Using my credit union's home banking service recently, I transferred money from checking to pay a credit union car loan. The system immediately reflected my checking withdrawal, loan payment, and new checking and loan balances. Two days later I paid my mortgage the same way. Problem is, the mortgage payment and new balance didn't show up right away (although the withdrawal did). Sound familiar? To members, this discrepancy is confusing and frustrating. To those of us in the business, it makes perfect sense: The mortgage is processed on a third-party system updated in batch mode overnight; the car loan, on the credit union's core system in real-time. But members don't know that, and they shouldn't have to. Which is just one of many reasons that EAI – Enterprise Application Integration – will be a dominant topic in credit union technology over the next several years. In its simplest terms, EAI enables the intelligent sharing of data and business processes among many applications or data sources within an organization. There are many reasons why EAI must mature and take hold in the credit union industry, but they boil down to two types of business issues: internal and external. On the internal side, every credit union has a constellation of different networked applications and systems – providing much-needed functionality to support the business, but inevitably involving inefficiencies that increase costs and introduce errors. You're thinking "workflow" as the solution, right? Unfortunately, workflow as we know it is far from sufficient, since it typically guides activities within a single system. But many business processes span multiple systems that don't interact. Opening a new member account is a prime example. It might require using separate systems for OFAC compliance, Patriot Act compliance, check ordering, signature card creation, plastic card issuance, credit bureau inquiry, Internet banking sign-up, even loan origination (if the member came through a dealer-direct program). Whew! Assuming your staff remembers what to do in what order, each step might require re-keying redundant data, trigger data entry mistakes, or otherwise slow the process. Consequently, the efficiencies, cost reductions, and convenience of self-service delivery become negated by slow, inefficient, error-prone business processes. Consider this analogy: If the "McServers" at McDonald's had to key every burger sale into a separate inventory system to ensure adequate product on hand, how fast would you get your order? And how often would they run out? High turnover and difficulty retaining skilled staff only exacerbate the problem. Externally, the issue is just as troublesome. As my opening example illustrates, members commonly see inconsistent data, through an inconsistent experience, from one channel to another. That's because most credit unions have multiple data stores, a natural by-product of the many self-service channels that often require different solutions from different suppliers – all using different databases, some updated in real-time, others batch. When service was primarily face-to-face, members could count on staff to interpret the discrepancies or reconcile differences between data sources. Now, with members directly accessing accounts remotely, there's no employee to help – creating more pressure for a seamless and consistent view, with accurate information, regardless of the channel or transaction. Looking Ahead While there's no magic bullet yet, there are steps that credit unions and their technology partners can take to prepare for EAI. One is the effective use and advancement of middleware – components that facilitate the integration of applications from different sources. If today's middleware is like basic plumbing, the industry needs to move toward "intelligent plumbing" – often referred to as a "service-oriented architecture" (SOA). That means giving middleware the intelligence and ability to make decisions that eliminate the internal and external problems outlined earlier. To do that, middleware must first become "aware" of the credit union's business processes and be able to facilitate their execution efficiently. Reaching beyond the limits of workflow, next-generation middleware must be smart enough to make automated, instant decisions across applications and systems – about which steps to take next, or which are mandatory vs. optional, for example. Second, middleware must become capable of accessing multiple data marts and abstracting information from them into a common format. This would eliminate the need for staff or members, and the applications they use, to remember special "rules" about data, as it all appears to come from a common source. One obvious obstacle is the variety of mechanisms used to make data from various sources available. Today, they range from the ideal (use of pervasive standards like XML, SQL and ODBC) to closed, proprietary interfaces, or even legacy user interfaces and screen scraping. While some suppliers are moving toward modern, standards-based, real-time data feeds, the practice is far from universal and not fully developed. If my credit union had intelligent middleware in place, I'd immediately see the impact of my car loan and mortgage payments, in real-time, through any channel. And the credit union would avoid a costly log-jam of calls from members questioning account balances and other data discrepancies. In the case of opening a new account, intelligent middleware would enable staff to interact with the many required systems much more efficiently – for example, only entering new data when prompted vs. re-keying redundant data. As suppliers move toward the deployment of such solutions, credit unions must position themselves now to capitalize on EAI in the future. One way is to carefully consider data access when choosing a technology supplier or solution. A long-term partner should make a commitment to enable data feeds in an increasingly modern, efficient manner, and to develop and deploy the intelligent middleware needed to make that possible. As factors in the purchase decision, data access and application integration are likely to become as important as any single system functionality. Every industry today seeks the holy grail of EAI. Like CRM was a few years ago, EAI is still in its infancy. But the right long-term approach will enable credit unions to leverage their current system investments and capitalize on all the bottom-line benefits that EAI offers.