Intelligent Enterprise

Better Insight for Business Decisions

RSS
Webcasts
Whitepapers
Subscribe
Home

December 10, 2003

Smart Synergy

With real-time business becoming a key objective, the worlds of enterprise application integration (EAI) and business intelligence (BI) are converging. What will it take to forge the most perfect union?

by David S. Linthicum

Two worlds have converged over the last year or so, those of application integration (or EAI) and business intelligence (or BI). The fit is logical. Application integration technology tends to deal with the transactional movement of information between source and target systems in support of real-time or near real-time business transactions within or between companies. BI deals with making sense of this information and using it for tactical and strategic decisions.

In the last several years, application integration — at least the notion — has worked its way into most IT departments, typically after the notion of business intelligence. Several emerging developments have driven this trend, including the need to expose information found in existing systems to the Web, the need to participate in electronic marketplaces, the need to integrate the supply chain, and perhaps more important, the need to enable existing enterprise systems to share information and common processes.

Application integration is less about J2EE vs. .Net and more about understanding the requirements and future growth of the problem domain, a not-so-sexy activity that is all too often avoided. Without this application integration infrastructure, the concept of real-time business won't work — we simply wouldn't be able to see into existing information systems. Other real-time technologies (such as real-time data warehousing) wouldn't be able to work either.

If we accept the need for real-time analytic visibility into existing information systems and the need for historical analytics, we therefore must accept the fact that application integration and business intelligence go hand in hand. The emerging synergy between these technologies is a foregone conclusion.

In this article, I'll explore how the worlds of application integration and business intelligence intersect and how your enterprise can benefit from the synergy between these two technologies.

Ready to Date

The synergy between EAI and BI is not at all new, at least in concept. Ever since we learned the value of using data warehouses and data marts to gain a better understanding of our business, we began to wonder how we could remove the latency from this information to compare the old with the new in the same analytic framework.

While BI systems give us access to information used for making critical strategic decisions, the latency of this information — at least in many businesses — has proven to be problematic. Information analyzed to determine the state of a business could be months old and much less valuable to organizations that need to be more proactive than reactive, such as retail and manufacturing.

The vision of this BI/EAI merger is easy to define. We know how to analyze historical information using the accepted analytic tools, so it's easy to see why a similar view of information in real time would be valuable. (See Figure 1) To this end, users would have the ability to view aggregated sales from, say, the last three years along with sales that occur at this very moment. They'd be able to understand the current state of the business as it happens as well as within the context of historical data, and be able to correct it in process.

For instance, a retail organization that depends on profits from razor-thin margins would have real-time visibility into current sales within all stores, as they occur, monitoring data points such as profitability per product. The organization would also understand how the currently observable behavior compares to the same behavior last month, last year, or perhaps for the last 10 years. Using such comparative analysis, business users could determine what is desirable through historical analysis and how that reflects on what is currently occurring in the business. My hypothetical retail business would be able to adjust prices, products, salesperson incentives, and even processes to maximize profitability using almost perfect information with sophisticated analytics.

It's important for both real-time and historical information to exist within the same analytic framework. Business activity monitoring (BAM), or the visibility and analysis of real-time information, loses its value when the real-time and historical information are decoupled.