Remember your high-school algebra teacher’s advice? Break down the problem into smaller parts. And when you tried it, the algebra changed from a complex web of Xs and Ys into several simpler problems.
Federal CIO Vivek Kundra and his boss, Chief Performance Officer Jeffery Zients, are giving the same kind of advice to agency leaders. They want agencies to stop building complex, multiyear, multibillion-dollar overhauls of major IT systems and infrastructure. Instead, they say, agencies should start taking much smaller bites.
“For too long, we have witnessed runaway projects that waste billions of dollars and are years behind schedule,” Kundra told the CIO Council in September. “By the time these projects launch — if they launch at all — they are obsolete.” He recounted the government’s history of failed IT projects, which stretches back to 1968.
When a problem has been around for decades, it means a change in thinking will be tough. Kundra maintains that agencies are stuck in think-big mode. Officials often set course on a far-reaching project rather than accomplishing essential needs. In too many instances, agencies are left with substantial cost overruns, lengthy delays in planned deployments and poor functionality in the fast-moving IT world, he said.
He’s arguing for an upheaval of the classic procurement thinking, especially in the technology world.
For example, the way the Defense Department buys its land vehicles and fighter airplanes does not fit with IT projects. In instances such as the Hummer or F-35, DOD officials set out all the specific requirements from the start. The department wouldn’t want to add requirements after production began because it would cost a lot of money to interrupt the production line.
But that approach doesn’t work as well for DOD’s IT acquisition projects. About one-third of the department’s IT projects get canceled before completion, according to reports. “You buy the same old-fashioned way, you get the same old-fashioned results,” said Jeremy Grant, chief development officer at ASI Government, formerly Acquisition Solutions.
These days, the government buys so much IT and services that agencies must buy differently. IT changes so fast that it requires a fast procurement process to keep in step. By scaling back programs, agencies will realize savings and perhaps more successes, Kundra said.
Apple’s iPhone is a good example. Apple outlined an application programming interface for the phone, even without fully testing all aspects. The developers changed various aspects in subsequent versions of the API. With each version, software makers built their ideas on the platform.
Talking About a Revolution
To be sure, smaller procurements might not save the procurement world from buying problems. There’s an underlying procurement problem even when officials buy with small bites in mind, experts say. Put another way, they’ve also got to be able to do the simple math.
The acquisition workforce is overwhelmed, said Larry Allen, president of the Coalition for Government Procurement. “The approach sounds really good on acquisition letterhead, but in reality, it has a lot of issues,” he added.
Whether big or small, the procurement process takes time. Mega contracts can force contracting officers, program managers and others to take months or even a year to develop requirements so the project goes smoothly. But smaller procurements also take time, Allen said.
Contracting officers will have more contracts to award. Program managers will have more to manage. Then the contracting officer’s technical representative, who oversees the progress of an awarded contract, will have more to watch. Many small procurements will add to the pressure on COTRs, for whom the COTR role is often only part of their duties.
Agencies also face more transitions between contractors as the project advances. There’s the ever-present risk of bid protests with each contract award. If a new contractor comes aboard, the changeover will delay the project because the new awardee can’t begin work the next day. As the old and new contractors switch places, there’s the potential for failures in continuity. Agencies will need to get companies to talk to each other in a transition, or problems are inevitable.
“It becomes like a game of telephone,” Allen said. “Some things will get lost in translation.”
Allen warned that a company might not put everything it has into a project if there is a potential for it to lose its place in future parts of the project. A contractor might keep back some of its ideas because it might later have to hand over trade secrets or strategies to a competitor.
Those are the kinds of algebra problems that defy easy solutions and short cuts. Perhaps that’s why Kundra is also trying to raise the stakes for agency leaders by putting his message into a much larger context of change.
“We are in the midst of the Information Revolution,” Kundra told the CIO Council, so the government must change as well, he said, “creating a culture of accountability with a focus on execution.”
— by Matthew Weigelt – Oct. 21, 2010 – Federal Computer Week.