Commerce procurement reinvention shows the way

The Commerce Department recently conducted its third annual joint training conference for information technology and procurement professionals. The conference, which was held in Williamsburg, Va., illustrates how times have changed. A decade ago, many government IT professionals would rather have spent time with their dentist than with their counterparts in the procurement profession.

The Commerce Department recently conducted its third annual joint training conference for information technology and procurement professionals. The conference, which was held in Williamsburg, Va., illustrates how times have changed. A decade ago, many government IT professionals would rather have spent time with their dentist than with their counterparts in the procurement profession. Now procurement professionals agree that such collaboration makes real sense and is the birthplace of true government reinvention. One of the topics discussed at the meeting was the ongoing re-engineering of Commerce's process for major system IT acquisitions, originally known under the name ConOps, or Concept of Operations. Individual Commerce bureaus have since adopted their own names for the innovation, such as ASAP (A Streamlined Acquisition Process) at the Census Bureau. ConOps began in l995 as a business process re-engineering effort initiated by Commerce's then-procurement executive and out-front reinvention advocate Shirl Kinney.

The basic structural change proposed under ConOps was something that had been tried for several years at the Defense Department: getting rid of the sequential review and signoff process for requirements documents, requests for proposals and source selection and replacing it with an empowered cross-functional team including IT, procurement, legal, and finance folks. The old stovepipes— still all too common in government— are a recipe for delay and ineffectiveness. Most of the time that it takes to get paper processed in a stovepiped world is "in-box" time, with documents waiting on people's desks. Separate reviews encourage adversarial relations among functional specialists, for whom "add value" comes to mean "raise objections." The reviews also produce a situation in which nobody really has responsibility for a project's success.

Beyond moving to teams, ConOps also sought to institutionalize early vendor involvement and the use of simplified performance-based specs (a statement of need), which would be developed with significant vendor input rather than a more detailed statement of work. Commerce decided to pilot test ConOps in six procurements, three at the Patent and Trademark Office and three at Census. Some were for commercial off-the-shelf hardware but most were for IT services.

A year ago, Commerce issued an evaluation of the pilots. It found that cycle time was way down— ranging from five months instead of three years for a data warehouse development contract to eight months instead of 11 months on a facilities management contract.

Other survey results showed that the teaming concept worked well, but it needed fine-tuning. Team members wanted just-in-time training as they needed it rather than concentrating training at the beginning of the process. There were concerns about how team assignments fit in with a team member's other duties, and some procurement folks fretted about a loss of authority in the team.

In Williamsburg, those experienced with ConOps spent time trying to persuade those from other Commerce bureaus that they also should try it. There was a joint presentation by Julie Hart and Wayman Baker, procurement and IT professionals at the National Oceanographic and Atmospheric Administration, about a supercomputer replacement buy using ConOps, the first such effort at NOAA. The two gave an upbeat presentation about their first foray into this new way of doing business.

"Empowering the team and including the appropriate people fostered efficiency by allowing us to create and approve our own work. We didn't have to formally create documents such as an acquisition plan or an agency procurement request, which are traditionally passed slowly back and forth through channels. Instead, we performed the actions required by those plans and obtained the necessary approvals by including reviewers on our team," they said.

Hart and Baker praised the use of a performance-based statement of needs and the instruction to vendors to bid a working solution rather than a contract line-item number structure that may or may not meet the government's needs. They discussed the open communication with vendors that preceded the RFP, and they spoke highly of the streamlined process, featuring oral presentations and fewer evaluation factors.

Hart, the contracting officer, impressed the crowd with her understanding of "techie" issues— being part of a team taught her IT knowledge she otherwise would never have gained and helped her help the team with the buy. One lesson she learned, Hart said, was the need to include department-level budget people on the team; the group included NOAA budget folks but no one from Commerce, and that produced a five-month hitch (which is now resolved) in getting funding for a site to house the new supercomputer.

This is how reinventing government happens. People on the front lines try new ideas, they present their experiences, others learn from them, and gradually, islands of innovation can turn into continents. That's the process in which Commerce IT and procurement leaders are currently engaged.

Kelman was the administrator of the Office of Federal Procurement Policy from 1993 to 1997. He is now Weatherhead Professor of Public Management at Harvard's Kennedy School of Government.