Expecting success 100% of the time, in an industry where variability can increase profits, incentivizes your employees to eliminate a profit generating event.
I have a hard time getting along with traditional project managers. They all seem to be working under two horrible misconceptions. First, that it is possible to eliminate all variability in a software project with adequate planning. Second, that it would be profitable to do so. These two misconceptions inevitably lead to incentive structures that reward innovation-less, safe projects, and punish risk taking.
Software development is product development. It isn’t manufacturing. It is absolutely and completely impossible to remove variability from this process. If you could, you wouldn’t be paying your software developers $80-120k a year. You would be paying $10-20 an hour for high-school graduates, or pushing the button on your nifty software development tool and having a complete and bug-free program pop out the other side. But you can’t do either of those things, because each new piece of software is new, and most of them are incredibly complex. Software development is a continuous chain of hypothesizing and testing. Each hypothesis may turn out to be false, which adds variability.
Regardless of whether you believe that it is possible to eliminate variability in software development or not, more important is the misconception around the value of variability. Again, software development is not manufacturing with its symmetrical payoff function where all variability is bad. The ability to vary scope and quality can allow you to push into a new market earlier (assuming of course you’re willing to pay back that technical debt). Trial-and-error innovation enables software companies to drive demand for feature sets that consumers didn’t even know they wanted. As the market evolves, your software must evolve along with it.
But each step on that evolutionary path requires doing something a little bit different, something that has never been done quite that way before. Doing something that has never been done before requires risk taken, to one degree or another. And they don’t call it risk because it guarantees success.
So cultivate a culture where calculated risk taking is celebrated, whether or not the outcome is successful. The only other option is innovation stagnation, and shrinking profit margins.