Ken Orr is the founder and chief scientist for The Ken Orr Institute, a business technology research organization. He is an internationally known and recognized expert on technology transfer, software engineering, information architecture, and data warehousing.
According to Orr, improving business processes is essential for all enterprises. BP are not that hard to conceive of, but they are very hard to manage in the real world and especially hard for big organizations. The trick is to create truly executable business models so that BP can be made to work. BP need to be integrated with the rest of the enterprise architectures. Good models are important, because they help people visualize new directions and procedures. Orr said having a one-page visual model is the best way to communicate this information. Visual models are important. They have to describe two concepts: the business processes, which is the understanding of what is needed from a business standpoint, and the Enterprise Architecture, which is what now exists in IT. Collaboration between the two gives the understanding of how to deliver the results.
Orr’s favorite quote on this is from Karl-Eric Sveiby, “Trust is the bandwidth of communications.” This is important, because strategic planning, enterprise architecture, business process and collaboration all come together, and communication is the glue that makes it work. Orr said some companies are outsourcing their knowledge, and this leaves them with the inability to actually understand their own businesses. He likened business to a baseball team, and said you really need to have a farm team, a group of young people coming up and learning the business and this isn’t happening in many places.
Orr listed five models:
- Benson and Parker’s Square Wheel
- Porter’s Business Value Chain
- Rummler and Brache’s Enterprise Feedback Model
- Nolan’s S-curves
- Orr’s Information Technology Architecture
Orr said that the Information Technology Model developed in the late 90s to describe technology architecture has proved useful in describing the complexity of technology options to management. Orr showed several dozen slides of how these models look.
Enterprise Architecture, the IT side of an organization, is always expanding. There are hundreds of applications with thousands of databases using more than 250 different technologies with more on the way all the time. And the older stuff never goes away completely. So changes in EA are difficult and expensive, and take a long time to implement.
According to Orr, collaboration is the key to what he calls, “Liberation Technology.” IT has to provide technology everywhere to everyone. When it is done right, it is indeed liberating.
Turning all of this into business processes that truly make the work easier, faster, and with more and better information is the promise of BPM. The reason it is difficult for large organizations is that BP is not intuitive and runs against existing organizational bias. Yet, it is still the key to business improvement, quality improvement, and leveraging technology. It is necessary for managers to know, in a detailed way, how their companies develop, make, and sell products.
The basics are fairly simple. There is a main process and there are inputs, outputs, and outcomes to this process. In order to clean and simplify what the organization does, it is necessary to get a good picture of the mainline process and how everything feeds into and out of it. Once the picture is complete, the organization can begin to improve it. Business processes and enterprise architecture cut across organizational units. This is their nature, according to Orr, and each is dependent on the other.
A common mistake is trying to automate existing processes without completely understanding how they fit into the main process. Another is nano-management, which works against seeing the big picture.
The new generation of applications includes process model-driven architecture and data-centric rather than document-centric applications. Orr predicts that business process will drive more and more initiatives and more industries will follow SCOR and eTOM models. Organizations will be purchasing processes rather than software suites. The difficult thing will still be to tie workflow to the data.
