And now for something completely different: honesty, courage, and starting over.

John McLaughlin, CIA deputy director, once warned, "Our country is vulnerable--if our intelligence analysts are not ready for something completely different from what they have experienced in the past." That was March 11, 2001. Exactly six months later, something completely different and apparently unanticipated did indeed happen.

[ILLUSTRATION OMITTED]

McLaughlin's prediction and warning has something to teach not only the intelligence community, but the technology development community as well.

Anticipating the Unexpected

The safest thing to say about the future is that it will be full of unexpected events. While the details of those events, activities, and developments are largely unknowable, no one should be surprised to discover that the future is going to be ... surprising. We may try to minimize the uncertainties and prepare for any possible outcome, but our crystal balls get murky the farther we try to look. That murkiness is one of life's great certainties, and it is an area deserving of our attention.

Of course, some future events can be predicted easily. But along with preparing for predictable outcomes, there is a full spectrum of possible surprises that may require a program manager (PM) to make a course correction. So PMs need to establish a mechanism--a flexible, simple mechanism--for responding quickly and smartly to life's inevitable surprises.

The SAWABI Approach

In the most extreme cases (which may or may not be uncommon) the recommended approach is called "SAWABI," which stands for "Start Again With A Better Idea." Once you decide to do it, implementing SAWABI is quite simple. The tricky part is determining that a SAWABI approach is necessary. Such a decision requires equal parts objectivity, honesty, and courage. Here are the steps for using this method.

Step One. Acknowledge that SAWABI is an option. PMs don't need to keep doing everything they are currently doing, particularly if there is a better idea out there somewhere. Given the dynamic nature of the unknown future, we can't expect always to find the best ideas on the first try. Software guru Eric Raymond recommends that programmers should "expect to start over at least once," and smart PM's should be willing to do so as well.

[ILLUSTRATION OMITTED]

It is important to understand that SAWABI covers a wide range of "starting over" activities, from a minor adjustment to a completely blank sheet. It doesn't always mean canceling an entire program; it could simply involve retooling a particular process or approach.

Step Two. Take an objective look at the situation and determine whether the current approach is the right one. Note we did not say "the best" or "the ideal": it is often sufficient to be adequate. Sometimes a better technical approach exists, but the cost of changing exceeds the benefit. A better idea, by definition, encompasses cost, schedule, and performance considerations. If a technology's corresponding impact on cost and/or schedule is unacceptable, then it is not really a better idea, just a better technology. And they're not the same thing.

Step Three. Make the call. Do your homework, get your ducks in a row, and start making the case for starting over. In the current acquisition framework, this is sometimes easier said than done, but despite the difficulty, it is indeed possible.

G. K. Chesterton warns against pulling something down "without even pausing to ask why it was put up" in the first place. He explains that unless we understand the reason for something's existence, we cannot "judge whether the reason was reasonable," and so we ought to be very reluctant to remove, replace, or destroy it. Thus, a SAWABI decision, which by definition involves abandoning an existing thing, must begin with an understanding of the thing's original purpose. As Chesterton goes on to explain, once we understand "how it arose, and what purposes it was supposed to serve, [we] may really be able to say that they were bad purposes, or that they have since become bad purposes, or that they are purposes which are no longer served."

Getting Out of Zimbardo's Prison

Anyone who sat through Psych 101 probably encountered the infamous prisoner/guard experiment performed by Stanford professor Phillip Zimbardo. In brief, Professor Zimbardo brought a group of undergrads together in a "prison" constructed in the basement of the university's Psychology Building and randomly assigned them to be either uniformed prison guards or prisoners. The experiment, intended to last two weeks, rapidly degenerated into a seething stew of cruelty and depression, and the experimenters were forced to cancel it after six days.

The most interesting and relevant point is that each participant could have opted out at any time, but almost all stayed--even the grossly mistreated prisoners--until the experimenters called it off. All the participants had to say was, "I'm done," and they'd go back to real life. They knew they had the authority and ability to cease their participation. Maybe they got caught up in the moment and forgot they did not have to proceed. Surely they were not comfortable, didn't think all was well and appropriate, particularly the unfortunate prisoners. Yet they didn't act on their own responsibility to put a stop to it.

[ILLUSTRATION OMITTED]

The lesson for PMs should be obvious. We may not be able to walk away from a bad program as easily as Zimbardo's subjects could have, but we do indeed have the ability and responsibility to speak up when a situation degenerates. We need to make the call and advise our superiors accordingly, waving the SAWABI flag whenever a current trajectory needs adjusting--or cancelling.

