Test Your Ejection Seat

Posted Monday, October 15, 2012

Allen Slade

Felix Baumgartner’s space jump had a dramatic buildup to a truly amazing physical feat. The video of Baumgartner spinning and tumbling at 700 m.p.h. is heart wrenching. And it ended with a perfect landing.

However, the “mission” was presented as a NASA-like experiment to make space jumping viable for escaping from a space vehicle. The mission was not very NASA-like. The flight director was no Chris Craft. And Baumgartner is an extreme athlete, not an astronaut. His poor checklist discipline and his concern with styling his hair after removing his helmet was more Tom Cruise than Alan Shepherd.

My father, Harry Slade, worked at NACA/NASA from the beginning of the jet age through the space shuttle. And, in 1976, in the New Mexico desert, he helped test one of the oddest emergency systems in the world: a helicopter ejection seat. How does Baumgartner’s space jump compare to a helicopter ejection seat?

The best ejection seats are virtually automatic – pull one lever and hang on. Baumgartner’s exit was slow and difficult: a long checklist and keeping his cool while tumbling at 700 m.p.h. But Baumgartner did not jump from a hypersonic spacecraft. He bunny hopped from a capsule floating gently beneath a balloon.

Space jumping will require more planning and testing before it provides real safety. In an emergency space jump, the craft would probably be spinning, disintegrating or on fire. A Baumgartner-like jump would be too complex and too physically stressful for a space tourist to survive.

NASA’s helicopter ejection seat was not like the space jump. The ejection seat was for a helicopter testing next generation rotor blades. Test rotors were thinner and lighter, increasing the risk of failure. Rotors are most likely to fail during takeoff and landing. A pilot could not eject down because of the low altitude. The pilot cold not eject sideways either, because rotor debris would be spiraling to every point of the compass. The ejection seat had to go up, through the path of the rotors. The ejection sequence was complex: pull the ejection handle, blow off the rotors with explosive bolts, blow the hatch, eject the seat, cut the seatbelts, then deploy the parachute. Asking a pilot to remember these steps during a crash was a recipe for failure. The pilot’s only job was to pull the ejection handle and let the automatic system do the rest.

For leaders, when the crisis happens, it is usually too late to build an ejection seat. When time and tempers are short, people go into survival mode. The brain shuts down cognitive processes, attention narrows and choices seem limited to fight or flight. What backup plans do you need to create? Your leadership crisis may be the loss of a key employee, an unexpected customer request, failure of essential equipment or overload of a critical operation. If the crisis happens quickly, you need to already have an ejection seat.

For a job interview in 2002, I made a presentation on leadership. I emailed my presentation to the host. By plan, I arrived 15 minutes early to check the room. When we turned everything on, my presentation was not on the laptop and my host did not know how to transfer the file. (It was 2002.) No problem – I had a floppy disk. (No USB in 2002.) She found an external disk drive, but the drive did not work. No problem – I had transparencies. They found a dusty overhead projector. When we turned it on, the bulb did not work. No problem – I activated the built in back up bulb. The second bulb worked. I tested the projected image, focused the projector, focused my emotions and was ready to go. (I had one more backup if the second bulb blew – use the whiteboard.) The presentation went well, and I was offered the job.

Lesson #1: For critical activities, have multiple backup plans before the crisis.

The helicopter ejection seat offers another lesson. My dad was in New Mexico to test the seat. The helicopter was bolted to a jet sled, accelerated to flight speed, then a test dummy was ejected. The first test was almost right. The rotors did not blow off before the seat was ejected, shredding the test dummy. (A watching test pilot walked away saying “I’m never flying that bird.”) The NASA team found and fixed the flaw in the timing circuits. All the other tests were successful.

When I joined Microsoft in 1999, my key task was to make the employee opinion survey a success. The 1998 survey locked up because of overloaded servers at the survey host. Under my direction, we created backup plans for every problem we could imagine. More importantly, we tested our back up plans. We did load testing, usability testing, data integrity testing and translation testing. We even tested our testing. In effect, we accelerated the survey to flight speed and pulled the ejection handle. The testing identified problems that we fixed, then we tested to the point of repeated perfection. The actual survey went well – on time, with virtually 100% uptime.

Lesson #2: For critical activities, test your backup plans before the crisis. Make the tests as realistic as possible. And test your tests.

As a leader, what are your key processes? How can they fail? What are your plans for overcoming failure? And how well tested are those plans?

Don’t rely on extreme leadership ability in a crisis. Few of us can tumble at supersonic speed like Felix Baumgartner. Plan for failure and test your plans until you know your ejection seat will save the day.