The almost deafening buzz these days around the "real-time enterprise" is drowning out important practical questions about what the term real time really means and the nature of real time's impact on business processes.
For example, a few weeks ago, I heard from Scott Hicar, the CIO at consumer electronics company Maxtor. His company has been using Data Integrator software from Business Objects to pull data from its various applications into an Oracle data warehouse. Maxtor executives were then pulling real-time reports from the data warehouse using Business Objects' WebIntelligence -- at least that's what they were doing early in the implementation.
The problem was that all those real-time reports were slightly different, since they had been produced at different times. So executives were spending a lot of their meeting time comparing reports and puzzling over data rather than making decisions, Hicar says. Once Maxtor recognized the root problem, the company easily solved it by tweaking the software so that it pushed reports to decision-makers at several set times throughout the day.
"Now people spend time acting on the information rather than questioning its validity. That was our early learning about the nature of real time," says Hicar. "Every business and business process has different criteria about how timely timely is."
Constantly updated, instantly available information is unnecessary for some business functions -- and sometimes it can be downright detrimental. That's because human decision-makers are still at the center of most business processes, and they need more than a stream of the most current data.
Promoters of real time sometimes forget that the notion had its beginnings in applications that were designed to eliminate the need for people to make decisions. Real time started out as a way to say that the response of the application to a particular event had to be both correct and timely, with timely usually meaning a very short time constraint. Early real-time applications were mostly designed for automated technologies like airplane navigation systems and missile-guidance systems or devices such as cardiac pacemakers.
The program controlling Uncle Ned's pacemaker not only has to trigger the correct electrical impulse; it also has to trigger it within a very precise time frame, or Uncle Ned is in big trouble. Likewise, the guidance system in a Tomahawk missile must respond to deviations in the missile's trajectory within a fraction of a second, or the corrections it makes will be useless to spare unintended targets.
Helping People Do Their Jobs
Clearly, some things can be done more effectively by computers than by people, and those tasks were the first prime targets for real-time applications. And the faster the information flows through those applications, the better.
But the IT systems in place at most companies are there to help people do their jobs, and people make better choices when they have access to context and analysis, which are pretty tough to serve up instantly.
The rise of increasingly more sophisticated real-time analytical tools may help, but there still remain the major issues of cost, data quality and the readiness of companies to make productive use of the analysis.
Smart technologists like Hicar know that their job isn't implementing software that makes decisions, but rather it's "presenting information to human experts to make decisions."
And those humans, expert or not, also create the cultures within organizations that will need to be changed to accommodate real-time information exchange. The utopian vision of the nimble, collaborative real-time enterprise with information flowing freely and at incredible speeds between departments is inspiring. But it ignores the work that needs to be done to tear down those walls and resolve the ongoing turf wars in most companies.
Gartner/G2 analyst Kraft Bell points out that the real-time enterprise is often mistaken for a business strategy, when it is in fact merely a "competency."
There's no doubt that gaining that competence and having high-speed access to information can make a company more competitive, as long as it chooses the appropriate spots to use real-time applications. Otherwise, it's like buying a Lamborghini to get yourself to the local grocery store and back.
As the economy has contracted, companies have felt the need for more visibility into their supply chains in order to decrease inventories, better track performance and sales opportunities, and have more information with which to make decisions. Implemented carefully, real-time applications can help with all of these business problems.
The key is to avoid unrealistic expectations for real-time applications and to take time to prepare your infrastructure or organization for them. Otherwise, you're just making it easier to make bad decisions more quickly.