It takes a special kind of person to be inspired by a mandate riddled with risk and having little margin for error, such as the one issued in the early 1990s by NASA to its Jet Propulsion Laboratory (JPL) in Pasadena, California: “Take risks but don’t fail.”

Such a person is Brian Muirhead, who at age 41 in 1993 accepted the job as flight systems manager of the Mars Pathfinder project and with it the NASA challenge to land a cutting-edge, remote-controlled robotic all-terrain rover on Mars that would reliably beam back images, collect samples, and return scientific data on the red planet.

The only catch: he was given just three years and $150 million to do it. The immediately preceding Mars Observer, which carried a $1 billion price tag and had taken 10 years, had just been lost in space, an embarrassing failure for the U.S. space program. No one in his or her right mind would want to manage the next Mars project, if indeed there was one.

Brian is a quiet, cerebral, and unassuming rocket scientist. Now chief engineer at JPL, he has a significantly bigger title, significantly less hair, and significantly more white in his beard than when I first met him, undoubtedly as a result of his almost 35 years of intense involvement with high-profile missions in pursuit of JPL’s mission to push the outer edge of space exploration.

Brian was a frequent visitor to Toyota’s Los Angeles campus during my tenure there as an advisor from 1999–2006. After reading his 1999 book High Velocity Leadership, we invited Brian to guest speak about the Mars Pathfinder project, and he soon he became a regular fixture in the University of Toyota’s “lean” leadership curriculum. Through the sessions I became quite familiar with Brian and his saga.

One of my all-time favorite stories is the one he told about the Pathfinder team’s approach to a “don’t fail” strategy. Brian would begin his story with how he was personally disrupted, in fact blindsided, by, of all things, one of his daughter’s kindergarten projects. The teacher gave the class an assignment right up Brian’s alley: design a package that would protect a raw egg from being dropped off the school roof. It was an annual event, affectionately referred to as the Great Egg-Drop Challenge.

It was right up Brian’s alley for two reasons: not only is he an expert on momentum, but the radical solution enabling the successful landing of the rover on Mars on July 4, 1994 addressed essentially the same problem. (If you don’t know, they used large airbags to “bounce” the rover on the surface of Mars, because the sparse budget precluded a rocket-powered solution.)

Brian had an answer in a snap, and coached his daughter in a rather conspiratorial way in order for her to arrive at the same solution, which involved a milk carton stuffed with newspaper.

Together they cut up newspaper, wadded it in the carton, put the uncooked egg in a plastic bag and set it on top of the loose packing. They tested it several times of their home’s high balcony to find just the right amount of padding to allow the egg to land untracked.

The whole school gathered for the event. But it was not the teacher who did the testing. It was the school principal, who tested the designs not by dropping them straight down, but by throwing them in what Brian described as a “big, high, looping arc.” Get the visual?

Needless to say, Brian’s daughter’s egg was crushed, as was his daughter’s morale. Brian himself was devastated: here he was building a spacecraft to travel 300 million miles to land safely on another planet and he couldn’t even help his daughter design a landing device to protect an egg.

Call it assuming, call it blindspots, call it oversight, call it taking unknown forces for granted. Brian realized a valuable lesson from the kindergarten experience, and applied it to the Pathfinder project: inevitable, unforeseen and disruptive forces could be the ruin of the project, so there had to be a way to mitigate their potential impact.

Enter what I call the “Gremlin” strategy.

Now, in space projects like the Pathfinder mission, it’s the job of someone called the fault protection engineer to look at possible failures of the spacecraft, understand how to recognize them, how the spacecraft would react, and what the consequences might be. But the focus of the job is on “what is true now,” rather than “what must be true” for things to go as planned.

Interestingly, the Pathfinder team had just lost their fault systems engineer, so Brian asked another team member to fill the spot. His name was Dave Gruel, but he eventually was nicknamed “Cruel Gruel,” because he was the ultimate Gremlin.

The term “Gremlin” was popularized during World War II, and referred to an imaginary creature that creates problems in normally reliable hardware.

