Yin Wong. A 'Pile' Metaphor for Supporting Casual Organization of Information. Apple Computer, Inc.., 1992.

McDonnell, Edwin D. and Gary E. Somerville. "Corporate Reengineering That Follows the Design of Document Imaging." Information Strategy: The Executive's Journal (Fall 1991): 5-10.

McLeod, Raymond Jr. and Jack William Jones. "A Framework for Office Automation." MIS Quarterly (March 1987): 87-104.

Meier, Johannes and Ralph H. Sprague. "Toward a Better Understanding of Electronic Document Management." Decision Science Department Working Paper, University of Hawaii, 1993.

Michalski, G. P. "The World of Documents." BYTE (April 1991): 159-170.

O'Gorman, Lawrence and Kasturi Rangachar. "Document Image Analysis Systems." Computer (July 1992): pp. 5-8.

Plesums, C.A. and R.W. Bartels. "Large Scale Image Systems: USAA Case Study." IBM Systems Journal 29 (3 1990): 343-355.

Popkin, J. and A. Cushman. Integrated Document Management - Controlling a Rising Tide . Stamford, Conn.: Gartner Group, 1993.

Rymer, John R. "Objects at Floodtide: Object Orientation Permeates New Development." Patricia Seybold's Office Computing Report , June 1989, 1-16.

Seymour, Jim. "An Interface You Won't Outgrow." PC Magazine , September 12, 1989,

Shank, M. and R. LaTarce. "Color in Communication Materials." Marketing News, August 6, 1990,

Shetler, Tim. "Birth of the BLOB." BYTE , 1990, 221-226.

Skapinker, Mark. "Warm for Forms." BYTE (April 1991): 166.

Sprague, Ralph H. "Converging Technologies: New Opportunities for Document Management." Journal of the Records Management Association of Australia (1990):

Sproull, Lee and Sara Kiesler. "Computers, Networks, and Work." Scientific American 265 (September 1991): 116-127.

Stanat, Ruth. "Building a Document Based Competitive Intelligence System." In Proceedings of the International Conference on Decision Support Systems , The Institute of Management Sciences, 1988.

Stewart, Jon A. A Proposed New Architecture for Open Information Exchange and Processing. 1992.

Swanson, E. Burton and Mary J. Culnan. "Document-Based Systems for Management Planning and Control: A Classification, Survey, and Assessment." MIS Quarterly (December 1978): 31-47.

Travis, Robert L. "CDA Overview." Digital Technical Journal 2 (Winter 1990): 8-15.

Vienzenu, A. "Color on Forms." Journal of Forms Management (August 1988): 13-14.

Wallace, Scott. "Image Archiving." Corporate Computing 1 (October 1992): 75-82.

Watson, Hugh J., Candice G. Harp, Gigi G. Kelly, and Margaret T. O'Hara. Soft Information in Executive Information Systems: Conceptualizations, Findings, and Propositions. College of Business, University of Georgia, 1993.

Weinberger, D. D. "The Active Document: Making Pages Smarter." The Futurist (July-August 1991):

Williams, Lowell. "Compound Documents." Dr. Dobb's Journal (March 1993): pp. 32-39.

Yam, P. "Plastics, Benjamin ... Data Storage in Photorefractive Polymers." Scientific American (June 1993):

Yankelovich, Nicole, Bernard J. Haan, Norman Meyerowitz, and Steven Drucker. "Intermedia: The Concept and the Construction of a Seamless Information Environment." IEEE Computer 21 (January 1988 1988): 81-96.

Zurier, Steve. "Here's the Straight Talk About Standards." Government Computer News, September 14, 1992,