It takes a gorilla to drive change

What's more

Age: 60

Family: Married; one son, 19, a college sophomore

Car currently driving: Saab 9-5

Last book read: Memoirs of a Geisha by Arthur Golden

Favorite Web site:

Leisure activities: Skiing, tennis and golf

Jim Champy, the x-engineer

You remember Jim Champy. Or at least you remember business process re-engineering, the movement'or fad, depending on your viewpoint'spawned by his 1993 bestseller, Re-engineering the Corporation: A Manifesto for Business Revolution, co-authored with Michael Hammer.

Champy is chairman of the Cambridge, Mass., consulting practice of Perot Systems Corp. of Dallas. He recently published a new book, X-Engineering the Corporation: Reinventing Your Business in the Digital Age. It describes how re-engineering can expand from a purely internal process to an organization's customers, suppliers, even competitors. Champy said he believes the prescription can apply to government agencies as they implement e-government.

A graduate of the Massachusetts Institute of Technology and Boston College's Law School, Champy was publisher of MIT's Technology Review magazine and, before joining Perot Systems, was chairman of Index Systems. Computer Sciences Corp. later acquired Index, which devised an early automated investment portfolio system.

GCN executive editor Thomas R. Temin interviewed Champy in Washington.

GCN: Do you have to re-engineer before you x-engineer?

CHAMPY: Yes, but not necessarily sequentially. If you have your internal processes lined up, it's easy to see how x-engineering works. It gives you a leg up.

If you missed re-engineering, I wouldn't say spend the next two years re-engineering. Be prepared to fix how you operate as you change processes. This is what I call deep harmonization. It's not just customer relationship management or supply chain management.

I'm arguing that you have to go far beyond that and get to a level of transparency where you open up your processes. Show me your costs, and I'll show you mine.

GCN: Have any government agencies re-engineered?

CHAMPY: Actually, yes. During Vice President Al Gore's National Performance Review there was a lot of effort. They were doing some very good things, but I would describe it as localized. As for re-engineering entire agencies, I didn't see one.

The solution has been to add more people'the traditional Washington approach.
I spent time at the IRS at the request of former Treasury Secretary Larry Summers as an observer. What I experienced at the operational level was that people understood processes and the relationships between process and technology.

At the executive level, it was as if they were from Mars. We would talk to the operational people and seem to make progress, but then at the leadership meetings we'd regress.

Veterans Affairs secretary Anthony Principi knows he has to do real x-engineering across agencies, or he can't serve veterans. His main boundary is with the Defense Department.

GCN: What are the obstacles to re-engineering and x-engineering?

CHAMPY: They require a shift in the way managers think. In the end, managers won't have a choice.

Integration and collaboration never happens unless it is top-down driven'a vice president or president who says, 'This is the way it is going to be.' I'm not sure even Gore understood it. He wasn't ambitious enough.

Government represents the ultimate in fragmentation. The question is whether managers have the appetite to go through what x-engineering requires.

There isn't any single, top-down view of how government should work. Maybe there never will be.

Sometimes people want to foster debate because we don't know the policy truth. But that gets in the way of harmonization, standardization and transparency: three notions of x-engineering. The private sector can build the business case. In government, it gets diverted to a policy debate.

GCN: So should government give up?

CHAMPY: I'm optimistic. We need a gorilla to drive it'perhaps a big agency that starts to do it right and builds some envy.

GCN: Can you give me some idea of the scope of the problems that x-engineering can help avoid?

CHAMPY: I typically find that 40 percent of industry spending is on administrative functions. That metric is about right'about true for Washington. In other words, 60 cents on the dollar is spent delivering service.

The question you have to ask is what value an agency has. That question isn't asked enough. There's no top-line pressure. It isn't bottom-line pressure that's missing. No, no, no, it's top-line pressure that's missing. Agencies always need to ask, what is the value proposition of a process?

GCN: Do you think Mark Forman, who heads the e-government and IT efforts at the Office of Management and Budget, is on the right track?

CHAMPY: I think he is. He's on to the standardization idea, one of the three ingredients of x-engineering. Underlying that is the argument that there's no justification for having a variety of processes or technologies.

GCN: Let's talk for a moment about the proposed Homeland Security Department. With so many disparate pieces supposedly coming together to form a new department, what should the IT people, the process and program owners and the executives do about their systems?

CHAMPY: Creating a homeland security agency out of so many parts is a massive challenge. Someone asked me how the private sector would handle this, and I said that a company in the private sector wouldn't even try it!

But in many ways, government has no choice. These various departments and functions do the work that now must come together. This is an attack on the organizational evils of fragmentation.

I would not begin this work with a focus on process or systems. I would begin by asking, what clients and what purpose must this agency serve? What are its clients' expectations and needs?

Before government designs a process, it must understand the customer pull'in as much detail as possible'that the process must respond to. Then you can design the principal processes of the new agency.

Next, the structure of the agency will have to be designed to reflect these processes. Just smashing the current pieces together'or leaving them as independent entities'won't work. Once the process and structure designs are done, we can design the systems that will handle the massive amounts of data from which this agency must glean information and knowledge. That, in my mind, is the order of things.

GCN: Where is the biggest opportunity for x-engineering: government to government, government to citizen, or government to business?

CHAMPY: It's different for each category. My inclination is that it is for government to individuals. In the end, the enterprise of government must respond to the demands of individuals. Government will learn the most and agencies will improve the most if they focus on the individuals.

But the easiest to pick off is government to business because companies are more likely to be in a position to respond to x-engineering. So government may learn more quickly dealing with business.

GCN: What about DOD?

CHAMPY: They should focus on the individual soldier. 'The Army of One' is a powerful slogan. Imagine the technology infrastructure to support an Army of one. Each agency must ask, who is the customer?

GCN: What should program or technology managers do tomorrow to start improving their performance?

CHAMPY: Start to chart the breakdowns in your organization, in service and capabilities. Be brutally honest about your agency's capabilities. Ask, what is truly distinctive and where are things breaking down?

If you develop a sense of where you are serving customers and where you are failing to meet customer expectations, you're setting up a way to understand real customer pull. I would attack issues I have to attack by building on my uniqueness.

Much of what agencies do is commodity work. That's where Forman is on to something. You could create universal processes. Will people let go of their processes? It will be a big battle.

Stay Connected

Sign up for our newsletter.

I agree to this site's Privacy Policy.