liveBook · Manning

Decision quality

As explained in Decision Engineering

The merit of a decision as judged by the reasoning and process behind it rather than solely by its eventual outcome.

Technical decisions are bets made under uncertainty, not simply choices between a right answer and a wrong one. Judge their quality by the reasoning and process behind them, not only by what happens afterward. A thoughtful decision can still lead to a poor outcome when circumstances change or uncertainty remains. section 1.1.1 section 1.5

To assess a decision, ask whether the process:

  • Clarified the problem and the constraints, assumptions, and expectations that matter. section 1.1.1 section 2.5
  • Compared alternatives through trade-offs, identified risks, and acknowledged uncertainty. section 1.1.1
  • Involved the right people and made its reasoning transparent. section 1.1.1

A good decision, a poor outcome

A team can carefully assess alternatives and risks, then choose an option that fails as circumstances change. That does not by itself make the decision poor. By contrast, rushing to a solution without understanding a constraint can mean solving the wrong problem or solving the right problem in the wrong way. section 1.1.1 section 2.5

Research, prototyping, and conversations with relevant people can reduce uncertainty, but cannot eliminate it. Calibrate the effort to the stakes: one-way-door decisions warrant careful analysis and broad involvement; two-way-door decisions benefit from speed. Treat decision-making as a skill that can improve, and revisit choices when prototypes, constraints, or execution reveal a need to reframe or reevaluate. section 1.1.1 section 1.5

How Decision Engineering covers it

How deep it goes
Examines problem understanding, constraints, alternatives, risks, and stakeholders while recognizing that uncertainty cannot be eliminated.
Examples
Contrasts a thoughtful bet that fails with a rushed solution that misses a constraint or solves the wrong problem.
What it assumes
Assumes a need to understand the problem and treats uncertainty as persistent rather than something analysis can eliminate.
What you can do after
Evaluate technical decisions by checking their problem framing, trade-offs, risks, uncertainty, and stakeholder involvement.
How it connects
Connects decision quality to requirements gathering, architectural significant requirements, uncertainty, and the decision lifecycle.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Decision Engineering. The section links point to the parts of the book it draws on.