4 steps toward better collaboration in government

Mark Forman is leader of the Federal Performance and Technology Advisory Services Practice at KPMG. This article represents the views of the author only and does not necessarily represent the views or professional advice of KPMG.

Anyone who follows how government agencies use technology knows we’re in the middle of a monumental shift. Social networking tools allow people to collaborate by rapidly bringing together a range of expertise and data to tackle today’s most pressing problems. That’s a big difference from the way things were in the 20th century, when government agencies took years to analyze policy issues and implement new programs.

It’s the budget season, which always revives questions about the effectiveness of government and how to improve it. So how could government agencies take advantage of collaboration tools to solve today’s problems? Here are four broad steps to consider.

1. Human judgment. Separate your agency’s transaction-based processes, such as writing purchase orders, from knowledge-based processes that involve judgment and experience, such as awarding grants. This kind of analysis is required to identify the knowledge processes that would most benefit from collaboration. During the past 30 years, agencies have boosted productivity by automating business processes but have rarely differentiated between a knowledge process and a transaction process. Automating knowledge processes hurts an agency’s ability to serve the public because computers can’t make human judgments, which are often critical to meeting a policy goal. CIOs need to recognize where knowledge processes have been inappropriately automated.
   
2. Data structure.  Work toward a new set of enterprise architecture reference models and tools to make data consistent and therefore usable across the enterprise. Earl Devaney, chairman of the Recovery Accountability and Transparency Board, best articulated that need when he told a Senate committee a few weeks ago that each of the 28 agencies that received federal grants under the economic stimulus program organized and tracked those grants differently, using unique systems and terminology. Oversight suffered as a result because employees who monitor for fraud and waste documents must check that information by hand. “While this may not sound like a big deal, it is,” Devaney told the committee. I agree. If government wants to effectively collaborate to solve problems and serve people better, we must unify data.
3. Identity Management. Establish policies to promote collaboration. Although we don’t need to rip out what’s in place, we do need to build on Homeland Security Presidential Directive 12 so workers can collaborate in teams across agencies. We’ve organized identity management for automation, not collaboration, and we need to shift to the latter. Technology is available to federate the government’s ID system. A governmentwide identity system would lead to role-based access control, thus facilitating virtual team building for problem solving.
4. Organizational change management.  Remember these three words. Agencies must be willing to surrender control over their policy prerogatives and become part of a governmentwide solution. Change management involves a willingness to accept other people’s data, including data from other agencies and external experts, in the interest of developing the best solution. No single organization can be expected to have all the knowledge needed to make the best policy decisions to solve complex problems. Imagine how much better the response to Hurricane Katrina could have been had agencies used modern collaboration tools to share knowledge and data. Driving change of that magnitude will take sustained involvement on the part of IT managers and high-level leaders.

At end of the day, what’s needed is governmentwide organizational and technology change management that rewards collaboration.