The other day, I was talking with a software engineer about a potential project. During the conversation, he asked if I was ever on a project that failed.

“Failed?”

I’ve worked on thousands of projects and given that more than 50% of project fail. I asked for a definition.

“The project didn’t launch,” he responded.

Given the complexity of even the simplest project, software or otherwise, it still seems like an odd question. The gestation of any project requires a commitment of time and effort – at any point, some external force could (and frequently does) warrant an end to the project. An engineer at Edward’s Air Force base even declared this a phenomenon a law.

All this assumes the project was a good idea to begin with. I’ve been on projects where it became very clear the project was doomed. Bad idea from the outset. It just took a little while to figure out exactly why.

Is this failure? I don’t think so. Especially if “failure” was found quickly. In the long run, stopping a project mid-stream saves time, effort, and probably a reputation or two. The trouble comes in when, despite all the red flags, the project continues onward. Unstoppable, yet acutely aware of the impending demise.

Measuring a success by whether or not a project saw the light of day is like judging how good your day was based on the thermostat reading. It’s just one of many factors.

I think a more interesting question would be, “Have you ever been on a project that succeeded.”

ELSEWHERE 28 Dec 2007:

Testing leads to failure, and failure leads to understanding.” – Burt Rutan