Stewart Mckie
This is an opportune time for developers who can advance the evolution of enterprise resource planning (ERP) applications. Most vendors of ERP software, in response to changing market demands, are now “extending” the domain of their ERP systems by integrating front-office and business intelligence applications. But this “band-aid” tactic is unlikely to keep the ERP ball rolling with any significant forward momentum. We need something beyond ERP: a new generation of business management applications built from the start with the goal of delivering business benefits.
Despite all the claims, ERP has basically been an expensive fix. ERP was an evolution, not a revolution, and has always been a tactical rather than strategic initiative. Admittedly, ERP solved a number of critical problems with traditional business management systems. ERP helped move IT shops off proprietary hardware platforms and databases onto more open systems and easier-to-access relational databases. ERP replaced messy combinations of loosely interfaced, built-and-bought systems with tightly integrated and relatively complete single-vendor application solutions.
But in the end, from an architecture perspective, ERP is simply another foundation layer that sits above the database, operating system, and hardware. (See Figure 1.
FIGURE 1 ERP as platform layer.
In reality, ERP software still does not manage all the resources of an enterprise and probably never will. The real return on investment from ERP comes when this foundation layer is in turn exploited by other maturing application suites such as those designed to deliver business intelligence, employee self-service and front-office systems, integrated supply chains, or Web-based electronic commerce. But the idea of snapping on new modules such as sales force automation (SFA) and customer relationship management (CRM) components can’t hope to revive the glory days that ERP enjoyed in the mid to late 1990s. We need new directions — a focus on asset management, cellular design, self-managing applications, and collaborative commerce, for example — to take us beyond ERP.
Asset Management Focus
In the pre-ERP days, business management software was organized into functional silos such as accounts payable, purchasing, and fixed assets. ERP vendors have been instrumental in breaking free of this silo mentality and introducing a design more focused on workflow and based on cross-module business processes such as procurement and fulfillment. Perhaps the time has come to move forward again and adopt an asset management perspective to the design of business management software, transforming the extended ERP scenario into the customer asset management solution. (See Figure 2.)
FIGURE 2 From extended ERP to customer asset management.
This method of designing and using business management software is new, and few businesses are prepared for it. How many businesses do you know with a VP of customers rather than a VP of sales and marketing, for example? The asset management approach to business management requires some significant realigning of organization and compensation structures before it can even be considered. Nevertheless, this asset management focus has great potential, considering how you could also apply it to the other assets that most businesses need to manage—products, people, partners, projects, plant, and processes, for instance. Maybe ERP’s next phase is, in fact, enterprise asset management (EAM), and the business benefit is a focus on asset management rather than application management.
Cellular Design
ERP applications are still largely monolithic in design. Such a design necessitates an all-or-nothing buying, deployment, and upgrading decision by ERP software customers. ERP is replacing monolithic designs with more flexible modular deliverables but now the time has come to move on to more granular design-and-deploy solutions that further break down modular software into “cellular” deliverables—whether you call them business objects, components, or functional granules. (See Figure 3.)
FIGURE 3 From monolithic to cellular software.
You build cellular applications using an object-oriented design that mandates the encapsulation of data and business logic by software “objects” that expose “methods” used to manipulate objects’ behavior. Because every object must have one or more methods to collaborate with other objects, every cellular application will offer a “cell-level” applications programming interface (API). This API opens up the possibility of building software applications that are assembled to fit the business need rather than the other way around. For example, vendors could offer pre-assembled cellular frameworks, or customers could purchase cell-assemblies in the form of processes or cell “clusters” as their business needs dictate. SAP’s Business Application Programming Interfaces (BAPIs) indicate the potential of future cellular software to integrate more easily with both internal and external application cells. There are significant business benefits to incremental rather than monumental ERP solutions. ERP customers can buy, deploy, and upgrade functional cells individually as needed and integrate them in application-service clusters, possibly integrating cells from specialist vendors via cell-level APIs.
Self-Managing Applications
Most ERP software still forces its users and managers to function largely in reactive mode. The software should be designed to be self managing; it should recognize and respond to a wide range of business events, making decisions based on preprogrammed rule logic that reflects knowledge about the business. Not all business intelligence needs to be done by analysts “mining” integrated or interfaced data marts manually and data warehouses using desktop online analytical processing (OLAP) tools. A great deal of business intelligence can be determined, implemented, and delivered by the ERP system itself if it lets a user program rules around known business events in order to trigger workflows of any complexity. Current ERP workflow and alert capabilities tend to focus on transaction workflow automation or generating messages triggered by the occurrence of basic exception conditions. However, at the core of a self-managing application is a rich set of out-of-the-box events that are user modifiable, with extensive rule-building logic and functional linking capabilities (to hook in screens and reports, for example).
ERP applications that have offered this type of capability for some years, such as Geac Computer Corp.’s SmartStream (www.geac.com), have not been especially popular in the marketplace. But they lack popularity because they focus on workflow capability rather than self management. Business event management requires a whole new layer of business logic that’s ideal for middle-tier application servers and could let ERP systems evolve into more proactive tools. Self-managing applications further reduce human involvement in systems administration and use, and help accelerate the transformation of staff from clerical task managers into knowledge workers who deal with exceptions.
Collaborative Commerce
The combination of workflow and the Web has the ability to transform business supply chains in ways that are difficult to predict in the early adoption phase of collaborative commerce over the Internet. Today, everybody accepts that the days of traditional electronic data interchange (EDI) are numbered and that new emerging standards for Internet data interchange, Web content aggregation, and programmatically-open, value-chain pipelines are where the action will be in the future. Current ERP suites provide an excellent integrated foundation for layering on new collaborative com- merce applications that take advantage of emerging standards, as well as the rapidly increasing number of application service providers (ASPs) on the Web. Collaborative commerce applications depend on cross-application workflow engines and the Internet to provide the end-to-end transaction integrity and communication channels required to minimize the distances among trading partners. Each of the “beyond ERP” directions outlined here really demand a major redesign of current ERP applications and are not quick fixes. The first psychological boundary for ERP vendors and their customers to cross is recognizing that ERP suites are already legacy applications and merely a means, not an end, for the next great leap in business management software. ERP’s current success means it will be a tough act to follow. Going beyond ERP, rather than simply extending it, will be a significant challenge for the current generation of ERP vendors. Some are going to fall by the wayside in the style of previous market leader Dun & Bradstreet Software, which was consumed in a fire sale by Geac. ERP stockholders hardly have a bright outlook, but for developers with ERP application expertise, the future looks bright.
Guest columnist Stewart McKie is an independent software analyst specializing in accounting software. You can reach him via his Web site www.cfoinfo.com.
|
