Crafting Engineering Strategy

Chapter 13. Strategy Testing for Iterative Refinement

If I could popularize only one idea about technical strategy, it would be this: prematurely rolling out a strategy prevents you from evaluating whether the strategy is effective. Pressure changes people’s behavior in profound ways, and they often make those changes to create the impression that they’re complying with your strategy while minimizing changes to the status quo (if you’re an executive) or getting your strategy repealed (if you’re not an executive). Neither is particularly helpful.

While some strategies are obviously wrong from the beginning, it’s much more common to see reasonable strategies fail because they didn’t get the small details right. Premature pressure is one common symptom of a more general phenomenon: most strategies are developed in a waterfall model, finalizing their approach before incorporating the lessons that reality teaches when you attempt the strategy in practice.

One effective way to avoid the waterfall strategy trap is explicitly testing your strategy to refine the details. This chapter describes the mechanics of strategy testing, including:

• When it’s important to test strategy (and when it isn’t)

• How to test strategy

• When you should stop testing

• Roles in strategy testing

• Metrics and meetings for strategy testing

• How to identify an untested strategy

• What to do when a strategy has progressed too far without testing

Many of the ideas in this chapter came together while I was ...