Death march projects are only viable because there are enough employees who tolerate them. Why risk crunch mode and possible failure? Yourdon lists the following reasons.
- The risks are high, but so are the rewards
- The "Mt. Everest" syndrome
- The "buzz" of working intensely with other committed people
- The naivete and optimism of youth
- The alternative is unemployment
- It's required in order to be considered for future advacement
- The alternative is bankruptcy or some other calamity
- It's an opportunity to escape the "normal" bureaucracy
Most people are gamblers at heart, and the promise of millions despite impossible odds has enticed a surprising number of software developers. Particularly for young programmers who haven't yet established family ties, the cost is much more likely to be acceptable.
Some people derive their self worth from competitors bested and obstacles overcome. Just as people climb Mt. Everest, they join software projects with unrealistic expectations because the goals are impressive. This is particularly dangerous when there is a maniacal vision and group mentality, such as a project where everyone is convinced that they will be the ones to revolutionize how people use the internet.
Epitomized by the term "skunkworks," some projects are given corporate blessing but exist outside the normal bureaucracy. More often than not, the developers are highly interested in the project and would work on it regardless of management's expectations. A particularly compelling example is the story of the Graphing Calculator application that shipped with the original Macintosh PowerPC computers.
Succeeding at a death march project disproves detractors and can silence critics. When those critics are vocal and in positions of power, the embarrassment to them that would follow success is often reason to attempt a death march project.