With the kindergarten experience fresh in his memory, Brian tasked Dave with duties beyond fault protection: he asked him dream up all sorts of challenges to throw at the team. It turned out that Dave Gruel had a real flair for the role, and spent his days and nights devising ways to disrupt the project. The tests were, as Brian told it, “cruel and almost pathological.” Dave “the Gremlin” Gruel regularly ruffled the feathers of engineers and controllers alike, by coming in at night while the team tucked in bed and monkeying with the rover or booby-trapping supporting equipment. Cruel Dave Gruel had an uncanny and rather Machiavellian ability to conjure up the most surprising ways to take the team by surprise.

For example, in the final preparation phase utilizing a sandy, rocky, imitation-Mars environment, the team was to test the ability of the rover to drive off the lander ramps onto the surface of Mars. There were two ramps, one in front and one in back of the landing module. From the control room, which utilized stereo 3D-view cameras to monitor the testing area, everything looked good, and the team was all set to drive the rover down the front ramp.

Brian urged them to be cautious, to think things well through, and to make sure they were making the right move. His words of warning went unheeded, though…the team was certain they had analyzed the situation thoroughly, and were ready to roll.

Except…the Gremlin had come in during the night.

As Brian tells it in High Velocity Leadership:

They were certain but entirely wrong. Cruel Dave Gruel had created a monster of a challenge — building a mound of sand in front of the lander, and creating a pit behind it. The rover team recognized the real danger of driving off the front ramp, leaving the rover to negotiate down a quite steep sandy hill on which it would have little traction. That meant they had to use the rear ramp. Because of the pit Dave had created, tho stamp was sitting at an extreme angle, but at least the rover’s unique metal wheels were designed to give good traction on the ramp.

We would later discover that the software program created to determine the steepness of the ramps had an error in it, and was giving the rover people incorrect data, making them think the angle was within safe parameters — shallow enough an incline for the rover to use.

Running a couple of available calibration checks had shown something wasn’t right. Yet the attitude of some of the rover people, so cocky, so certain of their judgment, kept them from stepping back and questioning their decision.

I see this attitude all the time in my role as a strategy advisor.

In the end, the Gremlin strategy was extremely effective in enabling the Pathfinder team to learn how to ferret out and deal with uncertainties in a way that positively neutralized disruptive forces. It allowed the team to build robustness, speed, and flexibility into their implementation.

The application to business should be clear. If you have a successful business, the chances are very good that somewhere someone or someones are dreaming up strategies that may just throw you for a loop.

So why not beat them to the punch?

Take a page from the Mars Pathfinder story and set up a Gremlin group in your company. Charge them with putting you out of business in new and innovative ways. The goal is not to break or undo anything that’s working well, it’s come at the business problem you’re trying to solve from an entirely different perspective. Done right, it will not only ward off disruption, it will do more to build innovation capability into your organization than any highfalutin innovation program some big name firm sold you on. Mars Pathfinder didn’t need such a program, and neither do you.

I recently worked with one company to set up a Gremlin group, and it has resulted in some brilliant strategies…ideas I doubt would have ever materialized otherwise. We rotate people through the Gremlin group, and it’s become a highly sought after assignment, even a badge of honor. It’s a great place for recent hires, after they familiarize themselves with the company but before they stop asking, “why do we do it that way?” And on an ongoing basis it’s a great place for the “lunatic fringe” ranks of the company to flourish.

What’s amazing to me is the level of passion, verve and vigor I see inside the Gremlin teams. The level of engagement is a full click above their engagement in their “real job.” I’m not quite sure I can draw an intelligent or meaningful conclusion about corporate cultures from this, because I simply don’t have enough evidence. But the early indications are that this may be a wonderful new wrinkle to the trend toward internal innovation and startup mechanisms like incubators and accelerators, which have replaced the older “skunkworks” approaches.

It certainly worked for the Pathfinder team: a successful template for innovation resulted, and the entire project from concept to touchdown was completed in 44 months, less than half the time of the previous Viking mission to Mars, using only 300 team members, versus the over 2000 assigned to the previous mission. And, they did it for less than it cost to produce the 1997 Hollywood blockbuster “Titanic.”

As for Brian Muirhead, an individual fond of dramatic destinations, I have no doubt that he’s hard at work on his favorite answer to the “what’s next?” question he was often asked: “I’m working on a project that will attempt to land a spacecraft on an active comet and analyze it.”

So, if the word “disruption” is being uttered in the halls of your company (as it seems to be almost every firm I visit), gather ye Gremlins, and go to work!

(Warning: brave heart, strong will, and intestinal fortitude required!)