SAWABI: Chaos and Innovation

SAWABI appears to inject a certain degree of chaos into a program by removing the assurance of continuity. What SAWABI actually does is acknowledge the ambiguity that is always there and enable a PM to respond appropriately. This is all to the good because the certainty inherent in some programs is unfortunate, unwarranted, and unwise. An assurance of programmatic continuity, regardless of performance, can have a numbing effect. The presence of a SAWABI mechanism removes that assurance and its associated numbness, thereby facilitating innovation and growth. When we know that any program we begin is probably destined to last indefinitely, there is little conscionable space for experimentation or error, for fear of propagating a poor result. But if we are free to start in one direction and then start again with a different tack later, we are more likely to explore new ground, make some interesting mistakes, learn something, and go on to discover the better idea we'd been seeking all along. Absent a SAWABI mechanism, we will find it much harder and slower to apply our learning in a timely manner or to grow.

Of course the opposite response is also possible. Once SAWABI is an option, some PMs may be reluctant to take risks or try something new, for fear that the program will be cancelled. That is why a SAWABI mechanism has to be relatively painless and not reflect poorly on the brave souls who attempt to use it. Without a painless SAWABI mechanism, fear of failure, fear of waste, and fear of getting it wrong will lead directly to waste, failure, and wrong answers.

Even if a SAWABI approach does cause some discomfort, there are times it must be pursued, nonetheless, with courage and honesty. Given the types of systems we develop in the DoD acquisition and development community, a lack of courage or a lack of honesty are frankly inexcusable. Lives and national security are at stake, so fear must not dictate our behavior or decisions. SAWABI, therefore, indicates the presence of these two key virtues--courage and honesty--both of which are absolutely vital attributes for a PM. Those who need a little more encouragement would do well to read the White House's National Security Strategy from September 2002, which explains: "The major institutions of American national security were designed in a different era to meet different requirements. All of them must be transformed" (emphasis added).

Sunk Cost and Cognitive Bias 

For the politically minded, abandoning previous investments appears to indicate poor judgment; in fact, making a course correction is actually a sign of good judgment. After 21 years and $8 billion, if we don't have a useful system, it is probably time to SAWABI. At some point, continuing on a fruitless trajectory is no longer admirable persistence, but rather a sign of possible mental illness. At the very least, it is evidence of an unfortunate cognitive bias.

PMs must wrestle with a common cognitive bias for programmatic stability--a preference for keeping programs alive even if they should be cancelled. Interestingly, this bias exists even if you know it's there. You've probably seen this illustration:

Those who aren't familiar with the optical illusion may need to be told the two lines are the same length. The funny thing is, even though you know the lines are the same length, they still don't look the same. Even after you measure them, your eyes will continue to insist the upper line is longer. In this simple scenario, your perception contradicts your intellectual knowledge. The question is, which will guide your actions? Now imagine the following situation:

[ILLUSTRATION OMITTED]

A hungry lion is behind Door # 1 if the lines are the same length. If they are different lengths, the lion is behind Door #2. You must open one door.

Which do you choose? Which source of data do you trust--your eyes or your intellect?

Similarly, a PM may know that Project X or Process Z needs to be cancelled/replaced/modified. According to my college economics professor, one is not supposed to take sunk costs into account when evaluating future options. But even though a PM may know intellectually not to include sunk costs in his or her calculations, there is a strong tendency to argue in favor of existing programs "because we've already spent $30 gazillion." What course of action should that PM take? Fear of failure and criticism leads in one direction and is supported by a cognitive bias for continuity. Honesty and courage point in the opposite direction, and require us to trust what we know to be true.

Letting Go of the Rope

There's an old saying that you can't unring a bell, but you can stop pulling the rope. In other words, we can't undo the past, but we can do something different in the future. Cancelling programs does not waste money: it prevents continued waste. Retooling a process, restarting a program, pursuing a demonstrably superior idea may indeed involve abandoning previous investment, but such courses of action also prevent throwing good money after gone money. SAWABI is indeed a fiscally responsible option when implemented judiciously.

[ILLUSTRATION OMITTED]

The British comedy troupe Monty Python's Flying Circus used the phrase. "And now for something completely different ..." as a segue between sketches. It's a phrase PMs should seriously consider adding to their vocabulary. Whether a SAWABI approach results in shorter meetings or an entirely new endeavor, when a better idea exists it is often worth pursuing. The key to maintaining this responsive, flexible posture is a firm commitment to honesty and a courageous objectivity. We ought not abandon every project at the first sign of difficulty, but we probably should exercise the option more often than we do. Rather than remain in Zimbardo's prison, we need to recognize our obligation to speak up and opt out when the situation warrants it.

And remember--just because one line looks longer doesn't make it so.

Editor's note: The author welcomes comments and questions. He can be reached at [e-mail].

Capt. Daniel Ward, USAF

Ward is an InnoVisioneer with the InnoVision Directorate of the National Geospatial-Intelligence Agency. He holds degrees in electrical engineering and engineering management, and is Level III certified in SPRDE. Level 1 in PM. T & E. and IT.