Identifying Systematic Biases

Beyond general overconfidence, retrospective analysis reveals specific biases in your intelligence practice. Anchoring bias: over-weighting the first signal you received about a trend, even when later signals contradict it. Availability bias: overestimating the importance of threats you can easily imagine. Status quo bias: underestimating the probability of disruptive change. Confirmation bias: seeking signals that confirm your existing thesis while discounting contradictory signals. Each bias produces a specific pattern in your prediction errors that the retrospective can detect.

1. Categorize Your Misses For each incorrect prediction from the quarter, categorize the error: did you overestimate the probability of change (status quo bias in reverse), underestimate the probability of change (status quo bias), overweight a single dramatic signal (availability or anchoring), or miss contradictory evidence (confirmation bias)? The distribution across categories reveals your dominant biases.
2. Test for Directional Bias Do you systematically overpredict threats (pessimistic bias) or underpredict threats (optimistic bias)? Do you overpredict the speed of change or underpredict it? Directional biases are the most actionable findings because they suggest specific corrections: if you consistently overpredict speed, add 3-6 months to every timeline estimate.
3. Implement Corrections For each identified bias, implement a specific correction in your analytical process. Status quo bias: explicitly consider "what if this changes dramatically?" for every assessment. Confirmation bias: actively seek signals that contradict your current thesis before finalizing any assessment. Anchoring bias: re-evaluate from scratch periodically instead of updating from your initial assessment.