Recency Bias in Option Comparisons
Counter recency bias by defining the evidence window before reviewing results and separating a new event from the longer base pattern.
Audit the time window
- Choose the period before viewing the trend.
- Compare recent and longer-run rates.
- Look for a mechanism that changed conditions.
- Run the same time rule for both options.
The latest event is easy to overweight
A recent failure, praise message, or price jump is vivid and available in memory. It may reveal a changed process, but it may also be ordinary variation. The decision needs a consistent window and a reason for treating the recent period differently.
Recent data can be the right data
A policy, product version, or team change can make older evidence less relevant. Document the break point instead of blending incompatible periods.
Example: replacing a service after one outage
A platform fails during an important launch. The team finds that its annual outage rate remains lower than the alternative, but the incident exposed a missing recovery plan. The action changes resilience controls rather than switching from one vivid event.
The decision rule
What to record before acting
Weight recent evidence more only when a named change makes the older reference class less comparable.
Apply the rule in context with audit evidence timing and fit. Keep the evidence and assumptions with the decision so an outcome review can distinguish reasoning from luck.
Continue within decision psychology
Optimism Bias in Project Decisions
Check optimism bias by starting with outcomes from comparable past projects before adjusting for the current plan's specific advantages.
Optimism Bias in Project DecisionsChoice-Supportive Bias After a Decision
Prevent choice-supportive bias by preserving the original alternatives, expectations, and evidence before learning the outcome.
Choice-Supportive Bias After a Decision