When Frank Kendall, the Undersecretary of Defense for Acquisition, Technology, and Logistics, submitted a legislative proposal to the House Armed Services Committee this past January, its theme was simplification. His proposal aimed to improve the way the Pentagon builds and buys new technologies, by simplifying the labyrinth of rules, clarifying oversight and decision-making roles, and reducing the amount of documentation required for these acquisition programs. Echoing the sentiments of countless practitioners and observers before him, Kendall told Congress that the way the Pentagon currently manages contracts is “still too complicated and burdensome.”
Kendall’s proposal is long overdue. Nobody likes the complexity of the defense acquisition system. Nobody can justify it. But how did the military’s technology development business get so tangled up in the first place? As I assess the problem, excessive complexity is often rooted in a lack of clear priorities. When we can’t tell which pieces of information or which aspects of our design matter, we try to include everything for fear of missing something important. The result—we miss almost everything.
I make this argument in my book, The Simplicity Cycle, and its applications are visible throughout the defense acquisition process. In the absence of clear judgment about the value of various activities, the illusion persists that everything is essential. Nothing is unimportant. The result is the brain-melting complexity Kendall’s proposal aims to reduce.
As the complexity of any process increases, the likelihood of missing an important step also increases simply because there are so many different steps.
Today’s acquisition process is the result of generations of legislators, Pentagon officials, and military leaders whose preferred problem-solving method has been to build new layers of complexity atop a process that is already overburdened by bureaucracy. For instance, a managerial misstep on a single project triggers a new, permanent level of oversight for the entire enterprise. In an attempt to prevent the repetition of an engineering mistake on a single project, senior leaders end up creating an entirely new technical review process to which every subsequent project must adhere. Ironically, the resulting complications tend to cause the very problems they were designed to solve.
As the complexity of any process increases, the likelihood of missing an important step also increases simply because there are so many different steps. Dr. Atul Gawande points out in his book The Checklist Manifesto that many of the difficulties in modern medicine are the result of “the complexity that science has dropped upon us.” The same dynamic occurs in military technology projects. It’s easy (and human nature) for anyone to overlook a step when working through a detailed process, but the sheer volume the steps in defense acquisition adds the difficulty of trying to determine which steps are truly essential and which are less so. Without clear priorities, we are more susceptible to missing the steps that might actually matter.
There is such a thing as too much detail, and in a similar way, long documents that try to include every possible detail are likely to obscure important information—they hide essential data amidst the trivial. When we produce a large stack of long documents, this problem is compounded further. Eventually, nobody reads any of it.
Complicated processes and organizations also inhibit quick problem-solving, because they make it harder for anyone to find the right person in the right office at the right time or to have anything signed off on in an efficient manner. For example, Colonel Thomas Todd, the Army’s program manager for utility helicopters, recently pointed out that his team is required to get approvals from 68 different offices for each step of his program. That’s multiple steps, each one multiplied by 68 phone calls or emails or forms to get signed—and his situation is far from unique.
When 68 offices all must approve a request, none of them can be held responsible for the actual outcome because overall accountability gets diluted by each additional approval required. And with such a large burden of coordination, it is possible that by the time he nears the finish line, Colonel Todd will actually need 69 or 70 approvals in order to cross it. Those missing voices or two may not be added or heard from until the last minute—or later.
Within the Pentagon, many sound decisions have been unnecessarily derailed at the last minute by an objection from an un-consulted corner of the building, despite the best efforts of staff officers to include all relevant parties. When I published an article titled "My Big Slow Fail " that described my experience with these beyond-the-last-minute changes and additions, readers from across the DoD flooded my inbox with notes saying “That exact same thing happened to me.”
The point is that in virtually every context, excessive complexity causes minutiae to prevail over the mission. The answer to dealing with excessive complexity is to remove it, so Kendall’s proposal to simplify is exactly what the Pentagon needs. However, simplicity advocates like Kendall must not only establish new simplifying mechanisms, they must also overcome a firmly entrenched cultural preference for complexity.
The point is that in virtually every context, excessive complexity causes minutiae to prevail over the mission.
I saw this dynamic firsthand when I was in uniform, but the starkest example occurred when I was stationed at the Pentagon itself. I was briefly assigned to a small tiger team charged with producing a recommendation on how to reduce the reporting requirements for technology development projects. A handful of us spent a couple weeks in a room with a big list of all the reports and documents an acquisition program must produce. We were supposed to analyze the list and pare it down to the essentials.
It looked easy at first. I immediately identified several reports that could be done away with. Based on my experience as an acquisition officer, I knew these particular reports did not help program managers or executives do their jobs. For the most part, they circulated data that was either redundant or obsolete. And although the Pentagon’s regulation (DoD Instruction 5000.02) explicitly says not to produce “compliance-only” documents whose sole purpose is to check a box, that is exactly what these reports did. They should have been easy kills.
Unfortunately and over my objections, each and every one of the items on that list were retained, with two reports added for good measure. This paradoxical and absurd outcome for a team tasked with streamlining—not increasing—reporting requirements was unbelievable but unsurprising to me. Every single item on that list was deemed vital to someone somewhere in the process so nothing could be let go. No one wanted to give up their own priorities or make a decision that would alienate others. And that is the problem Kendall will need to solve.
Today’s Pentagon needs the same thing my old tiger team lacked—priorities and the guts to stick to them. Once we decide what matters, simplifying things is quite easy. But without clear priorities, the process will continue to get more complicated. Kendall’s proposal is a step in the right direction, as is HASC Chairman Representative Thornberry’s recent acquisition reform bill, which emphasized “simplified decision-making and thinning out regulations.”
The Pentagon’s acquisition process won’t get fixed overnight, but the proposals by Kendall and Thornberry are an encouraging start. They invite the wider acquisition community to make the hard decisions about what matters and what can be set aside. Time will tell whether these proposals actually make things simpler and better, or whether the current inertia will continue to make things more complicated.
This post originally appeared in New America’s digital magazine, The Weekly Wonk, a Pacific Standard partner site. Sign up to get The Weekly Wonk delivered to your inbox, and follow @NewAmerica on Twitter.