A Simple Technique for Better Decisions

When I started in software development, projects went ahead with high certainty: we made big plans and then executed on them. That didn’t always turn out well.

Then Agile came along with its short feedback-rich cycles. That’s caught on only partially; in many organizations, teams work from hefty pre-populated backlogs.

Then product management took off. Concepts like experimentation, A/B testing, and MVP are now on everyone’s lips. And yet, execution still involves big and confident commitments and plans.

Unjustified certainty costs us, but experimentation isn’t free either. What we need is to balance predictability and evolution, depending on the given situation.

Creating this capability takes a while. I like the following technique, which is powerful yet simple:

As you contemplate a high-impact decision, ask: “What would have to be true for this decision to turn out great?”

Examples of such decisions: changing the behaviour of a popular feature, rewriting a chunk of code a certain way, prioritizing a project ahead of another one, moving a person to another team.

Use this question when you’re the sole decision-maker. Use it even more when multiple people weigh in on the decision, because the group might apply subtle pressure on its members to act confidently. The question tends to work well with a group because it floats assumptions and preconditions without calling them out as such.

Once those assumptions and preconditions are made explicit, you can decide what to do about them: note them and move on, test something, change course.

Asking this question injects humility and care into decision-making. It can quickly become a habit and a norm.