Decision Science
Decision Science
Decision Science: Making Decisions Under Uncertainty
Established Frameworks
Reversible vs. Irreversible (Type 1 / Type 2)
- Type 1: irreversible, high-consequence — deliberate carefully
- Type 2: reversible, recoverable — decide fast, correct later
- Most decisions are Type 2 but get treated like Type 1
Robust Decision Making (from DMDU)
Instead of predicting the “right” answer, pick the option that performs acceptably across the most scenarios. You’re not optimizing, you’re satisficing. Useful when you genuinely can’t predict outcomes.
Expected Value / Bayesian Thinking
Assign rough probabilities and rough costs/benefits. Even sloppy estimates beat pure gut feel because they force you to be explicit about what you think will happen.
Cognitive Biases Worth Knowing
- Status quo bias: doing nothing feels safer but is also a decision with consequences
- Information bias: gathering more info past a point doesn’t improve the decision, it just delays it
- Authority bias: wanting your boss’s input may be more about sharing blame than improving quality
Practical Heuristics
- 10/10/10 rule: How will you feel about this decision in 10 minutes, 10 months, 10 years? Helps calibrate actual stakes.
- Pre-mortem: Assume the decision failed. What went wrong? If you can mitigate the top failure modes, just go.
- Decision journal: Write down what you decided, why, and what you expected. Review later. Builds calibration over time and provides a paper trail.
Key Takeaway
Pick whatever is least bad if you’re wrong, document your reasoning, and move.
Further Reading
This post is licensed under CC BY 4.0 by the author.