You think you're being rational. You're not. Cognitive biases are invisible to the person experiencing them. The bias check is a structured set of questions designed to catch the ways you're fooling yourself - before you commit.
You can't eliminate bias. You can surface it.
The question isn't "am I biased?" The question is "which bias is operating?"
The Core Biases & Their Checks
You seek information that confirms what you already believe. If you can't name what would disprove your position, you're not thinking - you're rationalizing.
Past investment shouldn't affect future decisions. The money/time already spent is gone either way.
The first number you encounter biases all subsequent estimates. Generate your own estimate before looking at others.
Vivid examples feel more common than they are. The plane crash you heard about doesn't make flying dangerous.
Most people are wrong more often than they think. Widen your confidence intervals.
The default option gets unfair advantage. Evaluate as if all options were new.
Consensus feels like truth but isn't. Popularity is not evidence.
You underestimate time and cost. Always. Use reference class forecasting - base estimates on similar past projects, not optimistic plans.
Running A Bias Check
Before You Decide
Don't check for bias after you've committed - the check happens before. Once decided, you'll rationalize. The window is narrow.
You don't need to check every bias every time. Pick the 3-4 most likely to apply to this specific decision. Context matters.
Don't just think about it - write it down. Writing forces clarity. Vague answers in your head feel solid until you try to articulate them.
Show someone your answers. Ask them what you're missing. Biases are invisible from inside - you need external eyes.
The 5-Minute Version
If you only have five minutes, ask these three:
1. "What would change my mind?"
If nothing would, you're not reasoning - you're defending.
2. "What am I not seeing because I don't want to see it?"
The uncomfortable truth you're avoiding is usually the important one.
3. "What would a smart person who disagrees say?"
Steel-man the opposition. If you can't articulate their view, you don't understand the problem.
High-Stakes Moments
Money, time, reputation on the line. The cost of being wrong is high. Run the check.
People decisions are especially bias-prone. Similarity bias. Halo effect. Recency. Force the structured check.
Certainty is a red flag, not a green light. The more sure you feel, the more likely you're missing something. Check hardest when you're most confident.
Deciding to quit something is as prone to bias as starting. Sunk cost pulls you to stay. Frustration pulls you to leave. Neither is evidence.
You can't think yourself out of biases by trying harder. Bias operates below conscious awareness. What you can do is build systems that surface them - structured questions that force you to confront what you'd rather ignore.
Run the check. Answer honestly. Then decide. It won't make you perfect - but it'll make you less wrong.
The bias you don't check is the one that costs you.