There was a very interesting discussion in a work management study group a few years ago. Several members reported that industry analysts were describing the future of work management in terms of embedded workflow within business applications. Certainly those analysts saw a trend, and were accurately reporting it - many applications today are incorporating a "thin" embedded work management system.
After discussion, the study group concluded that
Upon further discussion, it appeared that several of the study group members had experiences similar to mine.
Late in 1984 or 1985, when we were testing one of the first production work management systems, we measured average turn-around time for processing insurance policy applications and changes. We were devastated to find the time using image and work management was longer than the time of the teams still using paper. This couldn't be!
Several members of my development team and I spent days standing behind the image and work management system users, to figure out what was wrong. The typical user would start with the image driven by a work manager (of course, since we were watching), and process the high priority work. After each item was completed, they would go to the next, and the next, until all the image work was done - even the lowest priority work, perhaps a day later. Suddenly they would remember their other work drivers - the paper in-box, the electronic mail, the system managed to-do list, the voice mail, and so forth. After some unrepeatable comment, they would start on their paper in-basket. By now, the rush work waiting in their paper in-basket was almost a day old, and all the high priority work might not be done today. So they struggled with the paper through the remainder of the day, and part of the next day, until all the paper was processed. Another expletive, and they would start on the system diary - their electronic to-do list. Of course, it had been a day and a half since they did any of that work, so deadlines were missed and urgent work was late. Repentant, they again completed all that work. Until they checked their E-mail, Voice Mail, etc. It was several days before they finally got back to the image system, and by now, as you expected, the high priority work on the image system was late.
What my team concluded in 1985, and the study group also concluded, was that people don't work well when driven by multiple workflow systems. What they need is an enterprise workflow system, that can prioritize work across many applications.
This is a minor issue for a data entry operator, a customer service representative, or a payroll clerk. They typically deal with homogeneous work from a limited number of sources. On the other hand, the toughest case I have seen is the line-of-business supervisor. They are driven by the widest variety of business cases (the hard decisions or the toughest customers). They are also driven by administrative workflows - personnel requisitions, supply orders, travel requests, payroll administration. They are subject to more than the usual amount of electronic mail and voice mail. They participate in ad-hoc work flows (e.g. "please review the revised procedure by Thursday"). What tool can help them balance their work?
The ideal solution is an enterprise work management system. One that supports all their systems, not just a single application. Available work management products are not perfect, but they do a reasonable job managing (and prioritizing and coordinating and tracking) the production work throughout the enterprise. Many support electronic forms, e-mail, and other sources of work. This is a much better solution that a separate workflow system embedded in each business application.
Of course, there will undoubtedly be companies which have multiple different work management products - perhaps due to the growth of independent departmental systems or business mergers and acquisitions. Therefore the ideal business application will be able to interface to multiple work management products. Workflow systems will need to interface to each other. The international Workflow Management Coalition (WfMC) has defined, and is continuing to refine, the interfaces to allow applications to interface to workflow systems, to allow workflow systems to talk to each other, etc. Not all vendors, even those vendors participating in the WfMC, conform to the standards, but even without formal adherence to the standards, the work of the coalition provides guidance to those interfaces.
What would be the ideal solution, given today's situation? An established work management system coupled with the key business systems. Coupled, so that they can be used together with minimal effort - so that it is easy to install both business and work management systems together. Coupled but not integrated. If they are totally integrated, then a company with the "wrong" work management system cannot choose a particular business processing system, or vice versa. How should they be coupled? Ideally by some industry standard. But lacking widespread acceptance of a standard, at least with interfaces that can be adapted in each environment.
Perhaps there should be a fourth option in the list at the beginning of this paper:
Back to the home page
Send e-mail comments to
©2002 by Charles A. Plesums, Austin, Texas USA. ALL RIGHTS RESERVED. You may license additional copies of this document through a nominal royalty payment as specified on www.plesums.com.