Contract management: Glass half empty or half full?

I recently came across a report from the Government Accountability Office on Defense Department services contracting ("Further Actions Need to Address Weaknesses in DOD's Management of Professional and Management Support Contracts"). The report is based on a review of a random sample of 64 task orders from five large management support contracts from 2004 to 2007. There is no doubt that we need to make significant progress in contract management in the government. But actually -- I'm guessing contrary to the intention of the study's authors -- I came away from reading the report feeling more optimistic about the state of contract management than before I had read it.

The report discusses use of performance metrics in the task orders. Sixty of the 64 included performance metrics for quality and 54 for cost and schedule performance. The report criticizes the fact that 40 of the 60 task orders used subjective quality metrics such as customer satisfaction and number of complaints.

Well, my reaction is that even doing that is a pretty decent first step, especially for the staff augmentation task orders included in the sample, which (though people don't like to say this) often involve contractor personnel simply doing various tasks as assigned to supplement the work of civil servants. The report suggests that contract management personnel frequently did make phone calls proactively to inquire about satisfaction with the contractor's work. Furthermore, about one-third of the task orders did have objective quality metrics.

(And sometimes the report picked nits -- for example, mentioning a case in which the contract called for a reduction of the government's unliquidated obligations by 25 percent in six months, and 50 percent in a year but criticizing the contract office for not including the metrics specifically as performance metrics for the task order, although government personnel indicated they had measured contractor performance against that standard.)

The report also found that every one of the 64 task orders had a person assigned to manage the order, and that 61 of those people had received at least some (two hours to a week!) training in task order management. This is better than I would have expected. Clearly, simply assigning people isn't enough (program offices for major weapons systems are filled with people, and of course the Defense Department still has problems -- though developing weapons systems is far more complex and difficult than these task orders). But to me the report suggests a decent baseline from which to build.