Enterprise application integration
EAI is the soluble glue needed for modular relationships that allow organizations to be flexible and responsive to market demands.
In the December 2000 issue of InfoPro, Bob Tillman, director of public relations and government advocacy for ARMA International, wrote of a visible future where new business models include modular corporations, outsourcing, and network relationships: “Modular relationships provide firms with greater flexibility, a more rapid response to market changes, a greater competitiveness in the short run, and less risky innovations. It is a future in which systems built to change are more valuable than systems built to last.”
Enterprise application integration (EAI) is a kind of technological Velcro, enabling computer systems to accommodate such change. EAI allows diverse systems to connect with one another quickly to share data, communications, and processes, alleviating the information silos that plague many businesses. The benefits of assimilating new systems without prolonged programming efforts are apparent following merger and acquisition activity. EAI solutions provide a way to connect the systems of collaborators, partners, and others for as long as necessary, decoupling when the relationship ends. EAI is, in essence, the soluble glue for the modular corporation.
In its April 2001 report for AIIM International, “Enterprise Applications: Adoption of E-Business and Document Technologies, 2000-2001: Worldwide Industry Study,” Gartner defines EAI as 11 the unrestricted sharing of data and business processes among any connected applications and data sources in the enterprise.” Gregg Wright, senior vice president for e-solutions services at McLean, Virginia-based Information Management Consultants Inc., defines EAI as “a strategy for making business processes more valuable… in its simplest form [it] involves passing data between application systems, and at its most complex, [it] involves rationalizing overlapping business processes, disparate data structures and diverse technologies.” Market forecasts for EAI predict it will be in the $11 billion to $12.5 billion range by 2005 with 22 percent annual growth, according to Meridien Research.
EAI is considered strategic because its potential contribution is measured in terms of attaining or exceeding key performance and competitive benchmarks for entire industries, as noted in the following examples.
Banking. The basis of competition in banking and financial services is customer retention. Customers with multiple accounts are less likely to change, but most financial institutions have stovepipe systems for credit cards, checking, savings, mortgage, brokerage, and other services. An EAI implementation would integrate the systems so that a data warehouse can aggregate account data, provide a single view to the customer, and recommend what additional products the customer should be offered. In EAI systems instituted at Bank of America and Royal Bank of Canada, a transaction in one account triggers an event in another process for a marketing contact.
Manufacturing. Manufacturing’s key competitive measure is cost reduction through inventory control and justin-time processes. Allowing outside suppliers to view inventory data is possible with enterprise resource planning (ERP) applications (see sidebar), but EAI provides a way for the manufacturer’s systems to communicate with external systems to track sales, forecast demand, and maintain a detailed view of pricing and availability among many suppliers. General Motors’ Covisint is a custom vehicle order management system that interfaces with inventory, planning, and logistics applications in the supply chain.
Life Sciences. In the life sciences industries, being first to market is critical since patents on new compounds expire within 17 years from the time the first documents are submitted to the Food and Drug Administration. EAI applications in pharmaceuticals, biotechnology companies, and research institutions integrate data from diverse laboratory systems with clinical data and other core systems so that it can be available to analytical applications. Several such projects are quietly underway within U.S. companies working on human and animal genome projects.
Reaping the Benefits
Christy Bass of Accenture, writing in the April 2001 issue of eAI Journa4 estimates that integrating enterprise application software with customer relationship management (CRM) functions can deliver as much as $17 million in profit to the average $1 billion business unit. Major savings include lower costs per transaction, decreased error correction costs, and order-processing speed. Whether such estimates are realistic or not, return on investment is crucial to project funding. EAI is a long-term investment, one in which software costs alone range from $400,000 to $700,000, with large implementations costing $1 million or more.
