Looking askance 3
Looking askance 3
January 2004
1. Process software compared
I am working on a book on BPM, aimed at general rather than technical managers. It will, I hope, help them make informed decisions over whether and how BPM is applied in their organizations.
The book necessarily deals with product matters. To help clarify the differences between the main types of 'white collar' process management software, I have compiled this table. It compares electronic document management (EDM), workflow automation ("workflow") and business process management (BPM) products. The shaded cells are where BPM software distinguishes itself from the two other kinds of software.
I have tried to be fair in this. It is easy in such exercises to exaggerate distinctions, as almost any supplier's literature makes clear. I have tried not to fall into that trap or to create straw men in my summaries of the earlier products' abilities. EDM and workflow still have their uses and, indeed, may be a more cost-effective option in some situations.
I'd be glad of constructive comments, to rgw@office-futures.com please. [The version below is based on the comments received.]
Ability
EDM
Workflow
BPM
Data handling
Makes electronic versions of paper documents?
Yes
Uses EDM for this
Uses EDM for this
Handles all kinds of digital document (e.g. word processor files)?
Yes
Yes
Yes
Indexes and searches digital documents?
Yes
Uses EDM for this
Uses EDM for this
Allows the reuse of documents, and the disassembly and recombination of their components?
Yes
Uses EDM for this
Uses EDM for this
Allows users exclusive access to master documents (check in/out)?
Yes
Uses EDM for this
Uses EDM for this
Links digital documents with other data?
Yes
Yes
Yes
Exchanges data with other software?
Yes
Yes
Yes
Routing
Routes digital documents around a network?
Yes
Yes
Yes
Manages the speed and direction of the route?
Sometimes
Yes
Yes
Sets up rules for managing routes, timing, actors, etc.?
Sometimes
Yes
Yes
Runs processes outside the organization?
No
Sometimes
Yes
Process management
Modifies processes on the fly?
No or seldom
Yes
Yes
Manages human activities?
Yes
Yes
Yes
Invokes the running of other software
No
Yes
Also integrates and interacts with other software
Manages an organization's other software (orchestration)?
No
No
Yes
Uses, and is used by, Web services?
No
No
Yes
Offers goal-directed processing?
No
No
Sometimes
Offers predictive processing?
No
No
Sometimes
Process design
Graphically models processes and creates the rules to run them?
Sometimes
Yes
Yes
Simulates and optimizes processes?
No
Sometimes
Yes
Manages the rules from application programs separately?
No
No
Yes
Allows analysts and users to add or modify rules?
Sometimes
Sometimes
Yes
Reporting
Provides after-the-fact reports and audit trails?
Sometimes
Yes
Yes
Provides real-time reports on progress and performance?
No
Yes
Yes. Also provides real-time reports on progress in all connected application programs (business activity monitoring or BAM)
Product design
Open programming interfaces?
Usually
Yes
Yes
Standards based?
Sometimes
Yes, usually
Yes, always
2. Where did human beings go?
British psychology is typically experimental psychology. One could read quite an appreciable portion of its literature without being acutely aware that it had anything to do with living beings at all. (J.A.C. Brown, Freud and the Post-Freudians, Penguin, 1964)
The same thought often crosses my mind when reading about business. For example, here is Judy Johnson, Sapiens America's vice-president of "Insurance Strategies":
It's not just the old systems, it's everything that's wrapped around them, including old processes, old work-arounds, and the white spaces where you have to stick clerks into the process because the systems are incomplete. (From "Insurance Industry IT Needs Significant Housecleaning", CIO, 28 July 2003)
I'd never before thought of people as a cross between Polyfilla and fuse wire.
Or take this extract from the blurb for a conference organized last year by the Balanced Scorecard Collaborative. In it, Cassandra Frangos, its Human Capital Practice Leader, says:
Today's most successful companies have realized that human capital represents the only real sustainable competitive advantage. Organizations that use the Balanced Scorecard to align their human capital with their strategy have achieved both bottom line and strategic results. Learn how these organizations use the Balanced Scorecard to assess the readiness of their human capital to execute their strategy and create a workforce that drives strategic success.
Could it be that it's written into Ms Frangos's job description not to say "people"?
