Kelman: Past performance revisited

At a conference I attended recently, a participant complained that vendors' past-performance evaluations look alike, making it hard for them to be a differentiator in source selection. Government agencies were putting considerable time into completing those evaluations, and the participant feared they weren't getting a good bang for the buck. Evaluating past performance is important, he concluded, but agencies weren't doing a good enough job.

I don't have a good feel for how true this observation is. I suspect offices vary in how hard they work to ensure honest judgments in past-performance report cards. I also suspect past performance is most effective when used informally — for example, in deciding which vendors to ask for bids on General Services Administration schedule orders or in evaluating those bids.

But the conference participant's complaint highlights a challenge for educating acquisition workers. Many procurement leaders, notably Deidre Lee, who recently joined GSA, have argued for years that contracting officers need to see themselves positively as business advisers to program customers, helping programs serve agencies and taxpayers rather than playing their traditional role of compliance police.

Supporting past-performance reports and evaluations is a prime example of the advisory role contracting officers can play. If we're failing here, then we're coming up short on the ambition to expand contracting's contribution to government.

So, people in charge of training acquisition workers — David Safavian, administrator of the Office of Federal Procurement Policy, and Frank Anderson, president of the Defense Acquisition University, for example — should make past-performance training into a model and a template for business adviser training.

Such training must be practical rather than a regurgitation of Federal Acquisition Regulation text. It should discuss how intelligent study of past performance can help the government get better value. Training should include a discussion of good practices for incorporating past performance in requests for proposals.

For example, I have never understood why the government allows bidders to cherry-pick references, almost ensuring that past performance won't be a discriminator — everybody can find somebody who likes them. Instead, the government should use standard criteria, such as the last five completed contracts in a similar area with a similar dollar value, for determining which references will be evaluated.

And there should be extensive, practical training about writing good pastperformance report cards, including discussion and role playing of the psychology of delivering bad news, along with other techniques to encourage differentiation.

Attention to past performance has been one of the most positive changes in the procurement system in the past decade. Let's train the workforce right, so we can get the use of past performance right.

Kelman is a professor of public management at Harvard University's Kennedy School and former administrator of the Office of Federal Procurement Policy. He can be reached at steve_kelman@harvard.edu.