The strategic advantage of EAI is not just the exchange of data but how it is achieved as well. Prior efforts to integrate computer systems involved writing code that allowed one computer to speak to another, as shown in the integration approaches listed on page 52. Each source application (the one sending data) needed a custom written connection to the target (the computer receiving the data). Custom connections are expensive because they take time to write and maintain. Whenever there is any change to either system (e.g., an upgrade), new connections have to be written.
In contrast, EAI makes use of middleware (a server and software) into which many applications may connect (see Figure 1). Functioning as a central hub, the EAI server transforms data from one application to another and can coordinate processes among applications. This means, for example, that when a bank customer changes the name on a personal checking account, it automatically changes on all of the customer’s accounts. The EAI system may also send a message for an insurance representative to contact the customer, as name changes usually signify a change in marital status.
Applications may be added to or removed from the middleware hub as situations dictate, without intricate programming. In industries where third parties provide critical services for example, an outsourced canning plant contracting with a major food processor, or a contract research organization working with a drug company – this is a boon. Web services, which are standards-based tools that can pull together information from multiple sources over the Internet, are garnering much attention as the focus of EAI projects shifts from integrating in-house systems to concentrating on collaboration with partners, co-marketers, suppliers, and customers.
Companies that have begun using Web services have quickly found it is the underlying infrastructure – workflow rules, validation routines, and the like — that is more complex than anticipated. According to Wright, “Understanding the existing technology environment, describing the business operations model, and determining which business units, processes, systems, and data are involved is rarely easy.” EAI projects must spread their considerable costs over multiple systems in order to produce meaningful return on investment, and most strive to show results within six months so as not to lose senior management’s attention.
Several information management issues are emergent in EAI’s expensive, highly visible, and complex projects. Sharing information raises concerns about systems access and security. Presumably, customers, suppliers, and collaborators cannot have unlimited access to data in connected systems. How to ensure that connected systems permit information sharing without information theft is a concern. Secondly, consumer data privacy is the subject of varying legislation throughout the world. If systems are connected, assuring compliance becomes that much more complicated. The idea of classifying data into categories so that access and security rules can be applied to each is beginning to take hold.
Records, in the sense of entities to be managed, are ephemeral. There is no single, complete consumer transaction record to manage because the transaction maybe spread over several connected systems – CRM, order entry, warehousing, shipping, and accounting. The transaction record actually consists of a series of field locators, each identifying the system and the database table in which a particular bit of information resides.
The good news is that EAI may be configured to query across all the connected systems; the bad news is that the record exists only in response to such queries. In addition, the records that outsiders and employees within the company see may vary considerably due to different security rules for users. The ability to capture a Web transaction and preserve it as a record is part of Bostonbased Tower Technology’s Web Capture offering in partnership with Canberrabased Tower Software’s TRIM.
It also is likely that connected systems will require that records be thought of as the products of specific processes. The idea that specific workflows generate particular classes of records is not lost on records management (RM) software companies and others. MDY Advanced Technologies’ File Surf is an early example.
On the plus side, could a middleware approach be used to integrate records management software systems with transaction-based e-business systems? Application programming interfaces have not been entirely worry-free for integrating RM software to other applications, such as document management systems. Watch for leading-edge RM software companies to explore a middleware approach.
Spencer Johnson’s Who Moved My Cheese (see review in this issue) admonishes everyone to adapt to change quickly. If the business world’s cheese is constantly in motion, EAI may be information technology’s best hope of being ready to adapt again and again.
(Editor’s Note: The products mentioned in this article serve as examples by the author and do not constitute endorsement by ARMA International.)
Julie Gable, CRM, CDIA
Julie Gable, CRM, CDIA, is the Principal of Gable Consulting, an independent firm that specializes in bringing strategic business perspectives to information management issues. Gable is also the Associate Executive Editor of The Information Management Journal and a Contributing Editor for Transform Magazine. She may be contacted at JulieGable@ aol.com.
Copyright Association of Records Managers and Administrators Inc. Mar/Apr 2002
Provided by ProQuest Information and Learning Company. All rights Reserved