Much the same applies here. This is Wendy Tan, the CEO of Moonshine, a failing dotcom, being interviewed in The Guardian a couple of years ago: "It was hard making positions redundant, while recruiting new skill sets." That's OK then - the organization chart redrawn and a few jobs reprofiled. No human implications at all.
Finally, here is an example from the ebPML.org Web site (see http://www.ebpml.org/pi-calculus.htm ).
At a high level, a company can be considered to be a very large automaton whose logical state consists of gigabytes or terabytes of data, and physical state is made of the raw materials, manufactured goods, people and money under its control.
This is at so high a level that it's almost out of sight. Only machine-readable data counts, it seems, and people are defined solely by their physical presence. This is to define them out of existence. Perhaps the unnamed author is making a debating point; I hope he is.
The difficulty is when people mistake such a map for the territory. It is so easy to think that the model you or someone else has created is as detailed and multi-dimensional as reality. It is not. Real life has a depth, complexity, variability, unpredictability and subtlety that no map or model can reproduce. If it could, then it would no longer be a model or a map but a separate reality. Such things occur only in fiction or in the minds of philosophers.
And what kind of system do they work in?
Unfortunately, experimental verification of the second law [of thermodynamics] is not practicable, since it requires dismantling an organism to its component molecules, which would result in its irreversible death. (Donald Voet, Fundamentals of Biochemistry, John Wiley, 1998)
More than a few books on process management fall into a similar trap. One I recently read was fundamentally an engineering text. (I shall save the author's blushes by not naming him. He's done me a favour or two!) It gives copious and well-illustrated detail on the most popular process methods and taxonomies. But, while doing so, it manages to ignore such matters as the influence of viewpoint, different organizational models, laws, regulations, international working, marketing (which is not even mentioned in the index), location, staffing plans, salaries, and culture and other aspects of human behaviour. In short, it is lifeless and hermetically sealed from a wider reality. Other ways of looking at organizations, of defining systems or solving problems are simply wished away.
Even some of the so-called classics are similarly lopsided. There is one entry for "marketing" in the index to Hammer and Champy's Reengineering the Corporation: A Manifesto for Business Revolution (2001 edition). There is none at all in Rummler and Brache's Improving Performance: Managing the White Space on the Organization Chart.
In all three books, the emphasis seems to go from the inside out rather than the other way. It is technology push rather than business demand pull. This is contrary to the experience of many managers, even middle managers. Their lives are concerned with the outside world - purchasing, logistics, marketing, sales, customer relations and so on. Despite this, marketing-based views of process management are hard to find.
I don't know why this should be. It is not hard to do and seems a natural way to think about organizations' internal workings. This is particularly so for companies that are or profess to be, in the jargon, marketing-led and customer-focussed. Even where an organization sees itself differently, its end to end processes must touch a supplier or a raw material at one end and a customer or trading partner at the other. Since it is the latter group that provides a trading organization's income, it makes sense to give priority to their requirements. An outside-in view does this.
And what system does that system work in?
Every company has a diagram of the universe in which they're the centre. That's never true. We're all a node in a mesh. Intel's customers are customers of each other, our suppliers are customers of each other... We all circle back to buy things from each other. It's an extraordinarily interdependent ecosystem we're a part of. (Douglas Busch, CIO of Intel Corp, quoted in "Era Of The E-Business Ecosystem", Datamation, 23 April 2002)
Here is a map I drew some years ago to depict what I saw as the main relationships in electronic commerce. (Yes, it was that long ago.)
Douglas Busch has put into words what I tried to describe pictorially. The basic four-blob map multiples and ramifies endlessly. This is, in a sense, also a picture of the virtual organization.
This also now seems to me to depict one of the objectives for business process management (except, perhaps, in the consumer-to-consumer link). BPM allows us treat the organization as a rapidly responding network of resources and actors. This network extends beyond the company's legal boundaries to suppliers, trading partners and customers. These participants can be anywhere in the world at any particular moment, using both fixed and mobile communications systems to connect to one another. Those systems are, these days, mostly reliable, efficient and pervasive. They are also cheaper than they used to be.
3. Coder Coda
Beware of bugs in the above code; I have only proved it correct, not tried it. (Donald Knuth, author of the TeX typesetting system and of The Art of Computer Programming.)
Roger Whitehead