the lectern banner

By Steve Kelman

Blog archive

Improving statements of work to improve contract management

In my early-October blog post, "To contract better, does the government need more in-house experts?" I promised I would continue to return to issues of improving contract management in government, with the hope of inspiring a dialogue and identifying improvements in practice. I am hereby starting to make good on my promise by reporting on a conversation I recently had with the founder and CEO (also an old friend) of a 75-person company that consults on strategy, IT and other topics to one large federal agency. She had asked to talk with me about some ideas she had for improving contract management.

I asked her at the beginning of our conversation her overall view of the state of contract management in her agency. "Terrible," she said. (More nuances surfaced later -- there were instances of good contract management, though they were not typical or numerous.)

My friend's ideas for improvement centered around writing and managing to statements of work (SOWs), which she felt put a contract on a good path for performance or a bad one. Statements of work themselves were often poor in her experience, written by program staff who were often talented, but overcommitted time-wise and neither trained nor invested in writing good requirements.

In most cases, what happened afterwards made the problem worse. The SOW writer ceased being associated with the contract and quickly moved on to something else (which hardly created an incentive to write a good SOW in the first place).

The SOW was passed on to a Contracting Officer's Representative, seldom a subject-matter expert and typically unenthusiastic about the idea of being a COR in the first place. Often, she noted -- and this was a shocker to me -- the agency would start the kickoff meeting for a new contract by asking the contractor, "So what are you guys supposed to be doing?"

The contracting people, she found, were generally spread too thin to be able to spend any time on contract management unless there was a crisis.

This is not popular to say, and it may go so far as to be a non-starter, but at some point I believe the government must deal with the fact that it is extremely difficult to have good contract management -- save for exceptional instances that rely civil servant heroes -- without putting more bodies into it.

It's hard to know where to start doing something about this dysfunction. Zeroing in on SOWs, my CEO friend had one interesting suggestion: Why not use the Federal Acquisition Regulation's advisory downselect procedure for important contracts to have a small number of finalists identified at the beginning?

Then get the government program person and contractor representatives involved in a collaborative effort to develop the SOW together. Immediately, there would be a major infusion of subject matter expertise, coming from the contractors, into the process. And having several contractors work on the joint SOW would guard against rigging the specs in favor of any one contractor's solution.

Would contractors be willing to invest time in this, I asked? Not for all contracts, certainly, but yes for contracts being bid through an agency strategic sourcing vehicle (or a governmentwide vehicle the agency was using as a preferred source). Those are vehicles in which contractors are deeply invested and care a great deal about their performance.

The CEO's comment about lack of COR commitment was depressing. She did advocate a cradle-to-grave system where the same people involved early in the cycle stayed involved post-award. But she felt getting good people to step up as CORs was a challenge.

I asked whether it would be a good idea to break up the COR function so that one person did the more administrative elements (such as voucher processing or handling of government-furnished equipment), while a subject-matter expert did the strategic elements of contract management. She did say that for about half the contracts, there was a quarterly "strategic review" of how the contract was progressing -- one of the few positive pieces of news she reported during our whole conversation.

I close this post with two suggestions to blog readers. One is that people join the dialogue by posting reactions to these suggestions on the blog comment space below. The second is that anyone -- be you government employees (contracting or program), industry, academia, and even non-American blog readers (I know this blog has non-U.S. readers, and many of whose countries face similar contracting challenges) -- with ideas for how to improve government contract management should contact me about setting up an interview to talk over the phone. I can be reach via email at

Posted by Steve Kelman on Nov 20, 2015 at 5:46 AM

Cyber. Covered.

Government Cyber Insider tracks the technologies, policies, threats and emerging solutions that shape the cybersecurity landscape.


Reader comments

Wed, Dec 9, 2015 Gorvan Metrov.

Re Mr Williston's 24 Nov comment. You lost me with splitting the COR function. You want one piece--"the PM"--to handle "strategic vision." Sir, the USG rarely buys anything requiring strategic vision. We buy mainly simple things and services. No strategy required. Just careful, workmanlike execution. The more we build up the high fallutin' attributes of what we buy, the more we waste $$$ and time.

Sun, Dec 6, 2015

Surely the statement of work, drafting process, could be improved; however, why is it that we spend more time in dialogue versus actioning what we already know? A great way to improve is simply practicing the "known" truths. We acknowledge limited resources, but keep asking for more ideas as if we have the resources to vet and implement them. Take the great ideas we already know about and do something with them. Once you do that, the real thinkers who have devoted significant amounts of time creating solutions and offering suggestions will be more prone to invest more energy. We are wise not to cast pearls to swine that don't appreciate or act upon the revelations already given. As long as the answers involve things such as using DAU's Services Acquisition Mall ARRT, or this unrealistic contractor involvement into an already bloated procurement acquisition lead time, we will never truly improve government contracting. The single easiest and most innovative thing we can do to improve the system is to empower contracting officers to do what it takes to award contracts within y the rules of the FAR. Nothing more, nothing less. If it's not prohibited, it's allowable...

Wed, Dec 2, 2015

I couldn't agree more; and I have always felt it to be the case that contracts don't yield the expected result because no one seems to know how to accurately articulate what they want the Contractor to do. Just got done submitting 47 Offeror questions about a nine-page RFQ SOW against which the Government wanted a quote back in 7 days. The SOW was an MS Word Search and Replace of "similar" work performed on another base 2700 miles away. Not only was the actual scope of work contradictory, but all the "bookend" requirements around it were not appropriate for, or applicable to, the base on which the work will be performed. Plus, some of the SOW requirements were also in direct conflict with parent IDIQ contract requirements.

Tue, Dec 1, 2015 Al

I feel for our project managers. Writing SOWs can be very difficult. If writing a SOW is too difficult, it can be a sign that either the project itself is a bad idea and should not happen, or it should be procured in smaller increments. Or should be performed in-house. I think one of the reasons the government's procurement system has such high transaction costs is that its competitor- the government's HR system, also has high transaction costs. Sure, there is still some competition, but it's like watching the two worst teams in the league play a playoff series.

Tue, Nov 24, 2015 John Inman WIlliston, Vermont

You wrote, "I asked whether it would be a good idea to break up the COR function so that one person did the more administrative elements (such as voucher processing or handling of government-furnished equipment), while a subject-matter expert did the strategic elements of contract management." -- I have always to have both an effective COR as well as an effective program manager on any major service or development contract -- the COR for day-to-day business, and the PM for the strategic vision and continuity.

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above

More from 1105 Public Sector Media Group